Search

Optical tissue transparency permits scalable cellular and molecular investigation of complex tissues in 3D. Adult human organs are particularly challenging to render transparent because of the accumulation of dense and sturdy molecules in decades-aged tissues. To overcome these challenges, we developed SHANEL, a method based on a new tissue permeabilization approach to clear and label stiff human organs. We used SHANEL to render the intact adult human brain and kidney transparent and perform 3D histology with antibodies and dyes in centimeters-depth. Thereby, we revealed structural details of the intact human eye, human thyroid, human kidney, and transgenic pig pancreas at the cellular resolution. Furthermore, we developed a deep learning pipeline to analyze millions of cells in cleared human brain tissues within hours with standard lab computers. Overall, SHANEL is a robust and unbiased technology to chart the cellular and molecular architecture of large intact mammalian organs., Competing Interests: Declaration of Interests A.E. has filed a patent on SHANEL technologies described in this study., (Copyright © 2020 Elsevier Inc. All rights reserved.)

We present a comprehensive analysis of the submissions to the first edition of the Endoscopy Artefact Detection challenge (EAD). Using crowd-sourcing, this initiative is a step towards understanding the limitations of existing state-of-the-art computer vision methods applied to endoscopy and promoting the development of new approaches suitable for clinical translation. Endoscopy is a routine imaging technique for the detection, diagnosis and treatment of diseases in hollow-organs; the esophagus, stomach, colon, uterus and the bladder. However the nature of these organs prevent imaged tissues to be free of imaging artefacts such as bubbles, pixel saturation, organ specularity and debris, all of which pose substantial challenges for any quantitative analysis. Consequently, the potential for improved clinical outcomes through quantitative assessment of abnormal mucosal surface observed in endoscopy videos is presently not realized accurately. The EAD challenge promotes awareness of and addresses this key bottleneck problem by investigating methods that can accurately classify, localize and segment artefacts in endoscopy frames as critical prerequisite tasks. Using a diverse curated multi-institutional, multi-modality, multi-organ dataset of video frames, the accuracy and performance of 23 algorithms were objectively ranked for artefact detection and segmentation. The ability of methods to generalize to unseen datasets was also evaluated. The best performing methods (top 15%) propose deep learning strategies to reconcile variabilities in artefact appearance with respect to size, modality, occurrence and organ type. However, no single method outperformed across all tasks. Detailed analyses reveal the shortcomings of current training strategies and highlight the need for developing new optimal metrics to accurately quantify the clinical applicability of methods.

Transcatheter left atrial appendage occlusion is an increasingly used alternative to oral anticoagulation in selected patients with atrial fibrillation. Pre-procedural imaging is a prerequisite to a successful intervention, with transesophageal echocardiography as the current gold standard. However, cardiac computed tomography offers improved imaging with high-quality multiplanar and 3-dimensional reconstructed images. Nevertheless, the lack of a standardized imaging protocol has slowed the adoption of cardiac computed tomography into clinical practice. On the basis of current research and expert consensus, this paper provides a protocol for the preparation, acquisition, and interpretation of cardiac computed tomographic imaging in pre-procedural planning of left atrial appendage occlusion., (Copyright © 2020 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)

Objectives: The evaluation of sources of error when preparing, printing, and using 3-dimensional (3D) printed head models for training purposes., Methods: Two 3D printed models were designed and fabricated using actual patient imaging data with reference marker points embedded artificially within these models that were then registered to a surgical navigation system using 3 different methods. The first method uses a conventional manual registration, using the actual patient's imaging data. The second method is done by directly scanning the created model using intraoperative computed tomography followed by registering the model to a new imaging dataset manually. The third is similar to the second method of scanning the model but eventually uses an automatic registration technique. The errors for each experiment were then calculated based on the distance of the surgical navigation probe from the respective positions of the embedded marker points., Results: Errors were found in the preparation and printing techniques, largely depending on the orientation of the printed segment and postprocessing, but these were relatively small. Larger errors were noted based on a couple of variables: if the models were registered using the original patient imaging data as opposed to using the imaging data from directly scanning the model (1.28 mm vs. 1.082 mm), and the accuracy was best using the automated registration techniques (0.74 mm)., Conclusion: Spatial accuracy errors occur consistently in every 3D fabricated model. These errors are derived from the fabrication process, the image registration process, and the surgical process of registration., (Copyright © 2019 Elsevier Inc. All rights reserved.)

Background: Gastrointestinal stromal tumors (GISTs), which are the most common mesenchymal tumors of the digestive system, are treated varyingly according to the malignancy. The purpose of this study is to develop and validate a nomogram for preoperative prediction of the malignant potential in patients with GIST., Methods: A total of 440 patients with pathologically confirmed GIST after surgery in our hospital from January 2011 to July 2019 were retrospectively analyzed. They were randomly divided into the training set (n = 308) and validation set (n = 132). CT signs and texture features of each patient were analyzed and predictive model were developed using the least absolute shrinkage and selection operator (lasso) regression. Then a nomogram based on selected parameters was developed. The predictive effectiveness of nomogram was evaluated by the area under receiver operating characteristic (ROC) curve (AUC). Concordance index (C-index) and calibration plots were formulated to evaluate the reliability and accuracy of the nomogram by bootstrapping based on internal (training set) and external (validation set) validity. The clinical application value of the nomogram was determined through the decision curve analysis (DCA)., Results: Totally 156 GIST patients with low-malignant (very low and low risk) and 284 ones with high-malignant potential (intermediate and high risk) are enrolled in this study. The prediction nomogram consisting of size, cystoid variation and meanValue had an excellent discrimination both in training and validation sets (AUCs (95% confidence interval(CI)): 0.935 (0.908, 0.961), 0.933 (0.892, 0.974); C-indices (95% CI): 0.941 (0.912, 0.956), 0.935 (0.901, 0.982); sensitivity: 81.4, 90.6%; specificity: 75.0, 75.7%; accuracy: 88.0, 88.6%, respectively). The calibration curves indicated a good consistency between the actual observation and nomogram prediction for differentiating GIST malignancy. Decision curve analysis demonstrated that the nomogram was clinically useful., Conclusion: This study presents a prediction nomogram that incorporates the CT signs and texture parameter, which can be conveniently used to facilitate the preoperative individualized prediction of malignancy in GIST patients.

Background: To evaluate radiomics analysis in neuro-oncologic studies according to a radiomics quality score (RQS) system to find room for improvement in clinical use., Methods: Pubmed and Embase were searched up the terms radiomics or radiogenomics and gliomas or glioblastomas until February 2019. From 189 articles, 51 original research articles reporting the diagnostic, prognostic, or predictive utility were selected. The quality of the methodology was evaluated according to the RQS. The adherence rates for the six key domains were evaluated: image protocol and reproducibility, feature reduction and validation, biologic/clinical utility, performance index, a high level of evidence, and open science. Subgroup analyses for journal type (imaging vs. clinical) and biomarker (diagnostic vs. prognostic/predictive) were performed., Results: The median RQS was 11 out of 36 and adherence rate was 37.1%. Only 29.4% performed external validation. The adherence rate was high for reporting imaging protocol (100%), feature reduction (94.1%), and discrimination statistics (96.1%), but low for conducting test-retest analysis (2%), prospective study (3.9%), demonstrating potential clinical utility (2%), and open science (5.9%). None of the studies conducted a phantom study or cost-effectiveness analysis. Prognostic/predictive studies received higher score than diagnostic studies in comparison to gold standard (P < .001), use of calibration (P = .02), and cut-off analysis (P = .001)., Conclusions: The quality of reporting of radiomics studies in neuro-oncology is currently insufficient. Validation is necessary using external dataset, and improvements need to be made to feature reproducibility, demonstrating clinical utility, pursuits of a higher level of evidence, and open science.

Background: The microTargetingTM MicrotableTM Platform is a novel stereotactic system that can be more rapidly fabricated than currently available 3D-printed alternatives. We present the first case series of patients who underwent deep brain stimulation (DBS) surgery guided by this platform and demonstrate its in vivo accuracy., Methods: Ten patients underwent DBS at a single institution by the senior author and 15 leads were placed. The mean age was 69.1 years; four were female. The ventralis intermedius nucleus was targeted for patients with essential tremor and the subthalamic nucleus was targeted for patients with Parkinson's disease., Results: Nine DBS leads in 6 patients were appropriately imaged to enable measurement of accuracy. The mean Euclidean electrode placement error (EPE) was 0.97 ± 0.37 mm, and the mean radial error was 0.80 ± 0.41 mm (n = 9). In the subset of CT scans performed greater than 1 month postoperatively (n = 3), the mean Euclidean EPE was 0.75 ± 0.17 mm and the mean radial error was 0.69 ± 0.17 mm. There were no surgical complications., Conclusion: The MicrotableTM platform is capable of submillimetric accuracy in patients undergoing stereotactic surgery. It has achieved clinical efficacy in our patients without surgical complications and has demonstrated the potential for superior accuracy compared to both traditional stereotactic frames and other common frameless systems., (© 2020 S. Karger AG, Basel.)

Motivation: Applying infrared microscopy in the context of tissue diagnostics heavily relies on computationally preprocessing the infrared pixel spectra that constitute an infrared microscopic image. Existing approaches involve physical models, which are non-linear in nature and lead to classifiers that do not generalize well, e.g. across different types of tissue preparation. Furthermore, existing preprocessing approaches involve iterative procedures that are computationally demanding, so that computation time required for preprocessing does not keep pace with recent progress in infrared microscopes which can capture whole-slide images within minutes., Results: We investigate the application of stacked contractive autoencoders as an unsupervised approach to preprocess infrared microscopic pixel spectra, followed by supervised fine-tuning to obtain neural networks that can reliably resolve tissue structure. To validate the robustness of the resulting classifier, we demonstrate that a network trained on embedded tissue can be transferred to classify fresh frozen tissue. The features obtained from unsupervised pretraining thus generalize across the large spectral differences between embedded and fresh frozen tissue, where under previous approaches separate classifiers had to be trained from scratch., Availability and Implementation: Our implementation can be downloaded from https://github.com/arnrau/SCAE&#95;IR&#95;Spectral&#95;Imaging., Supplementary Information: Supplementary data are available at Bioinformatics online., (© The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)

Objective: In this article, a craniocerebral model is introduced for neurosurgical training, which is patient-specific, tactile-realistic, and with adjustable intracranial pressure., Methods: The patient-specific feature is achieved by modeling from CT scans and magnetic resonance images (MRI). The brain tissue model is built by the hydrogel casting technique, while scalp, skull, vasculature, and lateral ventricles are all-in-one fabricated by three-dimensional (3D) printing. A closed-loop system is integrated to monitor and control the intracranial pressure. 3D measurements, mechanical tests, and simulated external ventricular drain (EVD) placement procedures are conducted on the model., Results: A neurosurgical training model is completed with high accuracy (mean deviation 0.36 mm). The hydrogel brain tissue has a stiffness more similar to that of a real brain than the common 3D printed materials. The elasticity modulus of hydrogel brain tissue model is E=25.71 kPa, compared with our softest 3D printed material with E=1.14×10 3  kPa. Ten experienced surgeons rate the tactile realness of the neurosurgical training model at an average point of 4.25 on a scale from 1 (strongly negative) to 5 (strongly positive). The neurosurgical training model is also rated to be realistic in size (4.82), anatomy (4.70), and effective as an aid to improve blind EVD placement skills (4.65)., Conclusions: The neurosurgical training model can provide trainee surgeons with realistic experience in both tactile feedbacks and craniocerebral anatomy, improving their surgical skills., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2020. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Current clinical guidelines for surgical repair of abdominal aortic aneurysms (AAAs) are primarily based on maximum diameter assessment. From a biomechanical point of view, not only the diameter but also peak wall stresses will play an important role in rupture risk assessment. These methods require patient specific geometry which typically uses computed tomography (CT) or magnetic resonance imaging. Recently, wall stress analysis based on 3D ultrasound (US) has been proposed, and shows promising results. However, the major limitations in these studies were the use of manual segmentation and the limiting field of view of US. Therefore in this study, the AAA is imaged with multiperspective 3D ultrasound, merged to obtain a large field of view, and afterwards automatically segmented. Geometry and wall stress results were validated using CT imaging., Methods: Three dimensional US and CT data were available for 40 AAA patients (maximum diameter 34-61 mm). The full US based AAA geometry was determined using automatic segmentation, and when the aneurysm exceeded a single 3D volume, automatic fusion of multiple 3D US volumes was used. Wall stress analysis was performed for all AAA patients and percentile wall stresses were derived. The accuracy of the US based geometry and wall stress prediction was measured by comparison with CT data., Results: Estimated geometries derived from 3D US and CT data showed good similarity, with an overall median similarity index (SI) of 0.89 and interquartile range of 0.87-0.92, whereas the median Hausdorff distances (HD), a measure for the maximum local mismatch, was 4.6 (4.0-5.9) mm for all AAA geometries. Thereby, the wall stress results based on merged multiperspective 3D US data revealed a greater similarity to CT than single 3D US data., Conclusion: This study showed that large volume geometry assessment of AAAs using multiperspective 3D ultrasound, segmentation and fusion, and wall stress analysis is feasible in a robust and labour efficient manner., (Copyright © 2019 European Society for Vascular Surgery. Published by Elsevier B.V. All rights reserved.)

Aims: The study aims to assess the accuracy of digital planning in dentistry, evaluating the characteristics of different intraoral 3D scanners and comparing it with traditional imaging 2D recording methods. Specifically, using computer aided design (CAD) software and measuring inside CAD software, authors want to verify the reliability of different models obtained with different techniques and machines., Methods: 12 patients that needed aesthetic restorative treatment were enrolled in the study. All the patients underwent recording data of the height and width dental elements 1.1, 1.2, and 1.3 size using different technologies and comparing 2D with 3D methods. A T test was then applied in order to verify whether there was a statistically significant difference between the measurements obtained, comparing the different tools data (Emerald, TRIOS, Photogrammetry and DSS (Digital Smile System)) with the reference values., Results: No significant differences emerged in the measurements made with the different scanners (Trios 3Shape ® , Planmeca Emerald ® ) and photogrammetry. Therefore, what should be underlined regarding the 2D measurements is the speed and simplicity compared to all 3D techniques, so this work can help to better define the field of application and the limits connected to 2D techniques, giving a good window of the technique., Conclusions: The low number of patients is not sufficient to provide statistically significant results, but the digital planning future prospects seem to be promising. This study results highlighted how a photogrammetric scanner for dental arches would only have a much smaller shooting field size and greater accuracy. Despite these considerations, the photogrammetric facial scanner provided excellent results for the measurement of individual teeth, showing a great versatility of use.

To develop a method to automatically determine intrafraction motion of the prostate based on soft tissue contrast on 3D cine-magnetic resonance (MR) images with high spatial and temporal resolution. Twenty-nine patients who underwent prostate stereotactic body radiotherapy (SBRT), with four implanted cylindrical gold fiducial markers (FMs), had cine-MR imaging sessions after each of five weekly fractions. Each cine-MR session consisted of 55 sequentially obtained 3D data sets ('dynamics') and was acquired over an 11 s period, covering a total of 10 min. The prostate was delineated on the first dynamic of every dataset and this delineation was used as the starting position for the soft tissue tracking (SST). Each subsequent dynamic was rigidly aligned to the first dynamic, based on the contrast of the prostate. The obtained translation and rotation describes the intrafraction motion of the prostate. The algorithm was applied to 6270 dynamics over 114 scans of 29 patients and the results were validated by comparing to previously obtained fiducial marker tracking data of the same dataset. Our proposed tracking method was also retro-perspectively applied to cine-MR images acquired during MR-guided radiotherapy of our first prostate patient treated on the MR-Linac. The difference in the 3D translation results between the soft tissue and marker tracking was below 1 mm for 98.2% of the time. The mean translation at 10 min were X: 0.0 [Formula: see text] 0.8 mm, Y: 1.0 [Formula: see text] 1.8 mm and Z: [Formula: see text] mm. The mean rotation results at 10 min were X: [Formula: see text], Y: 0.1 [Formula: see text] 0.6° and Z: 0.0 [Formula: see text] 0.7°. A fast, robust and accurate SST algorithm was developed which obviates the need for FMs during MR-guided prostate radiotherapy. To our knowledge, this is the first data using full 3D cine-MR images for real-time soft tissue prostate tracking, which is validated against previously obtained marker tracking data.

Three-dimensional printing is increasingly used in the health care industry. Making patient-specific anatomic task trainers has been one of the more commonly described uses of this technique specifically, allowing surgeons to perform complex procedures on patient-specific models in a nonoperative setting. With regard to transesophageal echocardiography (TEE) training, commercially available simulators have been increasingly used. Even though these simulators are haptic in nature and anatomically near realistic, they lack patient specificity and the training of the dynamic workflow and imaging protocol used in the operative setting. Herein a customized pulsatile left-sided heart model that uses patient-specific 3-dimensional printed valves under physiological intracardiac pressures as a TEE task trainer is described. With this model, dynamic patient-specific valvular anatomy can be visualized with actual TEE machines by trainees to familiarize themselves with the surgery equipment and the imaging protocol., (Copyright © 2019 Elsevier Inc. All rights reserved.)

A 3-dimensional (3D) convolutional neural network is presented for the segmentation and quantification of spontaneous intracerebral haemorrhage (ICH) in non-contrast computed tomography (NCCT). The method utilises a combination of contextual information on multiple scales for fast and fully automatic dense predictions. To handle a large class imbalance present in the data, a weight map is introduced during training. The method was evaluated on two datasets of 25 and 50 patients respectively. The reference standard consisted of manual annotations for each ICH in the dataset. Quantitative analysis showed a median Dice similarity coefficient of 0.91 [0.87-0.94] and 0.90 [0.85-0.92] for the two test datasets in comparison to the reference standards. Evaluation of a separate dataset of 5 patients for the assessment of the observer variability produced a mean Dice similarity coefficient of 0.95 ± 0.02 for the inter-observer variability and 0.97 ± 0.01 for the intra-observer variability. The average prediction time for an entire volume was 104 ± 15 seconds. The results demonstrate that the method is accurate and approaches the performance of expert manual annotation.

Background: Assimilating diagnostic radiology education into undergraduate medical education remains a challenge. This challenge places a greater emphasis in surgical residency to ensure this education occurs. The objective of this study is to determine whether a 3D-reconstruction may improve surgical residents' assessment of resectability of pancreatic lesions., Methods: Four cases were identified of patients with a pancreatic lesion; high-quality, triphasic abdominal CT scans were obtained and evaluated to ensure sufficient resolution and slice thickness. The images then were used for 3D-reconstruction of the cases., Results: PGY3-5 residents had a statistically-significant higher percentage of correct answers on objective questionnaire items using CT in conjunction with 3D-reconstruction software versus CT only. PGY1-2 residents had a higher percentage of correct answers using 3D-reconstruction software, but the difference was not statistically significant., Conclusions: 3D-reconstruction software could be a viable tool to augment radiology education within a surgery residency, especially in CT interpretation, but there appears to be a minimum threshold of knowledge needed for meaningful improvement; therefore, this software may be more useful for PGY3-5 residents than PGY1-2 residents., (Copyright © 2019 Elsevier Inc. All rights reserved.)

The conventional reading of the skin prick test (SPT) for diagnosing allergies is prone to inter- and intra-observer variations. Drawing the contours of the skin wheals from the SPT and scanning them for computer processing is cumbersome. However, 3D scanning technology promises the best results in terms of accuracy, fast acquisition, and processing. In this work, we present a wide-field 3D imaging system for the 3D reconstruction of the SPT, and we propose an automated method for the measurement of the skin wheals. The automated measurement is based on pyramidal decomposition and parametric 3D surface fitting for estimating the sizes of the wheals directly. We proposed two parametric models for the diameter estimation. Model 1 is based on an inverted Elliptical Paraboloid function, and model 2 on a super-Gaussian function. The accuracy of the 3D imaging system was evaluated with validation objects obtaining transversal and depth accuracies within ± 0.1 mm and ± 0.01 mm, respectively. We tested the method on 80 SPTs conducted in volunteer subjects, which resulted in 61 detected wheals. We analyzed the accuracy of the models against manual reference measurements from a physician and obtained that the parametric model 2 on average yields diameters closer to the reference measurements (model 1: -0.398 mm vs. model 2: -0.339 mm) with narrower 95% limits of agreement (model 1: [-1.58, 0.78] mm vs. model 2: [-1.39, 0.71] mm) in a Bland-Altman analysis. In one subject, we tested the reproducibility of the method by registering the forearm under five different poses obtaining a maximum coefficient of variation of 5.24% in the estimated wheal diameters. The proposed method delivers accurate and reproducible measurements of the SPT., Competing Interests: AGM, LAR, J. Marrugo and J, Meneses are co-inventors of a patent application related to this study, Dispositivo y método de reconstrucción 3D para la medición de pápulas en la piel. Colombian Patent App. NC2018/0007546; Inventors: Andres G. Marrugo, Lenny A. Romero, Javier Marrugo, Jaime Meneses. Assignee: Universidad Tecnologica de Bolivar, Universidad de Cartagena, Universidad Industrial de Santander. There are no further patents, products in development or marketed products to declare. This does not alter our adherence to all the PLOS ONE policies on sharing data and materials, as detailed online in the guide for authors.

Background: Anterior cervical discectomy and fusion (ACDF) remains one of the most commonly taught procedures during residency and one of the most frequently performed by neurosurgeons. Neurosurgeons use microscopes to perform surgery and to train other surgeons. Although the microscope provides excellent illumination and magnification, its use will be limited to 2 people: the surgeon and the assistant. Consequently, the scrub nurse and residents watching 2-dimensional images on monitors will have a reduced perception of the surgical field depth and anatomical details. The exoscope has been introduced as an alternative to microscopes and endoscopes. We used a 3-dimensional (3D), high-definition exoscope (3D Vitom [Karl Storz, Tuttlingen, Germany]) in 2 patients undergoing 2-level ACDF for cervical myeloradiculopathy., Methods: The exoscope was used during soft tissue dissection, discectomy, osteophytectomy, and cage insertion. Microsurgical drilling of the posterior osteophytes, which will usually require adequate magnification and proper microscope angulation, was also performed using the exoscope., Results: The exoscope provided a 3D view of the surgical field similar to that provided by a microscope and allowed us to effectively and safely perform the required surgical steps. The main advantage of 3D exoscope-assisted surgery, compared with microscope-assisted surgery, is the possibility to generate videos with a similar view and image quality as perceived by the surgeon. Therefore, the didactic capabilities of exoscopic videos are greater than those provided by microscopic videos. Exoscopes are also smaller compared with microscopes, allowing for comfortable use from the early surgical steps to device implantation., Conclusion: We believe that exoscope-assisted surgery could become a safe and effective alternative to microscope-assisted surgery in ACDF., (Copyright © 2019 Elsevier Inc. All rights reserved.)

The goal of objective stereoscopic image quality assessment (SIQA) is to predict the human perceptual quality of stereoscopic/3D images automatically and accurately. Compared with traditional 2D image quality assessment, the quality assessment of stereoscopic images is more challenging because of complex binocular vision mechanisms and multiple quality dimensions. In this paper, inspired by the hierarchical dual-stream interactive nature of the human visual system, we propose a stereoscopic image quality assessment network (StereoQA-Net) for no-reference stereoscopic image quality assessment. The proposed StereoQA-Net is an end-to-end dual-stream interactive network containing left and right view sub-networks, where the interaction of the two sub-networks exists in multiple layers. We evaluate our method on the LIVE stereoscopic image quality databases. The experimental results show that our proposed StereoQA-Net outperforms state-of-the-art algorithms on both symmetrically and asymmetrically distorted stereoscopic image pairs of various distortion types. In a more general case, the proposed StereoQA-Net can effectively predict the perceptual quality of local regions. In addition, cross-dataset experiments also demonstrate the generalization ability of our algorithm.

Background: To explore a new quantitative index to assess the matching performance of anatomical bone plates using digital technology., Methods: CT data of normal tibias of 40 adults were collected. Two brands of medial distal tibia plates were digitized. Two trained orthopedists simulated the surgical operation in Rhino 5.1 software by placing the plate curve on the medial distal tibia surface. The volume of the interstice between the plate curve and the bone surface was measured. The inverse value of this average interstice distance was used as the matching performance index (MPI). A wall thickness analysis tool was used to mark various interstice distances with varied colors., Results: The Kangli medial distal tibia plate had a MPI of 0.55 ± 0.08 by operator A and 0.55 ± 0.06 by operator B. The general care medial distal tibia plate had a MPI of 0.32 ± 0.06 by operator A and 0.31 ± 0.05 by operator B. There were significant variations in the MPI between the two types of plates by both operators (p < 0.001). And significant variations were observed in the MPI of general care medial distal tibia plates among various operator groups (p = 0.028)., Conclusion: This quantitative index of matching performance is straightforward and intuitive. However, we still need a method to improve the experimental repeatability, especially when it comes to a plate with poor matching performance.

Background: A novel acoustic coupling fluid (ACF), with the potential to reduce surgically induced image artefacts during intraoperative ultrasound imaging in brain tumour surgery, has been evaluated with respect to image quality and safety in a clinical phase 1 study., Methods: Fifteen patients with glioblastoma (WHO grade IV) were included. All adverse events were registered in a 6-month study period. During acquisition of 3D ultrasound image volumes, three different concentrations of the ACF and Ringer's solution were filled into the resection cavity. The effect of ACF on the ultrasound images was rated by the operating surgeon, and by five independent neurosurgeons evaluating a pair of blinded images from all patients. Images from all patients were analysed by comparing pixel brightness in a noise-affected region and a reference region., Results: The operating surgeon deemed the ACF images to have less noise than images obtained with Ringers's solution. The blinded evaluations by the independent neurosurgeons were significantly in favour of ACF (p < 0.0001). The analyses of pixel intensities showed that the ACF images had lower amount of noise than images obtained with Ringer's solution. No radiological sign of inflammation nor circulatory changes was found in the early postoperative MR images. Of the nine complications registered as serious events in the study period, none was deemed to be caused by the ACF., Conclusion: The ultrasound (US) images obtained using ACF have significantly less noise than US images obtained with Ringer's solution. The rate of adverse events was comparable to what has been reported for similar groups of patients.

Background: Three-dimensional (3D) endoscopy is an emerging tool in ear-nose-throat (ENT) and skull base surgery with the benefit of providing real-time depth perception. Several investigators have reported that the field of view (FOV) is reduced in 3D endoscopes compared with regular 2-dimensional (2D) endoscopes. Thus, we objectively compared the FOV of 2D and 3D endoscopes., Methods: Using a standard 2D and 2 different 3D ENT endoscopes, images were captured of 1-mm graph paper from a set distance of 6 cm. The FOV was calculated from these images and compared among the endoscopes., Results: The VisionSense 3D endoscope had a slightly smaller FOV (9.1 cm vs. 10.1 cm; -9.9%), and the Karl Storz 3D endoscope showed a slightly larger FOV (10.4 cm vs. 10.1 cm; +3.0%). However, the results were complicated by the different-shaped images produced by the 3D endoscopes., Conclusion: The differences in the FOV between the 2D and 3D endoscopes used in ENT surgery were not clinically significant. Thus, the FOV should not be considered a limitation of 3D technology., (Copyright © 2019 Elsevier Inc. All rights reserved.)

Background: The measurement of finger and wrist range of motion (ROM) is of great importance to clinicians when assessing functional outcomes of therapeutic interventions and surgical procedures. The purpose of the study was to assess the repeatability of ROM measurements of the hand joints with manual goniometer and 3D motion capture system and to calculate the minimal detectable difference for both methods., Methods: Active finger and wrist joints ROM of 20 healthy volunteers were assessed using a manual goniometer and 3D motion capture system. Minimal detectable difference (MDD) and standard error of measurement (SEM) were calculated for both measurement systems and compared within the same task. Maximal ROM of all joints was registered twice on two different days to evaluate the test-retest repeatability. The intraclass correlation coefficients (ICC) was calculated and examined to determine if reliability ≥ 0.70 existed., Results: MDD for the 3D motion capture was between 5 and 12° except for the metacarpophalangeal joint (MCP) 1, interphalangeal joint (IP), and MCP5. SEM values lay between 2 and 4° for all joints except for the MCP5, IP, and MCP1. For the goniometric measurements, MDD and SEM were between 12-30° and 4-11°, respectively. The reliability criterion (ICC > 0.7) was achieved for the ROM measurement with the 3D motion capture system for 94% of the joints and in only 65% of the joints with the manual goniometer., Conclusions: Joint ROM assessed with 3D motion analysis showed higher test-retest agreement demonstrating overall better repeatability for this method. Because of the smaller measurement error, the 3D motion capture system has a smaller MDD. Only individual test-rest differences bigger than the MDD can be considered as real changes, and therefore, in an experimental situation, the use of a more precise measurement method can greatly reduce the number of subjects needed for a statistical significance. Goniometer measurements of some joints should be carefully interpreted, due to a low repeatability and reliability., Trial Registration: This study is approved by the Ethical Committee Zurich ( Kek-ZH-Nr: 2015-0395 ).

Background: Tracking and predicting the growth performance of plants in different environments is critical for predicting the impact of global climate change. Automated approaches for image capture and analysis have allowed for substantial increases in the throughput of quantitative growth trait measurements compared with manual assessments. Recent work has focused on adopting computer vision and machine learning approaches to improve the accuracy of automated plant phenotyping. Here we present PS-Plant, a low-cost and portable 3D plant phenotyping platform based on an imaging technique novel to plant phenotyping called photometric stereo (PS)., Results: We calibrated PS-Plant to track the model plant Arabidopsis thaliana throughout the day-night (diel) cycle and investigated growth architecture under a variety of conditions to illustrate the dramatic effect of the environment on plant phenotype. We developed bespoke computer vision algorithms and assessed available deep neural network architectures to automate the segmentation of rosettes and individual leaves, and extract basic and more advanced traits from PS-derived data, including the tracking of 3D plant growth and diel leaf hyponastic movement. Furthermore, we have produced the first PS training data set, which includes 221 manually annotated Arabidopsis rosettes that were used for training and data analysis (1,768 images in total). A full protocol is provided, including all software components and an additional test data set., Conclusions: PS-Plant is a powerful new phenotyping tool for plant research that provides robust data at high temporal and spatial resolutions. The system is well-suited for small- and large-scale research and will help to accelerate bridging of the phenotype-to-genotype gap., (© The Author(s) 2019. Published by Oxford University Press.)

Background/aim: Methods to assess three-dimensionally the breast surface are increasingly used in plastic and reconstructive surgery. The aim of this study was to validate the use of the Structure Sensor 3D scanner (Occipital, Inc., Boulder, CO, USA) connected to an iPad Pro (Apple, Inc., Cupertino, CA, USA) as a novel, inexpensive and handheld three-dimensional scanning process., Materials and Methods: Surface images of a medical human female anatomy torso model of rigid plastic were repeatedly acquired with Structure Sensor 3D scanner and compared with those obtained using two clinically established 3D imaging systems. Digital measurements of vector and surface breast distances were analyzed using Mimics® Innovation Suite 20 medical imaging software (Materialise, Leuven, Belgium)., Results: The analysis of variance (ANOVA) revealed no statistically significant difference among measurements obtained using different scanning processes for all the variables examined (p>0.05)., Conclusion: The study demonstrates analogous practicability and reliability for surface image acquisition using the newly introduced Structure Sensor 3D scanner and other clinically established scanners., (Copyright© 2019, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.)

Background: Neurological development is determined by brain growth. Methods to measure total brain volume (TBV) in clinical settings are limited. MR-imaging represents the gold-standard., Aims: The present study tests the hypotheses that in infants without any brain pathology, TBV - as determined by MRI - can be accurately estimated by cranial volume (CrV), measured by 3D-laser scanning. In case of good correlation of CrV with TBV it was further tested, whether CrV can be also estimated by (I) head circumference (HC) or (II) by other technology than laser scanning., Study Design & Subjects: To test the hypothesis, that TBV can be reliably estimated by CrV-measurement, data from routine MRI and 3D-laser-scanner measurements were analyzed in infants if no major structural brain anomaly was found in MR-imaging. To test whether CrV can be predicted by HC-measurements during infancy, data from routine follow-up visits were used from preterms born in a two year period. Preterms are invited for a routine follow-up visit (which includes laser scanning of the head) at an age of 3 months and, for further follow-ups at 6, 9 or 12 months. To compare accuracy of CrV measurement by other techniques, a puppet head was measured, using different 3D-measurement principles: (i) Structured light projection system, (ii) The non-invasive laser-shape-digitizer, and (iii) Structure-from-motion (SFM) technique., Outcome Measures: TBV was compared with CrV using a Passing-Bablok-Regression. To determine how well HC predicts CrV, the coefficient of determinant (R 2 ) was calculated for each age group., Results: CrV and TBV of 25 infants (median age 19 month, body weight of 11 kg) showed a median bias of -86.7 mm 3 with a slightly smaller TBV (median of 1034.1 mm 3 , IQR 875.9 … 1179.8 mm 3 ) than CrV (median 1092.2, IQR 950.5 … 1258.4 mm 3 ). CrV was poorly estimated by HC, with R 2 between 0.79 and 0.87 at 3 and 9 month of age respectively. For the non-invasive laser-shape-digitizer and the SFM-technique the accuracy was good (radial coordinate differences ±0.3 vs. ±0.5 mm)., Conclusion: The present study provides convincing evidence that CrV can be used to estimate TBV in routine care, whereas HC is a poor predictor of individual CrV., (Copyright © 2019 Elsevier B.V. All rights reserved.)

Background: Minimally-invasive sacrocolpopexy is the gold standard procedure for advanced apical prolapse. Nonetheless, sacrocolpopexy has potential serious complications leading many surgeons to avoid this excellent surgical procedure. To overcome these limitations, preoperative planning with 3D models of the female pelvis is proposed. The aim of the study is to evaluate the feasibility of pelvic anatomy reconstruction with the ITK-SNAP software and highlight its potential benefits in this intervention., Methods: Thirty patient-specific 3D models of the female pelvis were created using ITK-SNAP and the EndoCAS Segmentation Pipeline extension for image segmentation: contrast-enhanced computed tomography (CE-CT) data sets of women who underwent examinations for reasons other than prolapse were used. The distances of pelvic structures from the sacral promontory were standardised and measured, and correlations among these distances were evaluated with Spearman's correlation coefficient., Results: Pelvic anatomy reconstruction was feasible for all CE-CT data sets. A statistically significant correlation was found between the distances of the cava bifurcation and common iliac vessels from the sacral promontory. An area for proximal mesh attachment was defined: it is free from the passage of iliac vessels in 97.5% of cases. A significant statistical correlation was found between the distances of the midpoint of the bispinous diameter and the uterine cervix from the sacral promontory; a process of linear regression showed that the latter measure can be estimated by multiplying the first one by 0.86., Conclusions: Pre-surgical 3D reconstructions of the female pelvis using ITK-SNAP could help achieve widespread use of sacrocolpopexy: further comparative studies are needed to evaluate the outcomes with and without their use.

Background: External ventricular drain (EVD) placement is one of the most commonly performed procedures in neurosurgery, frequently by the junior neurosurgery resident. Simulators for EVD placement are often costly, time-intensive to create, and complicated to set up., Objective: To describe creation of a simulator that is inexpensive, time-efficient, and simple to set up., Methods: This simulator involves printing a hollow head using a desktop 3-dimensional (3D) printer. This head is registered to a commercially available image-guidance system. A total of 11 participants volunteered for this simulation module. EVD placement was assessed at baseline, after verbal teaching, and after live 3D view instruction., Results: Accurate placement of an EVD on the right side at the foramen of Monro or the frontal horn of the lateral ventricle increased from 44% to 98% with training. Similarly, accurate placement on the left increased from 42% to 85% with training., Conclusion: During participation in the simulation, accurate placement of EVDs increased significantly. All participants believed that they had a better understanding of ventricular anatomy and that this module would be useful as a teaching tool for neurosurgery interns., (Copyright © 2018 by the Congress of Neurological Surgeons.)

This paper describes an automated method for extracting the apparent positions of fiducial points from 2D or 3D images of a phantom. We consider a 3D-lattice phantom for two main reasons: first, ease of manufacture and isotropy of its structure with respect to coordinate projections; second, a connected structure allowing to uniquely assess the adjacency relationship between fiducial points even if geometric distortions arising from main magnet inhomogeneity and gradient fields non-linearity is severe as observed in open-bore systems. In order to validate our proposed method and compare different choices for the parameters of our phantom (i.e. number and distance between grids and thickness of its branches) we developed in-house a software for simulating 2D or 3D volume images of the phantom, using customizable MRI sequence parameters and Spherical Harmonic Coefficients for the fields. We deem worthy of note that using simulated images is the only way to evaluate the estimated position error, since it allows to compare the estimates to their theoretical counterparts. Furthermore, the use of simulated images allows to evaluate the robustness of the method with respect to image quality in terms of Signal-to-Noise Ratios and geometric distortion, and allows to evaluate different phantom geometries without having to manufacture them. The proposed method can be easily extended to phantoms having an arbitrary overall shape, as long as it is a fully connected structure. Specifically, it is easy to design a phantom with fiducial points laying outside of the homogeneity sphere, so that indirect measurement of the fields becomes possible, for example by using the recent method proposed by Acquaviva et al. To the best of our knowledge, the proposed method outperforms other state-of-the-art methods, with an average positioning offset of 0.052 mm (with a 0.99 quantile of 0.12 mm) when working on images featuring a differential Signal-to-Noise Ratio within Region-of-Interest (ROI) equal to 105 (20.2 dB) and a ROI-to-background SNR of 20 dB. Estimating the positions of 6859 fiducial points in a volume, our highest density case, was carried out in less than 30 min on a desktop personal computer., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Purpose of the Study: Medical field has highly evolved with advancements in the technologies which prove to be beneficial for radiologists and patients for better diagnosis. The era of medical science provides best healthcare solutions with the help of medical images. Till now, 2D MRIs played a prominent role in early detection of disease but with latest technologies taking over the charge, 3D MRIs are highly effective and great in demand nowadays. With the aid of advanced techniques such as edge detection, segmentation and texture analysis on these images, the disease detection may become much easier., Materials and Methods: Texture of any image is recognized by distribution of gray levels in the neighborhood. The Texture Analysis plays an important role in study of medical images. It identifies the prominent features of an image and highlights the same using different feature extraction technique. In this paper, 3D MRI of human brain is considered and texture analysis based on Haralick's and GLCM texture features is performed. Haralick's feature explains the image intensities of each pixel and their relationship with neighborhood pixels. The entire data set consists of 40 brain tumor patients, out of which a sample has been depicted., Results: The analysis of different features such as Contrast, Correlation, Energy, Homogeneity and Entropy is carried out., Conclusion: Further, the study highlights about the highly useful features for early detection of brain tumor disease.

The aim of this study was to evaluate the factors affecting intra-oral scanner accuracy by analyzing variation in measurements of a dental model according to scanning distance. A dental cast, including a prepared left mandibular first molar, was used. Rectangular frames measuring 20 × 30 mm with heights of 2.5, 5.0, and 7.5 mm were made. The model was scanned 10 times with a reference scanner to obtain the true value. Scanning was performed 10 times at four distances of 0, 2.5, 5.0, and 7.5 mm with the frame of each height using the following intra-oral scanners: TRIOS; CS 3500; and PlanScan. In the linear distance measurement method (2D), measurements were taken at five parameters using the Rapidform software. In the best-fit alignment method (3D), using the Geomagic Control X, the root mean square values of the two scan images were calculated. In the 2D comparison, the different from the reference value was the smallest at 2.5 and 5.0 mm. In the 3D comparison, 2.5 and 5.0 mm were the most accurate, and 0 mm was the least accurate among the four distances. To the best of our knowledge, this study was the first to evaluate the accuracy of scanning distances, and found a difference between the accuracy of the scanning distance and the accuracy of the scanner. Moreover, the results of this study indicated that the scanning distance was a variable affecting accuracy. Clin. Anat. 32:430-438, 2019. © 2019 Wiley Periodicals, Inc., (© 2019 Wiley Periodicals, Inc.)

Purpose: Manganese-enhanced magnetic resonance imaging (MEMRI) can help us trace the active neurons and neuronal pathway in transgenic mouse AD model. 5XFAD has been widespread accepted as a valuable model system for studying brain dysfunction progresses in the courses of AD. To further understand the development of AD at early stages, an effective and objective data analysis platform for MEMRI studies should be constructed., Materials and Methods: A set of stereotaxic templates of mouse brain in Paxinos and Franklin space, "the Institute of High Energy Physics Mouse Template", or IMT for short, was constructed by iteratively registration and averaging. An atlas image was reconstructed from the Paxinos and Franklin atlas figures and each sub-anatomical segmentation was assigning a unique integer. An analysis SPM plug-in toolbox was further created, that automates and standardizes the time-consuming processes of brain extraction, tissue segmentation, and statistical analysis for MEMRI scans., Results: The IMT comprised a T2WI template image, a MEMRI template image, intracranial tissue segmentations, and accompany with a digital mouse brain atlas image, in which 707 sub-anatomical brain regions are delineated. Data analyses were performed on groups of developing 5XFAD mice to demonstrate the usage of IMT, and the results shows that abnormal neuronal activity occurs at early stage in 5XFAD mice., Conclusion: We have constructed a stereotaxic template set of mouse brain named IMT with fine delineations of sub-anatomical structures, which is compatible with SPM. It will give a widely range of researchers a standardized coordinate system for localization of any mouse brain related data., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Background: The surgical microscope is used primarily for microsurgeries, which are more complicated than other surgical procedures and require delicate tasks for a long time. Therefore, during these surgical procedures, surgeons experience back and neck pain. To solve this problem, new technology, such as wearable displays, is required to help surgeons maintain comfortable postures and enjoy advanced functionality during microsurgery., Objective: The objective of this study was to develop a surgical microscope system that would work with wearable devices. It would include a head-mounted display (HMD) that can offer 3D surgical images and allow a flexible and comfortable posture instead of fixed eyepieces of surgical microscope and can also provide peripheral visual field with its optical see-through function., Methods: We designed and fabricated a surgical microscope system that incorporates a see-through type 3D HMD, and we developed an image processing software to provide better image quality. The usability of the proposed system was confirmed with preclinical examination. Seven ENT (ear, nose, and throat) surgical specialists and 8 residents performed a mock surgery-axillary lymph node dissection on a rat. They alternated between looking through the eyepieces of the surgical microscope and viewing a 3D HMD screen connected to the surgical microscope. We examined the success of the surgery and asked the specialists and residents to grade eye fatigue on a scale of 0 (none) to 6 (severe) and posture discomfort on a scale of 1 (none) to 5 (severe). Furthermore, a statistical comparison was performed using 2-tailed paired t test, and P=.00083 was considered significant., Results: Although 3D HMD case showed a slightly better result regarding visual discomfort (P=.097), the average eye fatigue was not significantly different between eyepiece and 3D HMD cases (P=.79). However, the average posture discomfort, especially in neck and shoulder, was lower with 3D HMD display use than with eyepiece use (P=.00083)., Conclusions: We developed a see-through type 3D HMD-based surgical microscope system and showed through preclinical testing that the system could help reduce posture discomfort. The proposed system, with its advanced functions, could be a promising new technique for microsurgery., (©Cheol-Hwan Kim, Seon-Young Ryu, Ji-Young Yoon, Hyoung-Kwon Lee, Nak-Gu Choi, Il-Ho Park, Hae-Young Choi. Originally published in JMIR Mhealth and Uhealth (http://mhealth.jmir.org), 07.03.2019.)

Background: Use of 3-dimensional (3D) computed tomography (CT) preoperative planning and patient-specific instrumentation has been demonstrated to improve the accuracy of glenoid implant placement in total shoulder arthroplasty (TSA). The purpose of this study was to compare the accuracy of glenoid implant placement in primary TSA among different types of instrumentation used with the 3D CT preoperative planning., Methods: One hundred and seventy-three patients with end-stage glenohumeral arthritis were enrolled in 3 prospective studies evaluating patient-specific instrumentation and 3D preoperative planning. All patients underwent preoperative 3D CT planning to determine optimal glenoid component and guide pin position based on surgeon preference. Patients were placed into 1 of 5 instrument groups used for intraoperative guide pin placement: (1) standard instrumentation, (2) standard instrumentation combined with use of a 3D glenoid bone model containing the guide pin, (3) use of the 3D glenoid bone model combined with single-use patient-specific instrumentation, (4) use of the 3D glenoid bone model combined with reusable patient-specific instrumentation, and (5) use of reusable patient-specific instrumentation with an adjustable, reusable base. Postoperatively, all patients underwent 3D CT to compare actual versus planned glenoid component position. Deviation from the plan (in terms of orientation and location) was compared across groups on the basis of absolute differences and outlier analysis. Univariable and multivariable comparisons were performed. As the initial analyses showed no significant differences in preoperative factors or in deviation from the plan between Groups 1 and 2 or between Groups 4 and 5 across studies, the final analysis was across 3 major treatment groups: standard instrumentation (Groups 1 and 2), single-use patient-specific instrumentation (Group 3), and reusable patient-specific instrumentation (Groups 4 and 5)., Results: In nearly all comparisons, there were no significant differences in the deviation from the plan (absolute differences or outlier frequency) for glenoid implant orientation or location across the 3 major treatment groups., Conclusions: This study did not demonstrate that any type of patient-specific instrumentation resulted in consistent differences in accuracy of glenoid implant placement in primary TSA with 3D CT preoperative planning. Surgeons have multiple patient-specific instrumentation options available for improving accuracy of glenoid implant placement when compared with 2D imaging without patient-specific instrumentation., Level of Evidence: Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.

Medical image classification is currently a challenging task that can be used to aid the diagnosis of different brain diseases. Thus, exploratory and discriminative analysis techniques aiming to obtain representative features from the images play a decisive role in the design of effective Computer Aided Diagnosis (CAD) systems, which is especially important in the early diagnosis of dementia. In this work, we present a technique that allows using specific time series analysis techniques with 3D images. This is achieved by sampling the image using a fractal-based method which preserves the spatial relationship among voxels. In addition, a method called Empirical functional PCA (EfPCA) is presented, which combines Empirical Mode Decomposition (EMD) with functional PCA to express an image in the space spanned by a basis of empirical functions, instead of using components computed by a predefined basis as in Fourier or Wavelet analysis. The devised technique has been used to classify images from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Parkinson Progression Markers Initiative (PPMI), achieving accuracies up to 93% and 92% differential diagnosis tasks (AD versus controls and PD versus Controls, respectively). The results obtained validate the method, proving that the information retrieved by our methodology is significantly linked to the diseases.

Purpose: To assess safety, efficacy, and outcomes of vitreoretinal surgery for macular pathology using a 3-dimensional heads-up display (3D HUD) surgical platform compared with a standard operating microscope (SOM)., Design: Prospective, single-center, unmasked, randomized study., Participants: Patients undergoing pars plana vitrectomy (PPV) for epiretinal membrane (ERM) or full-thickness macular hole (MH) at Wills Eye Hospital., Methods: Patients were randomized 1:1 to undergo surgery with a 3D HUD surgical platform or SOM. Patients who had previous PPV were excluded. Surgical choices, including PPV gauge, were based on surgeon preference. Standard surgical safety parameters, Early Treatment Diabetic Retinopathy Study visual acuity (VA), minimum required endoillumination levels, operative times, and surgeon "ease of use" of the viewing platform were recorded. Patients were followed up to postoperative month 3 (POM3)., Main Outcome Measures: The main outcome measures were total operative time, macular peel time, surgeon rating of viewing system ease of use, minimum required endoillumination, intraoperative complication rate, and postoperative VA., Results: Thirty-nine eyes of 39 patients with a mean age of 67.60±8.21 SD years were enrolled. Indications included ERM (n = 26 [3D HUD = 14, SOM = 12]) and MH (n = 13 [3D HUD = 9, SOM = 4]). Minimum required endoillumination was significantly lower with 3D HUD (mean 22.70%±15.10% SD) compared with SOM (mean 39.06%±2.72%; P < 0.001). There was no significant difference in overall operative time, but macular peel time was significantly longer using 3D HUD (mean 14.76±4.79 minutes) than SOM (11.87±8.07 minutes; P = 0.004). Surgeon-reported ease of use was significantly higher (easier) using SOM compared with 3D HUD (P = 0.004). There was no statistically significant difference between the groups in POM3 logarithm of the minimum angle of resolution (logMAR) VA or change in logMAR VA from baseline (all P > 0.681). There were no clinically significant intraoperative complications in either group., Conclusions: Three-dimensional heads-up display surgical visualization is an evolving technology demonstrating comparable efficacy to the SOM for macular surgery. Although overall surgical times were similar, 3D HUD macular peel times were longer and associated with less ease of use in this study, which may partly be due to a learning curve with new technology., (Copyright © 2018 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.)

This paper presents a new approach to automatic three-dimensional (3D) cephalometric annotation for diagnosis, surgical planning, and treatment evaluation. There has long been considerable demand for automated cephalometric landmarking, since manual landmarking requires considerable time and experience as well as objectivity and scrupulous error avoidance. Due to the inherent limitation of two-dimensional (2D) cephalometry and the 3D nature of surgical simulation, there is a trend away from current 2D to 3D cephalometry. Deep learning approaches to cephalometric landmarking seem highly promising, but there exist serious difficulties in handling high dimensional 3D CT data, dimension referring to the number of voxels. To address this issue of dimensionality, this paper proposes a shadowed 2D image-based machine learning method which uses multiple shadowed 2D images with various lighting and view directions to capture 3D geometric cues. The proposed method using VGG-net was trained and tested using 2700 shadowed 2D images and corresponding manual landmarkings. Test data evaluation shows that our method achieved an average point-to-point error of 1.5 mm for the seven major landmarks.

Background: With application of 3T magnetic resonance imaging (MRI) to functional neurosurgery procedures and given the inherent requirement of millimetric precision, the need to develop a method for correction of geometric image distortion emerged. The aim of this study was to demonstrate clinical safety and practical viability of a correction protocol in patients scheduled to undergo stereotactic procedures using 3T MRI., Methods: This prospective study comprised 20 patients scheduled to undergo computed tomography (CT) stereotactic functional procedures or encephalic brain lesion biopsies. The CT images were references for MRI geometric accuracy calculations. For each scan, 2 images were obtained: normal and reversed images. Eight distinct points on CT and MRI were selected summing 152 points that were based on a power analysis calculation value >0.999. One patient was excluded because of the inability to find reliable common landmark points on CT and MRI., Results: The distortion range was 0-5.6 mm and increased proportionally with stereotactic isocenter distance, meaning the distortion was greater in the periphery. After correction, the minimum and maximum distortion found was 0 mm and 3.5 mm, respectively. There was no significant difference between CT and MRI corrected x-coordinates (P > 0.05)., Conclusions: The proposed method can satisfactorily correct geometric distortions in clinical 3T MRI studies. Clinical use of the technique can be practical and efficient after software automation of the process. The method can be applied to all spin-echo MRI sequences., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Based on a recently introduced immunohistochemical panel (Bachmann et al. 2015) for aganglionic megacolon (AM), also known as Hirschsprung disease, histopathological diagnosis, we evaluated whether the use of digital pathology and 'machine learning' could help to obtain a reliable diagnosis. Slides were obtained from 31 specimens of 27 patients immunohistochemically stained for MAP2, calretinin, S100β and GLUT1. Slides were digitized by whole slide scanning. We used a Definiens Developer Tissue Studios as software for analysis. We configured necessary parameters in combination with 'machine learning' to identify pathological aberrations. A significant difference between AM- and non-AM-affected tissues was found for calretinin (AM 0.55% vs. non-AM 1.44%) and MAP2 (AM 0.004% vs. non-AM 0.07%) staining measurements and software-based evaluations. In contrast, S100β and GLUT1 staining measurements and software-based evaluations showed no significant differences between AM- and non-AM-affected tissues. However, no difference was found in comparison of suction biopsies with resections. Applying machine learning via an ensemble voting classifier, we achieved an accuracy of 87.5% on the test set. Automated diagnosis of AM by applying digital pathology on immunohistochemical panels was successful for calretinin and MAP2, whereas S100β and GLUT1 were not effective in diagnosis. Our method suggests that software-based approaches are capable of diagnosing AM. Our future challenge will be the improvement of efficiency by reduction of the time-consuming need for large pre-labelled training data. With increasing technical improvement, especially in unsupervised training procedures, this method could be helpful in the future.

Background: Precise robotic or stereotactic implantation of stereoelectroencephalography (sEEG) electrodes relies on the exact referencing of the planning images in order to match the patient's anatomy to the stereotactic device or robot. We compared the accuracy of sEEG electrode implantation with stereotactic frame versus laser scanning of the face based on computed tomography (CT) or magnetic resonance imaging (MRI) datasets for referencing., Methods: The accuracy was determined by calculating the Euclidian distance between the planned trajectory and the postoperative position of the sEEG electrode, defining the entry point error (EPE) and the target point error (TPE). The sEEG electrodes (n = 171) were implanted with the robotic surgery assistant (ROSA) in 19 patients. Preoperative trajectory planning was performed on three-dimensional (3D) MRI datasets. Referencing was accomplished either by performing (A) 1.25-mm slice CT with the patient's head fixed in a Leksell stereotactic frame (CT-frame, n = 49), fused with a 3D-T1-weighted, contrast enhanced- and T2-weighted 1.5 Tesla (T) MRI; (B) 1.25 mm CT (CT-laser, n = 60), fused with 3D-3.0-T MRI; (C) 3.0-T MRI T1-based laser scan (3.0-T MRI-laser, n = 56) or (D) in one single patient, because of a pacemaker, 3D-1.5-T MRI T1-based laser scan (1.5-T MRI-laser, n = 6)., Results: In (A) CT-frame referencing, the mean EPE amounted to 0.86 mm and the mean TPE amounted to 2.28 mm (n = 49). In (B) CT-laser referencing, the EPE amounted to 1.85 mm and the TPE to 2.41 mm (n = 60). In (C) 3.0-T MRI-laser referencing, the mean EPE amounted to 3.02 mm and the mean TPE to 3.51 mm (n = 56). In (D) 1.5-T MRI, surprisingly the mean EPE amounted only to 0.97 mm and the TPE to 1.71 mm (n = 6). In 3 cases using CT-laser and 1 case using 3.0 T MRI-laser for referencing, small asymptomatic intracerebral hemorrhages were detected. No further complications were observed., Conclusion: Robot-guided sEEG electrode implantation using CT-frame referencing and CT-laser-based referencing is most accurate and can serve for high precision placement of electrodes. In contrast, 3.0-T MRI-laser-based referencing is less accurate, but saves radiation. Most trajectories can be reached if alternative routes over less vascularized brain areas are used. This article is part of the Special Issue "Individualized Epilepsy Management: Medicines, Surgery and Beyond"., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Background: Frameless stereotactic surgery utilizing fiducial-based (FB) registration is an established tool in the armamentarium of deep brain stimulation (DBS) surgeons. Fiducial-less (FL) registration via intraoperative CT, such as the O-arm, has been routinely used in spine surgery, but its accuracy for DBS surgery has not been studied in a clinical setting., Objective: We undertook a study to analyze the accuracy of the FL technique in DBS surgery and compare it to the FB method., Methods: In this prospective cohort study, 97 patients underwent DBS surgery using the NexFrame and the O-arm registration stereotactic system. Patients underwent FB (n = 50) registration from 2015 to 2016 and FL (n = 47) O-arm registration from 2016 to 2017., Results: The radial errors (RE) and vector/euclidean errors of FB and FL registration were not significantly different. There was no difference in additional passes between methods, but there was an increase in the number of RE ≥2.5 mm in the FL method., Conclusion: Although there was no statistically significant difference in RE or the need for additional passes, the increased number of errors ≥2.5 mm with the FL method (17 vs. 4% in FB) indicates the need for further study. We concluded that O-arm images of the implants should be utilized to assess and correct for this error., (© 2019 S. Karger AG, Basel.)

Background: Many surgeons utilize assistants to perform procedures in more than one operating room at a given time using a practice known as overlapping surgery. Debate has continued as to whether overlapping surgery improves the efficiency and access to care or risks patient safety and outcomes., Objective: To examine effects of overlapping surgery in deep brain stimulation (DBS) for movement disorders., Methods: In this retrospective analysis of overlapping and non-overlapping cases, we evaluated stereotactic accuracy, operative duration, length of hospital stay, and the presence of hemorrhage, wound-related complications, and hardware-related complications requiring revision in adults with movement disorders undergoing DBS., Results: Of 324 cases, 141 (43.5%) were overlapping and 183 (56.5%) non-overlapping. Stereotactic error, number of brain penetrations, and postoperative length of hospitalization did not differ significantly (p ≥ 0.08) between the overlapping and non-overlapping groups. Mean operative duration was significantly longer for overlapping (81/141 [57.4%], 189.5 ± 10.8 min) than for non-overlapping cases (79/183 [43.2%], 169.9 ± 7.6 min; p = 0.004). There were no differences in rates of wound-related complications or hemorrhages, but overlapping cases had a significantly higher rate of hardware-related complications requiring revision (7/141 [5.0%] vs. 0/183 [0%]; p = 0.002)., Conclusions: Overlapping and non-overlapping cases had comparable DBS lead placement accuracy. Overlapping cases had a longer operative duration and had a higher rate of hardware-related complications requiring revision., (© 2019 S. Karger AG, Basel.)

Wound area measurements provide an objective assessment of wound healing; however, most commonly used measurement techniques are imprecise., Purpose: A new portable 3-dimensional (3D) wound measurement device was tested against laser- and hand-measurement methods., Methods: A retrospective comparative analysis was conducted to analyze the difference in wound measurements using records of patients seen at the University of Miami Hospital (Miami, FL) outpatient wound healing clinic between November 2017 and February 2018 who had wounds of various etiologies measured using 3 different techniques during a single visit: the 3D device, a laser-assisted wound measurement device (laser), and standard hand measurements. Patients with circumferential wounds were excluded (the laser and 3D devices are incapable of assessing these wounds). Differences were compared using paired t tests., Results: The wounds ranged in area from 0.8 cm² (hand measurements) and 0.2 cm² (3D and laser devices) to 100.94 cm², 61.9 cm², and 65 cm² by hand measurement, 3D, and laser device, respectively. Among the 23 wounds measured, the majority (16) were venous ulcers. No statistically significant difference was noted between the 3D measurements compared with the laser (P = .340). Statistically significant differences in the measurements between the 3D device and hand measurements (P = .008) and the laser device and hand measurements (P = .006) were found., Conclusion: Measurements of the 3D device appear analogous to laser devices, making it an alternative tool for clinicians interested in monitoring wound progression. Because the 3D device has the capacity to examine wound volume, prospective comparative trials should be used to examine the accuracy and precision of the device to measure volume.

Objective: Stereoelectroencephalography (SEEG) was first developed in the 1950s by Jean Talairach using 2D angiography and a frame-based, orthogonal approach through a metallic grid. Since then, various other frame-based and frameless techniques have been described. In this study the authors sought to compare the traditional orthogonal Talairach 2D angiographic approach with a frame-based 3D robotic procedure that included 3D angiographic interoperative imaging guidance. MRI was used for both procedures during surgery, but MRI preplanning was done only in the robotic 3D technique., Methods: All study patients suffered from drug-resistant focal epilepsy and were treated at the same center by the same neurosurgical team. Fifty patients who underwent the 3D robotic procedure were compared to the same number of historical controls who had previously been successfully treated with the Talairach orthogonal procedure. The effectiveness and absolute accuracy, as well as safety, of the two procedures were compared. Moreover, in the 3D robotic group, the reliability of the preoperative MRI to avoid vascular structures was evaluated by studying the rate of trajectory modification following the coregistration of the intraoperative 3D angiographic data onto the preoperative MRI-based trajectory plans., Results: Effective accuracy (96.5% vs 13.7%) and absolute accuracy (1.15 mm vs 4.00 mm) were significantly higher in the 3D robotic group than in the Talairach orthogonal group. Both procedures showed excellent safety results (no major complications). The rate of electrode modification after 3D angiography was 43.8%, and it was highest for frontal and insular locations., Conclusions: The frame-based, 3D angiographic, robotic procedure described here provided better accuracy for SEEG implantations than the traditional Talairach approach. This study also highlights the potential safety advantage of trajectory planning using intraoperative frame-based 3D angiography over preoperative MRI alone.

MRI observations following in vivo administration of Mn 2+ [manganese (Mn)-enhanced MRI, MEMRI] have been used as an excellent morphological and functional MRI tool for in vivo preclinical studies. To detect brain three-dimensional (3D) microstructures, we improved the ex vivo MEMRI method for mouse brains after in vivo Mn administration and obtained high-resolution MRIs using a cryogenic radiofrequency (RF) coil. Male C57BL/6 mice ( n = 8) were injected with 50 mM MnCl 2 intravenously and MEMRIs of the brain were acquired in vivo after 24 h, followed by perfusion fixation with a 4% paraformaldehyde (PFA) solution. High-resolution 25-μm isotropic MRIs were successfully acquired from the extracted brain tissue and could identify the brain microstructures, especially in the hippocampus [the pyramidal cell layer through CA1-3 and the dentate gyrus (DG) granular layers (GLs)], cell layers of cerebellum, three sub-regions of the deep cerebellar nucleus, and white matter (WM) structures [e.g., the fasciculus retroflexus (fr) and optic tract in the thalamus]. The following technical conditions were also examined: (i) the longitudinal stability of Mn-enhanced ex vivo tissue after in vivo administration; and (ii) the effects of mixing glutaraldehyde (GA) with the fixative solution for the preservation of in vivo MEMRI contrast. Our results indicate that ex vivo MEMRI observations made shortly after fixation maintain the contrast observed in vivo . This research will be useful for non-destructive whole-brain pathological analysis.

Background: In the past, measurement of upper limb lymphedema was done by water displacement (WD), which is frequently cited as the gold standard. For various reasons, however, the use of WD is restricted in clinical settings. A more precise and easy-to-use method would be favorable. The high precision of three-dimensional (3D) imaging in comparison to WD has already been reported for healthy subjects. The aim of this study is to determine the validity and reliability of 3D imaging by comparing it to the WD method in women with unilateral upper limb lymphedema. Methods and Results: Thirty-nine women with unilateral breast cancer-related lymphedema (BCRL) were included, of which 37 completed two volume measurement techniques (3D and WD) on the BCRL and contralateral healthy arm. Slightly larger volumes were measured by the WD method in healthy arms (+9.8 mL; p  = 0.058) and also in BCRL arms (+18.5 mL; p  < 0.001). All measurements were performed twice by the same researcher to evaluate reliability. There was no significant difference between the two measurements for healthy arms ( p  = 0.323) or BCRL arms ( p  = 0.807) in 3D imaging. Bland-Altman plots showed a high limit of agreement between the single measurements. 3D imaging had a high intrarater reliability (Intraclass Correlation Coefficient = 0.999). Conclusion: Results show that 3D imaging is an innovative method for measuring upper limb volume in BCRL patients. Even though image processing is time consuming, 3D imaging combines high reproducibility with high precision. By software automation, this technique could easily be integrated into clinical routine. It is for this reason that we would recommend implementing the Vectra 3D imaging technique for measurement of BCRL.

Introduction: A novel real-time lesion size index (LSI) that incorporates contact force (CF), time, and power has been developed for safe and effective catheter ablation. The optimal LSI was evaluated to eliminate gap formation during pulmonary vein isolation (PVI)., Methods and Results: Consecutive patients were enrolled, who underwent their first PVI using a fiber-optic CF-sensing catheter for atrial fibrillation between December 2016 and October 2017. The CF parameters, force-time integral (FTI), and LSI for 3095 ablation points in 34 patients were evaluated. The FTI and LSI in the lesions with gaps or dormant conduction (gaps/DC) were significantly lower than those in the lesion without gaps/DC (FTI: 140.5 ± 54.5 and 232.4 ± 121.4 g s, P < 0.0001; LSI: 4.0 ± 0.6 and 4.7 ± 0.9, P < 0.0001, respectively). On receiver operating characteristic curve analysis, the optimal LSI threshold was 4.05 (sensitivity, 63.4%; specificity, 76.3%). The LSI of <5.25 predicted a gap or DC with a high sensitivity (sensitivity, 97.6%; specificity, 25.7%). In the posterior wall, which was 37% thinner than the nonposterior wall, a lower LSI of <3.95 showed a relatively high sensitivity (92.3%) and specificity (65.6%)., Conclusions: The LSI can be used to predict gaps/DC during the PVI procedure. An LSI of 5.2 may be a suitable target for effective lesion formation. An LSI of 4.0 may be acceptable in the posterior wall, especially in areas adjacent to the esophagus., (© 2018 Wiley Periodicals, Inc.)

Purposes: Conventionally, lower limb circumference is measured using a measuring tape; however, a measuring tape may not provide sufficient precision to measure the magnitude of the lymphedema. We report the use of three-dimensional photography (VECTRA ® ) for the evaluation of lymphedema in patients after lymphovenous anastomosis (LVA). Methods: We calculated the percent error in measuring lower limb circumference with a measuring tape. We performed VECTRA before and after LVA and measured the change in volume. Because of the limitation of photographic range, the measurement of the entire lower limb was difficult using this approach. We were limited to thigh measurements. Results: The ratio of error was calculated using the largest and smallest mean measurements. The largest measurement errors, 4.3%-5.8%, were observed for the measurement of thigh circumferences, whereas the smallest measurement errors, 2.3%-2.9%, were observed in the foot and lower leg areas. The change in volume postoperatively measured using VECTRA decreased by an average of 35.1 cc. Conclusions: The evaluation with a measuring tape for the foot and lower leg region was useful because a lesser content of soft tissue was not influenced to the error by power. Contrarily, VECTRA was useful for measuring the thigh region because of the higher quantity of soft tissue, and the combination of both reduced the error.

Objective: The aim of this study was to find the optimal balance among tube voltage (kV), tube current (mA), and exposure time (s) in cone beam computed tomography (CBCT)., Study Design: Three human hemimandibles were scanned by using the 3D Accuitomo 170 scanner (J. Morita, Kyoto, Japan). First, 3 combinations of kilovolt (kV) and milliampere (mA) were used at a constant radiation dose. Seven observers evaluated the images; the kV of the highest-scoring scan was considered optimal. Second, the lowest acceptable mA for visualizing different anatomic structures was determined. Finally, the samples were scanned by using 3 combinations of tube current and exposure time; the observers determined the combination with the highest image quality., Results: At a constant radiation dose, the highest available voltage (i.e., 90 kV) resulted in the highest image quality in terms of general impression, sharpness, noise, and artefacts. Depending on the anatomic structure, mA reductions of 20% to 40% compared with the default setting were possible. Fast-scan protocols showed equal or slightly better image quality compared with the standard-scan mode., Conclusions: For the CBCT model used in this study, optimization implies the use of the highest kV along with the shortest exposure time and a task-specific mA. The proposed stepwise optimization approach could be applied to any CBCT unit, preferably during commissioning., (Copyright © 2018 Elsevier Inc. All rights reserved.)

For newborns and neonates, ultrasound (US) is the most common imaging modality used for examinations due to its accessibility and ease of use. However, precise volume measurements remain limited in 2D, while MRI in newborns is typically avoided because of immobilization issues which may require sedation. The objective of this study is to assess and validate the lateral ventricular and total brain volumes obtained with an automatic segmentation method using cerebral trans-fontanelle 3D US. Infants aged between 2 and 8.5 months old were recruited, with both MRI and 3D US acquired on the same day was used to validate ventricular and brain volume measurements in comparison to MRI. Lateral ventricles were segmented on both the US (manually and with a proposed automatic fusion-based approach) and MRI, while brain volumes were estimated with an automatic segmentation method. Volumetric 3D US measurements were then evaluated with respect to age distribution. For the comparison between MRI and 3D US, strong inter-class correlations (ICC) were found for the ventricle volumes (manual: 5.9%  ±  2.5% difference (ICC  =  0.99); automatic: 6.0%  ±  2.6% difference (ICC  =  0.98)), as well as the total brain size, with a 3.0%  ±  1.3% difference (ICC  =  0.98). There was no statistically significant difference based on t-test and f-test for the lateral ventricles volume (t-test: p  =  0.542) and (f-test: p  =  0.738) and for the total brain volume (t-test: p  =  0.412) and (f-test: p  =  0.685) between MRI and 3D US. This study demonstrates that 3D US can be used to automatically assess lateral ventricular and total brain volumes with no significant difference to the MRI acquisitions. The highest correlations were obtained for infants under 8 months when the fontanelle is open.