Search

Multi-parametric MRI (mpMRI) studies of the body are routinely acquired in clinical practice. However, a standardized naming convention for MRI protocols and series does not exist currently. Conflicts in the series descriptions present in the DICOM headers arise due to myriad MRI scanners from various manufacturers used for imaging, wide variations in imaging practices across institutions, and technologist preferences. These conflicts affect the hanging protocol, which dictates the arrangement of sequences for the reading radiologist. At present, clinician supervision is necessary to ensure that the correct sequence is being read and used for diagnosis. This pilot work seeks to classify five different series in mpMRI studies acquired at the levels of the chest, abdomen, and pelvis. First, 2D and 3D classification networks were compared using data acquired by Siemens scanners and the optimal network was identified. Then, its performance was analyzed when trained with different training data quantities. The out-of-distribution (OOD) robustness on data acquired by a Philips scanner was also measured. In addition, the effect of data augmentation on model training was studied. The model was also tested with smaller input volumes through downsampling or cropping. Finally, the model was trained on combined data from both Siemens and Philips scanners to bridge the performance gap between different scanners. Among 2D and 3D networks of ResNet-50, ResNet-101, DenseNet- 121, and EfficientNet-BN0, the 3D DenseNet-121 ensemble achieved an F 1 score of 99.5% when tested on data from the Siemens scanners. The model performed well on OOD data from the Philips scanner and achieved an F 1 score of 86.5%. There was no statistically significant difference between the models trained with and without data augmentation, and between the models trained with original-sized input and with smaller-sized input. When training the model with combined data, the F 1 score improved to 98.8% for the Philips test set and 99.3% for the Siemens test set respectively. Our pilot work is useful for the classification of MRI sequences in studies acquired at the level of the chest, abdomen, and pelvis. It has the potential for robust automation of hanging protocols and the creation of large-scale data cohorts for pre-clinical research. [ABSTRACT FROM AUTHOR]

In the United States, cirrhosis was the 12th leading cause of death in 2016. Despite end-stage cirrhosis being irreversible, earlier stages of hepatic fibrosis can be reversed via early diagnosis and intervention. The objective is to investigate the utility of a fully automated technique to measure liver surface nodularity (LSN) for staging hepatic fibrosis (stages F0–F4). In this retrospective study, a dataset consisting of patients with multiple etiologies of liver disease collected at Institution-A (METAVIR F0–F4, 2000–2016) was used. The LSN was automatically measured in contrast-enhanced CT volumes and compared against scores from a manual tool. Area under the receiver operating characteristics curve (AUC) was used to distinguish between clinically significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and end-stage cirrhosis (F4). The study sample had 480 patients (304 men, 176 women, mean age, 49 ± 9). Automatically derived LSN scores progressively increased with the fibrosis stage: F0 (1.64 [mean] ± 1.13 [standard deviation]), F1 (2.16 ± 2.39), F2 (2.17 ± 2.55), F3 (2.23 ± 2.52), and F4 (4.21 ± 2.94). For discriminating significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and cirrhosis (F4), the automated tool achieved ROC AUCs of 73.9%, 82.5%, and 87.8% respectively. The sensitivity and specificity for significant fibrosis (nodularity threshold 1.51) was 85.2% and 73.3%, advanced fibrosis (nodularity threshold 1.73) was 84.2% and 79.5%, and cirrhosis (nodularity threshold 2.18) was 86.5% and 79.5%. Statistical tests revealed that the automated LSN scores distinguished patients with advanced fibrosis (p <.001) and cirrhosis (p <.001). The fully automated LSN measurement retained its predictive power for distinguishing between advanced fibrosis and cirrhosis. The clinical impact is that the fully automated LSN measurement may be useful for early interventions and population-based studies. It can automatically predict the fibrosis stage in ∼45 s in comparison to the ∼2 min needed to manually measure the LSN in a CT volume. [ABSTRACT FROM AUTHOR]

Pancreas segmentation accuracy at CT is critical for the identification of pancreatic pathologies and is essential for the development of imaging biomarkers. Our objective was to benchmark the performance of five high-performing pancreas segmentation models across multiple metrics stratified by scan and patient/pancreatic characteristics that may affect segmentation performance. In this retrospective study, PubMed and ArXiv searches were conducted to identify pancreas segmentation models which were then evaluated on a set of annotated imaging datasets. Results (Dice score, Hausdorff distance [HD], average surface distance [ASD]) were stratified by contrast status and quartiles of peri-pancreatic attenuation (5 mm region around pancreas). Multivariate regression was performed to identify imaging characteristics and biomarkers (n = 9) that were significantly associated with Dice score. Five pancreas segmentation models were identified: Abdomen Atlas [AAUNet, AASwin, trained on 8448 scans], TotalSegmentator [TS, 1204 scans], nnUNetv1 [MSD-nnUNet, 282 scans], and a U-Net based model for predicting diabetes [DM-UNet, 427 scans]. These were evaluated on 352 CT scans (30 females, 25 males, 297 sex unknown; age 58 ± 7 years [ ± 1 SD], 327 age unknown) from 2000–2023. Overall, TS, AAUNet, and AASwin were the best performers, Dice= 80 ± 11%, 79 ± 16%, and 77 ± 18%, respectively (pairwise Sidak test not-significantly different). AASwin and MSD-nnUNet performed worse (for all metrics) on non-contrast scans (vs contrast, P <.001). The worst performer was DM-UNet (Dice=67 ± 16%). All algorithms except TS showed lower Dice scores with increasing peri-pancreatic attenuation (P <.01). Multivariate regression showed non-contrast scans, (P <.001; MSD-nnUNet), smaller pancreatic length (P =.005, MSD-nnUNet), and height (P =.003, DM-UNet) were associated with lower Dice scores. The convolutional neural network-based models trained on a diverse set of scans performed best (TS, AAUnet, and AASwin). TS performed equivalently to AAUnet and AASwin with only 13% of the training set size (8488 vs 1204 scans). Though trained on the same dataset, a transformer network (AASwin) had poorer performance on non-contrast scans whereas its convolutional network counterpart (AAUNet) did not. This study highlights how aggregate assessment metrics of pancreatic segmentation algorithms seen in other literature are not enough to capture differential performance across common patient and scanning characteristics in clinical populations. [ABSTRACT FROM AUTHOR]

Our review shows that AI-based analysis of lymphoma whole-body FDG-PET/CT can inform all phases of clinical management including staging, prognostication, treatment planning, and treatment response evaluation. We highlight advancements in the role of neural networks for performing automated image segmentation to calculate PET-based imaging biomarkers such as the total metabolic tumor volume (TMTV). AI-based image segmentation methods are at levels where they can be semi-automatically implemented with minimal human inputs and nearing the level of a second-opinion radiologist. Advances in automated segmentation methods are particularly apparent in the discrimination of lymphomatous vs non-lymphomatous FDG-avid regions, which carries through to automated staging. Automated TMTV calculators, in addition to automated calculation of measures such as Dmax are informing robust models of progression-free survival which can then feed into improved treatment planning. [ABSTRACT FROM AUTHOR]

Background: Abdominal aortic atherosclerotic plaque burden may have clinical significance but manual measurement is time-consuming and impractical.Purpose: To perform external validation on an automated atherosclerotic plaque detector for noncontrast and postcontrast abdominal CT.Materials and Methods: The training data consisted of 114 noncontrast CT scans and 23 postcontrast CT urography scans. The testing data set consisted of 922 CT colonography (CTC) scans, and 1207 paired noncontrast and postcontrast CT scans from renal donors from a second institution. Reference standard data included manual plaque segmentations in the 137 training scans and manual plaque burden measurements in the 922 CTC scans. The total Agatston score and group (0-3) was determined using fully-automated deep learning software. Performance was assessed by measures of agreement, linear regression, and paired evaluations.Results: On CTC scans, automated Agatston scoring correlated highly with manual assessment (R2 = 0.94). On paired renal donor CT scans, automated Agatston scoring on postcontrast CT correlated highly with noncontrast CT (R2 = 0.95). When plaque burden was expressed as a group score, there was excellent agreement for both the CTC (weighted kappa 0.80 ± 0.01 [95% confidence interval: 0.78-0.83]) and renal donor (0.83 ± 0.02 [0.79-0.86]) assessments.Conclusion: Fully automated detection, segmentation, and scoring of abdominal aortic atherosclerotic plaques on both pre- and post-contrast CT was validated and may have application for population-based studies. [ABSTRACT FROM AUTHOR]

Purpose: Accurate pancreas segmentation has application in surgical planning, assessment of diabetes, and detection and analysis of pancreatic tumors. Factors that affect pancreas segmentation accuracy have not been previously reported. The purpose of this study is to identify technical and clinical factors that adversely affect the accuracy of pancreas segmentation on CT.Method and Materials: In this IRB and HIPAA compliant study, a deep convolutional neural network was used for pancreas segmentation in a publicly available archive of 82 portal-venous phase abdominal CT scans of 53 men and 29 women. The accuracies of the segmentations were evaluated by the Dice similarity coefficient (DSC). The DSC was then correlated with demographic and clinical data (age, gender, height, weight, body mass index), CT technical factors (image pixel size, slice thickness, presence or absence of oral contrast), and CT imaging findings (volume and attenuation of pancreas, visceral abdominal fat, and CT attenuation of the structures within a 5 mm neighborhood of the pancreas).Results: The average DSC was 78% ± 8%. Factors that were statistically significantly correlated with DSC included body mass index (r = 0.34, p < 0.01), visceral abdominal fat (r = 0.51, p < 0.0001), volume of the pancreas (r = 0.41, p = 0.001), standard deviation of CT attenuation within the pancreas (r = 0.30, p = 0.01), and median and average CT attenuation in the immediate neighborhood of the pancreas (r = -0.53, p < 0.0001 and r = -0.52, p < 0.0001). There were no significant correlations between the DSC and the height, gender, or mean CT attenuation of the pancreas.Conclusion: Increased visceral abdominal fat and accumulation of fat within or around the pancreas are major factors associated with more accurate segmentation of the pancreas. Potential applications of our findings include assessment of pancreas segmentation difficulty of a particular scan or dataset and identification of methods that work better for more challenging pancreas segmentations. [ABSTRACT FROM AUTHOR]

Rationale and Objectives: To assess whether a fully-automated deep learning system can accurately detect and analyze truncal musculature at multiple lumbar vertebral levels and muscle groupings on abdominal CT for potential use in the detection of central sarcopenia.Materials and Methods: A computer system for automated segmentation of truncal musculature groups was designed and created. Abdominal CT scans of 102 sequential patients (mean age 68 years, range 59-81 years; 53 women, 49 men) conducted between January 2015 and February 2015 were assembled as a data set. Truncal musculature was manually segmented on axial CT images at multiple lumbar vertebral levels as reference standard data, divided into training and testing subsets, and analyzed by the system. Dice similarity coefficients were calculated to evaluate system performance. IRB approval was obtained, with waiver of informed consent in this retrospective study.Results: System performance as gauged by the Dice coefficients, for detecting the total abdominal muscle cross-section at the level of the third and fourth lumbar vertebrae, were, respectively, 0.953 ± 0.015 and 0.953 ± 0.011 for the training set, and 0.938 ± 0.028 and 0.940 ± 0.026 for the testing set. Dice coefficients for detecting total psoas muscle cross-section at the level of the third and fourth lumbar vertebrae, were, respectively, 0.942 ± 0.040 and 0.951 ± 0.037 for the training set, and 0.939 ± 0.028 and 0.946 ± 0.032 for the testing set.Conclusion: This system fully-automatically and accurately segments multiple muscle groups at all lumbar spine levels on abdominal CT for detection of sarcopenia. [ABSTRACT FROM AUTHOR]

Advances in machine learning in medical imaging are occurring at a rapid pace in research laboratories both at academic institutions and in industry. Important artificial intelligence (AI) tools for diagnostic imaging include algorithms for disease detection and classification, image optimization, radiation reduction, and workflow enhancement. Although advances in foundational research are occurring rapidly, translation to routine clinical practice has been slower. In August 2018, the National Institutes of Health assembled multiple relevant stakeholders at a public meeting to discuss the current state of knowledge, infrastructure gaps, and challenges to wider implementation. The conclusions of that meeting are summarized in two publications that identify and prioritize initiatives to accelerate foundational and translational research in AI for medical imaging. This publication summarizes key priorities for translational research developed at the workshop including: (1) creating structured AI use cases, defining and highlighting clinical challenges potentially solvable by AI; (2) establishing methods to encourage data sharing for training and testing AI algorithms to promote generalizability to widespread clinical practice and mitigate unintended bias; (3) establishing tools for validation and performance monitoring of AI algorithms to facilitate regulatory approval; and (4) developing standards and common data elements for seamless integration of AI tools into existing clinical workflows. An important goal of the resulting road map is to grow an ecosystem, facilitated by professional societies, industry, and government agencies, that will allow robust collaborations between practicing clinicians and AI researchers to advance foundational and translational research relevant to medical imaging. [ABSTRACT FROM AUTHOR]

Rationale and Objectives: This study aims to compare the speed and accuracy of three different software packages in segmenting the liver and the spleen.Materials and Methods: The three software packages are Advantage Workstation Solutions (AWS), Claron Technology (Claron) Liver Segmentor, and Vitrea Core Fx (Vitrea). The dataset consisted of abdominal computed tomography scans of 30 patients obtained from the portal venous phase. All but two of the patients had a cancer diagnosis. The livers of 14 patients and the spleens of 24 patients were reported as normal; the remaining livers and spleens contained one or more abnormalities. The initial segmentation times and volumes were recorded in Claron and Vitrea as these created automatic segmentations. The total segmentation times and volumes following corrections were recorded. The livers and spleens were segmented separately by two radiologists who used all three packages. Accuracy was assessed by comparing volumes measured using fully manual segmentation on the AWS.Results: Claron could not segment the spleen in four subjects for the first reader and in two subjects for the second reader. The final mean segmentation times for the liver for both readers were 6.5 and 5.5 minutes for AWS, 4.4 and 3.6 minutes for Claron, and 5.1 and 4.2 minutes for Vitrea. The final mean segmentation times for the spleen were 2.7 and 2.1 minutes for AWS, 2.1 and 1.4 minutes for Claron, and 1.8 and 1.2 minutes for Vitrea. No statistically significant difference was found between the organ volumes measured by the two readers when using Vitrea. The mean differences between the initial and final segmentation volumes ranged from -1.2% to 0.4% for the liver and from -4.0% to 9.8% for the spleen. The mean differences between the automated liver segmentation volumes and the AWS volumes were 2.5%-2.9% for Claron and 4.9%-6.6% for Vitrea. The mean differences between the automated splenic segmentation volumes and the AWS volumes were 5.0%-6.2% for Claron and 10.6%-12.0% for Vitrea.Conclusions: Both automated packages (Claron and Vitrea) measured liver and spleen volumes that were accurate and quick before manual correction. Volumes for the liver were more accurate than those for the spleen, perhaps due to the much smaller splenic volumes compared to those of the liver. For both liver and spleen, manual corrections were time consuming and for most subjects did not significantly change the volume measurement. [ABSTRACT FROM AUTHOR]

Rationale and Objectives: To define systematic volumetric thresholds to identify and grade splenomegaly and retrospectively evaluate the performance of radiologists to assess splenomegaly in computed tomography (CT) image data. Materials and Methods: A clinical tool was developed to segment spleens from 172 contrast-enhanced clinical CT studies. There were 45 normal and 127 splenomegaly cases confirmed by radiological reports. Spleen volumes were compared to manual measurements using overlap/error. Volumetric thresholds for mild/massive splenomegaly were defined at 1/2.5 standard deviations above the average splenic volume of the healthy population. The thresholds were validated against consensus reports. The performance of radiologists in assessing splenomegaly was retrospectively evaluated. Results: The automated segmentation of spleens was robust with volume overlap/error of 95.2/3.3%. There were no significant differences (P > .2) between manual and automated segmentations for either normal/splenomegaly subgroups. Comparable correlations between interobserver and manual-automated measurements were found (r = 0.99 for all). The average volume of normal spleens was 236.89 ± 77.58 mL. For splenomegaly, average volume was 1004.75 ± 644.27 mL. Volumetric thresholds of 314.47/430.84 mL were used to define mild/massive splenomegaly (±18.86 mL, 95% CI). Radiologists disagreed in 23.25% (n = 40) of the diagnosed cases. The area under the receiver operating characteristic curve of the volumetric criterion for splenomegaly detection was 0.96. Using the volumetric thresholds as the reference standard, the sensitivity of radiologists in detecting all/mild/massive splenomegaly was 95.0/66.6/99.0% at 78.0% specificity, respectively. Conclusion: Thresholds for the identification and grading of splenomegaly from automatic volumetric spleen assessment were introduced. The volumetric thresholds match well with clinical interpretations for splenomegaly and may improve splenomegaly detection compared with splenic cephalocaudal height measurements or visual inspection commonly used in current clinical practice. [Copyright &y& Elsevier]

Abstract: Background: Myocardial fibrosis leads to impaired cardiac function and events. Extracellular volume fraction (ECV) assessed with an iodinated contrast agent and measured by cardiac CT may be a useful noninvasive marker of fibrosis. Objective: The purpose of this study was to develop and evaluate a 3-dimensional (3D) ECV calculation toolkit (ECVTK) for ECV determination by cardiac CT. Methods: Twenty-four subjects (10 systolic heart failure, age, 60 ± 17 years; 5 diastolic failure, age 56 ± 20 years; 9 matched healthy subjects, age 59 ± 7 years) were evaluated. Cardiac CT examinations were done on a 320-multidetector CT scanner before and after 130 mL of iopamidol (Isovue-370; Bracco Diagnostics, Plainsboro, NJ, USA) was administered. A calcium score type sequence was performed before and 7 minutes after contrast with single gantry rotation during 1 breath hold and single cardiac phase acquisition. ECV was calculated as (ΔHUmyocardium/ΔHUblood) × (1 − Hct) where Hct is the hematocrit, and ΔHU is the change in Hounsfield unit attenuation = HUafter iodine − HUbefore iodine. Cardiac magnetic resonance imaging was performed to assess myocardial structure and function. Results: Mean 3D ECV values were significantly higher in the subjects with systolic heart failure than in healthy subjects and subjects with diastolic heart failure (mean, 41% ± 6%, 33% ± 2%, and 35% ± 5%, respectively; P = 0.02). Interobserver and intraobserver agreements were excellent for myocardial, blood pool, and ECV (intraclass correlation coefficient, >0.90 for all). Higher 3D ECV by cardiac CT was associated with reduced systolic circumferential strain, greater end-diastolic and -systolic volumes, and lower ejection fraction (r = 0.70, r = 0.60, r = 0.73, and r = −0.68, respectively; all P < 0.001). Conclusion: 3D ECV by cardiac CT can be performed with ECVTK. We demonstrated increased ECV in subjects with systolic heart failure compared with healthy subjects. Cardiac CT results also showed good correlation with important functional heart biomarkers, suggesting the potential for myocardial tissue characterization with the use of 3D ECV by cardiac CT. This trial is registered at www.ClinicalTrials.gov as NCT01160471. [Copyright &y& Elsevier]

Rationale and Objectives: The aims of this study were to define volumetric nomograms for identifying hepatomegaly and to retrospectively evaluate the performance of radiologists in assessing hepatomegaly. Materials and Methods: Livers were automatically segmented from 148 abdominal contrast-enhanced computed tomographic scans: 77 normal livers and 71 cases of hepatomegaly (diagnosed by visual inspection and/or linear liver height by radiologists). Quantified liver volumes were compared to manual measurements using volume overlap and error. Liver volumes were normalized to body surface area, from which hepatomegaly nomograms were defined (H scores) by analyzing the distribution of liver sizes in the healthy population. H scores were validated against consensus reports. The performance of radiologists in diagnosing hepatomegaly was retrospectively evaluated. Results: The automated segmentation of livers was robust, with volume overlap and error of 96.2% and 2.2%, respectively. There were no significant differences (P > .10) between manual and automated segmentation for either the normal or the hepatomegaly subgroup. The average volumes of normal and enlarged livers were 1.51 ± 0.25 and 2.32 ± 0.75 L, respectively. One-way analysis of variance found that body surface area (P = .004) and gender (P = .02), but not age, significantly affected normal liver volume. No significant effects were observed for two-way and three-way interactions among the three variables (P > .18). H-score cutoffs of 0.92 and 1.08 L/m2 were used to define mild and massive hepatomegaly (95% confidence interval, ±0.02 L/m2). Using the H score as the reference standard, the sensitivity of radiologists in detecting all, mild, and massive hepatomegaly was 84.4%, 56.7%, and 100.0% at 90.1% specificity, respectively. Radiologists disagreed on 20.9% of the diagnosed cases (n = 31). The area under the receiver-operating characteristic curve of the H-score criterion for hepatomegaly detection was 0.98. Conclusions: Nomograms for the identification and grading of hepatomegaly from automatic volumetric liver assessment normalized to body surface area (H scores) are introduced. H scores match well with clinical interpretations for hepatomegaly and may improve hepatomegaly detection compared with height measurements or visual inspection, commonly used in current clinical practice. [Copyright &y& Elsevier]

Rationale and Objectives: The aims of this study were to develop and validate an automated method to segment the renal cortex on contrast-enhanced abdominal computed tomographic images from kidney donors and to track cortex volume change after donation. Materials and Methods: A three-dimensional fully automated renal cortex segmentation method was developed and validated on 37 arterial phase computed tomographic data sets (27 patients, 10 of whom underwent two computed tomographic scans before and after nephrectomy) using leave-one-out strategy. Two expert interpreters manually segmented the cortex slice by slice, and linear regression analysis and Bland-Altman plots were used to compare automated and manual segmentation. The true-positive and false-positive volume fractions were also calculated to evaluate the accuracy of the proposed method. Cortex volume changes in 10 subjects were also calculated. Results: The linear regression analysis results showed that the automated and manual segmentation methods had strong correlations, with Pearson''s correlations of 0.9529, 0.9309, 0.9283, and 0.9124 between intraobserver variation, interobserver variation, automated and user 1, and automated and user 2, respectively (P < .001 for all analyses). The Bland-Altman plots for cortex segmentation also showed that the automated and manual methods had agreeable segmentation. The mean volume increase of the cortex for the 10 subjects was 35.1 ± 13.2% (P < .01 by paired t test). The overall true-positive and false-positive volume fractions for cortex segmentation were 90.15 ± 3.11% and 0.85 ± 0.05%. With the proposed automated method, the time for cortex segmentation was reduced from 20 minutes for manual segmentation to 2 minutes. Conclusions: The proposed method was accurate and efficient and can replace the current subjective and time-consuming manual procedure. The computer measurement confirms the volume of renal cortex increases after kidney donation. [Copyright &y& Elsevier]

Rationale and Objectives: The purpose of this study was to develop and test a computer-assisted detection method for the identification and measurement of pulmonary abnormalities on chest computed tomographic (CT) imaging in cases of infection, such as novel H1N1 influenza. The method developed could be a potentially useful tool for classifying and quantifying pulmonary infectious disease on CT imaging. Materials and Methods: Forty chest CT examinations were studied using texture analysis and support vector machine classification to differentiate normal from abnormal lung regions on CT imaging, including 10 patients with immunohistochemistry-proven infection, 10 normal controls, and 20 patients with fibrosis. Results: Statistically significant differences in the receiver-operating characteristic curves for detecting abnormal regions in H1N1 infection were obtained between normal lung and regions of fibrosis, with significant differences in texture features of different infections. These differences enabled the quantification of abnormal lung volumes on CT imaging. Conclusion: Texture analysis and support vector machine classification can distinguish between areas of abnormality in acute infection and areas of chronic fibrosis, differentiate lesions having consolidative and ground-glass appearances, and quantify those texture features to increase the precision of CT scoring as a potential tool for measuring disease progression and severity. [ABSTRACT FROM AUTHOR]

Rationale and Objectives: To determine whether the display of computer-aided detection (CAD) marks on individual polyps on both the supine and prone scans leads to improved polyp detection by radiologists compared to the display of CAD marks on individual polyps on either the supine or the prone scan, but not both. Materials and Methods: The acquisition of patient data for this study was approved by the Institutional Review Board and was Health Insurance Portability and Accountability Act–compliant. Subsequently, the use of the data was declared exempt from further institutional review board review. Four radiologists interpreted 33 computed tomography colonography cases, 21 of which had one adenoma 6–9 mm in size, with the assistance of a CAD system in the first reader mode (ie, the radiologists reviewed only the CAD marks). The radiologists were shown each case twice, with different sets of CAD marks for each of the two readings. In one reading, a true-positive CAD mark for the same polyp was displayed on both the supine and prone scans (a double-mark reading). In the other reading, a true-positive CAD mark was displayed either on the supine or prone scan, but not both (a single-mark reading). True-positive marks were randomized between readings and there was at least a 1-month delay between readings to minimize recall bias. Sensitivity and specificity were determined and receiver operating characteristic (ROC) and multiple-reader multiple-case analyses were performed. Results: The average per polyp sensitivities were 60% (38%–81%) versus 71% (52%–91%) (P = .03) for single-mark and double-mark readings, respectively. The areas (95% confidence intervals) under the ROC curves were 0.76 (0.62–0.88) and 0.79 (0.58–0.96), respectively (P = NS). Specificities were similar for the single-mark compared with the double-mark readings. Conclusion: The display of CAD marks on a polyp on both the supine and prone scans led to more frequent detection of polyps by radiologists without adversely affecting specificity for detecting 6–9 mm adenomas. [Copyright &y& Elsevier]

Rationale and Objectives: The factors that influence the conspicuity of polyps on computed tomographic (CT) colonography (CTC) are poorly understood. The aim of this study is to compare radiologists'' visual assessment of polyp conspicuity to quantitative image features and show the relationship between visual conspicuity and the detection of colonic polyps by computer-aided detection (CAD) on CTC. Methods: One polyp (size range 6—10 mm) was selected from the CTC examination of each of 29 patients from a larger cohort. All patients underwent oral contrast-enhanced CTC with same-day optical colonoscopy with segmental unblinding. The polyps were analyzed by a previously validated CAD system and placed into one of two groups (detected [n = 12] or not detected [n = 17] by CAD). The study population was intentionally enriched with polyps that were not detected by the CAD system. Four board-certified radiologists, blinded to the CAD results, reviewed two- and three-dimensional CTC images of the polyps and scored the conspicuity of the polyps using a 4-point scale (0 = least conspicuous, 3 = most conspicuous). Polyp height and width were measured by a trained observer. A t-test (two-tailed, unpaired equal variance) was done to determine statistical significance. Intra- and interobserver variabilities of the conspicuity scores were assessed using the weighted κ test. Regression analysis was used to investigate the relationship of conspicuity to polyp height and width. Results: A statistically significant difference was found between the average conspicuity scores for polyps that were detected by CAD compared to those that were not (2.3 ± 0.6 vs. 1.4 ± 0.8) (P = .004). There was moderate intraobserver agreement of the conspicuity scores (weighted κ 0.57 ± 0.09). Interobserver agreement was fair (average weighted κ for six pair-wise comparisons, 0.38 ± 0.15). Conspicuity was correlated with manual measurement of polyp height (r 2 = 0.38–0.56, P < .001). Conclusions: This CAD system tends to detect 6—10 mm polyps that are more visually conspicuous. Polyp height is a major determinant of visual conspicuity. The generalizability of these findings to other CAD systems is currently unknown. Nevertheless, CAD developers may need to specifically target flatter and less conspicuous polyps for CAD to better assist the radiologist to find polyps in this clinically important size category. [Copyright &y& Elsevier]

Oral contrast agents used during CT colonography (CTC) are valuable and may reduce false positive and false negative detections due to stool and residual fluid. Electronic cleansing algorithms are feasible, and oral contrast agents can eliminate the CTC requirement for a clean colon. Recent work shows oral contrast frequently adheres to polyps, with a preference for those with villous histology, a characteristic of advanced polyps. This finding encourages the development of contrast agents that highlight polyps at greatest risk for progression to malignancy. Our review summarizes numerous aspects of oral barium sulfate contrast agents as well as tests to assess adherence and coating ability of the agents, offering arenas to explore and tools for evaluation. [Copyright &y& Elsevier]

Rationale and Objectives: We sought to demonstrate that intravenous contrast-enhanced CT colonography (CTC) can distinguish colonic adenomas from carcinomas. Methods: Supine intravenous contrast-enhanced CTC with colonoscopic and/or surgical correlation was performed on 25 patients with colonic adenomas or carcinomas. Standard deviation of mean polyp CT attenuation was computed and assessed using ANOVA and receiver-operating characteristic analyses. Results: Colonoscopy confirmed 32 polyps or masses 1 to 8 cm in size. The standard deviations of CT attenuation were carcinomas (n = 13; 36 +/- 6 HU; range 28-48 HU) and adenomas (n = 19; 49 +/- 14 HU; range 31-100 HU) (P = 0.005). At a standard deviation threshold of 42 HU, the sensitivity and specificity for classifying a polyp or mass as a carcinoma were 92% and 79%, respectively. The area under the receiver-operating characteristic curve was 0.89 +/- 0.06 (95% confidence interval 0.73-0.96). Conclusions: Measurement of the standard deviation of CT attenuation on intravenous contrast-enhanced CTC permits histopathologic classification of polyps 1 cm or larger as carcinomas versus adenomas. The presence of ulceration or absence of muscular invasion in carcinomas creates overlap with adenomas, reducing the specificity of carcinoma classification. [ABSTRACT FROM AUTHOR]

Background & Aims: The sensitivity of computed tomographic (CT) virtual colonoscopy (CT colonography) for detecting polyps varies widely in recently reported large clinical trials. Our objective was to determine whether a computer program is as sensitive as optical colonoscopy for the detection of adenomatous colonic polyps on CT virtual colonoscopy. Methods: The data set was a cohort of 1186 screening patients at 3 medical centers. All patients underwent same-day virtual and optical colonoscopy. Our enhanced gold standard combined segmental unblinded optical colonoscopy and retrospective identification of precise polyp locations. The data were randomized into separate training (n = 394) and test (n = 792) sets for analysis by a computer-aided polyp detection (CAD) program. Results: For the test set, per-polyp and per-patient sensitivities for CAD were both 89.3% (25/28; 95% confidence interval, 71.8%–97.7%) for detecting retrospectively identifiable adenomatous polyps at least 1 cm in size. The false-positive rate was 2.1 (95% confidence interval, 2.0–2.2) false polyps per patient. Both carcinomas were detected by CAD at a false-positive rate of 0.7 per patient; only 1 of 2 was detected by optical colonoscopy before segmental unblinding. At both 8-mm and 10-mm adenoma size thresholds, the per-patient sensitivities of CAD were not significantly different from those of optical colonoscopy before segmental unblinding. Conclusions: The per-patient sensitivity of CT virtual colonoscopy CAD in an asymptomatic screening population is comparable to that of optical colonoscopy for adenomas ≥8 mm and is generalizable to new CT virtual colonoscopy data. [Copyright &y& Elsevier]

Rationale and Objectives: A new classification scheme for the computer-aided detection of colonic polyps in computed tomographic colonography is proposed. Materials and Methods: The scheme involves an ensemble of support vector machines (SVMs) for classification, a smoothed leave-one-out (SLOO) cross-validation method for obtaining error estimates, and use of a bootstrap aggregation method for training and model selection. Our use of an ensemble of SVM classifiers with bagging (bootstrap aggregation), built on different feature subsets, is intended to improve classification performance compared with single SVMs and reduce the number of false-positive detections. The bootstrap-based model-selection technique is used for tuning SVM parameters. In our first experiment, two independent data sets were used: the first, for feature and model selection, and the second, for testing to evaluate the generalizability of our model. In the second experiment, the test set that contained higher resolution data was used for training and testing (using the SLOO method) to compare SVM committee and single SVM performance. Results: The overall sensitivity on independent test set was 75%, with 1.5 false-positive detections/study, compared with 76%-78% sensitivity and 4.5 false-positive detections/study estimated using the SLOO method on the training set. The sensitivity of the SVM ensemble retrained on the former test set estimated using the SLOO method was 81%, which is 7%-10% greater than the sensitivity of a single SVM. The number of false-positive detections per study was 2.6, a 1.5 times reduction compared with a single SVM. Conclusion: Training an SVM ensemble on one data set and testing it on the independent data has shown that the SVM committee classification method has good generalizability and achieves high sensitivity and a low false-positive rate. The model selection and improved error estimation method are effective for computer-aided polyp detection. [ABSTRACT FROM AUTHOR]

Rationale and Objective: To present an algorithm to automatically locate seeds for colon segmentation in computed tomography colonography (CTC). Materials and Methods: The algorithm automatically locates two points (seeds) inside the colon lumen. Because of their high distention and fixed anatomic position, we focus on the cecum and rectum for automatic seed placement. We use two-dimensional morphological operators that find pockets of colonic air of sufficient size. For the rectum, we search within an inferiorly and centrally located CT slice. For the cecum, we search in a group of CT slices in the middle of the scanned volume on the patient's right side. We applied our automated algorithm to segment the colon in 292 consecutive cases of CTC (146 prone, 146 supine). Results: After automated seed placement, 83.2% (243 of 292) of the colons were segmented completely and 9.6% (28 of 292) were segmented partially. The unsegmented colon parts were present in datasets where the colon was collapsed in more than one place or because seeds could not be placed in regions filled with fluid. In the remaining 7.2% (21 of 292) of cases, the automatic segmentation leaked outside the colon because of a limitation of the contrast-enhanced fluid detection algorithm. Conclusion: Fully automatic seed placement for colonic segmentation is feasible in the majority of cases without seeding of undesired extracolonic air. [ABSTRACT FROM AUTHOR]

We examine a method of curvature estimation that computes the curvature directly from three dimensional data. We refer to this as the kernel method of curvature estimation and our experiments indicate that several parameters be modified from those originally suggested to achieve more accurate and reliable results. This improved performance is essential for analysis of medical volume data by computer aided diagnosis algorithms, many of which use curvature estimation for shape computations and pattern recognition. We also examine cases in which the kernel method yields inaccurate responses based on specific topologies. [Copyright &y& Elsevier]

Rationale and Objectives: Subcutaneous melanomas may be missed on computed tomography because of their peripheral location or perceived unimportance, yet they can have clinical significance. The use of a novel computer-assisted detection scheme to locate subcutaneous melanoma lesions in body CT images was investigated. Materials and Methods: The detection software segments subcutaneous fat from the rest of the body and searches for soft tissue density lesions that match a size and shape constraint. Sensitivity and specificity of the proposed method was analyzed by comparing automated lesion detection results in eight patients with 118 subcutaneous melanomas with ground truth data derived from manual tracings of a trained observer. Results: The sensitivity of subcutaneous melanoma detection was 86%. The false-positive rate was 3.1 per slice. Analysis of the false-positives showed that the most common cause was incorrect classification of muscle as a nodule. Conclusion: This study showed the feasibility of a fully automatic subcutaneous melanoma lesion detection system having good sensitivity. The false-positive rate was high, but avenues for further reduction were identified. [ABSTRACT FROM AUTHOR]

Rationale and Objectives. A novel method to compute the centerline of the human colon obtained from computed tomography colonography is proposed. Two applications of this method are demonstrated: to compute local colonic distension (caliber), and to match polyps on supine and prone images.Materials and Methods. The centerline algorithm involves multiple steps including simplification of the colonic surface by decimation; thinning of the decimated colon to create a preliminary centerline; selection of equally spaced points on the preliminary centerline; grouping neighboring points; and mapping them back to rings in the original colon. This method was tested on 20 human computed tomography colonography datasets (supine and prone examinations of 10 patients) and on a computer-generated colon phantom.Results. Visual inspection of the colons and their centerlines showed the centerline to be accurate. For the colon phantom, the average error was only 1 mm. For 11 polyps visualized in both the supine and prone positions and found by computer-aided detection, the normalized distance along the centerline to each polyp was not significantly different on the supine and prone views (r = 0.999; P < .001).Conclusion. This method produces an accurate colon centerline that may be useful for flight path planning, matching detections on the supine and prone views, and computing local colonic distension. [Copyright &y& Elsevier]

: Rationale and ObjectivesA new classification system for colonic polyp detection, designed to increase sensitivity and reduce the number of false-positive findings with computed tomographic colonography, was developed and tested in this study.: Materials and MethodsThe system involves classification by a committee of neural networks (NNs), each using largely distinct subsets of features selected from a general set. Back-propagation NNs trained with the Levenberg-Marquardt algorithm were used as primary classifiers (committee members). The set of features included region density, Gaussian and mean curvature and sphericity, lesion size, colon wall thickness, and the means and standard deviations of all of these values. Subsets of variables were initially selected because of their effectiveness according to training and test sample misclassification rates. The final decision for each case is based on the majority vote across the networks and reflects the weighted votes of all networks. The authors also introduce a smoothed cross-validation method designed to improve estimation of the true misclassification rates by reducing bias and variance.: ResultsThis committee method reduced the false-positive rate by 36%, a clinically meaningful reduction, and improved sensitivity by an average of 6.9% compared with decisions made by any single NN. The overall sensitivity and specificity were 82.9% and 95.3%, respectively, when sensitivity was estimated by means of smoothed cross-validation.: ConclusionThe proposed method of using multiple classifiers and majority voting is recommended for classification tasks with large sets of input features, particularly when selected feature subsets may not be equally effective and do not provide satisfactory true- and false-positive rates. This approach reduces variance in estimates of misclassification rates. [Copyright &y& Elsevier]