Search

Large-scale Artificial General Intelligence (AGI) models, including Large Language Models (LLMs) such as ChatGPT/GPT-4, have achieved unprecedented success in a variety of general domain tasks. Yet, when applied directly to specialized domains like medical imaging, which require in-depth expertise, these models face notable challenges arising from the medical field's inherent complexities and unique characteristics. In this review, we delve into the potential applications of AGI models in medical imaging and healthcare, with a primary focus on LLMs, Large Vision Models, and Large Multimodal Models. We provide a thorough overview of the key features and enabling techniques of LLMs and AGI, and further examine the roadmaps guiding the evolution and implementation of AGI models in the medical sector, summarizing their present applications, potentialities, and associated challenges. In addition, we highlight potential future research directions, offering a holistic view on upcoming ventures. This comprehensive review aims to offer insights into the future implications of AGI in medical imaging, healthcare, and beyond. [ABSTRACT FROM AUTHOR]

Increasing demands on medical imaging departments are taking a toll on the radiologist's ability to deliver timely and accurate reports. Recent technological advances in artificial intelligence have demonstrated great potential for automatic radiology report generation (ARRG), sparking an explosion of research. This survey paper conducts a methodological review of contemporary ARRG approaches by way of (i) assessing datasets based on characteristics, such as availability, size, and adoption rate, (ii) examining deep learning training methods, such as contrastive learning and reinforcement learning, (iii) exploring state-of-the-art model architectures, including variations of CNN and transformer models, (iv) outlining techniques integrating clinical knowledge through multimodal inputs and knowledge graphs, and (v) scrutinising current model evaluation techniques, including commonly applied NLP metrics and qualitative clinical reviews. Furthermore, the quantitative results of the reviewed models are analysed, where the top performing models are examined to seek further insights. Finally, potential new directions are highlighted, with the adoption of additional datasets from other radiological modalities and improved evaluation methods predicted as important areas of future development. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is becoming all-pervasive, impacting our lives in ways that are sometimes visible, but at other times hidden. Algorithms that learn from data—be it consumer preferences, medical imaging data, or any other dataset—are influencing decisions and choices in every industry, and health care is no exception. When patient outcomes are at stake this raises questions of ethics, accountability, and accuracy. This article looks at how AI is currently being put to work in cardiology, how its range of applications might expand, and how health care professionals and technologists can work together to ensure that it is deployed in a way that is safe, efficient, and beneficial for all patients. [ABSTRACT FROM AUTHOR]

We designed a generative adversarial network and an attention network to solve the brainwave emotion-classification problem. Using spatial attention and channel attention superposition to normalize and enhance the raw EEG data, we effectively solved the defects in the EEG data with weak features and easily disturbed them. First, a cognitive map of the brain in the emotional state was constructed by extracting graphical features from EEG signals. Simultaneously, generative adversarial networks are used to add noise to the cognitive map to generate similar data. The volume of the brain cognitive map has been expanded. The problem of insufficient EEG signal data was solved, and the accuracy and robustness were improved. Finally, we compared the processing abilities of different neural networks using EEG and adversarial signals. Compared with other deep learning models and parameter optimization methods, the proposed model achieved a detection accuracy of 94.87% on the SEED dataset.

The instance segmentation of overlapping cells in smear images of epithelial cells is challenging due to the significant overlap and adhesion between the cells’ translucent cytoplasm. In this paper, an improved single-stage instance segmentation network called VoVNet-BiFPN-SOLO (VB-SOLO) is proposed to address this problem. The model takes SOLOv2 model as its main frame. Firstly, the backbone network uses Efficient Channel Attention (ECA) to optimize the VoVNetv2 network to increase the information interaction across channels and enhance the extraction of cell instance features. Secondly, the bi-directional feature pyramid network (BiFPN) is introduced to connect with the new backbone. BiFPN can achieve the weighted fusion of features with different resolutions from bottom to top and keep more shallow semantic information in the network. Finally, the Convolutional Block Attention Module (CBAM) is added to the mask branch to improve cell segmentation results in feature maps. Experimental results on the publicly available datasets CISD and Cx22 demonstrate the effectiveness of the VB-SOLO model, achieving a DCP of 0.966 and 0.940 and a FNRO of 0.055 and 0.03. Compared to the original SOLOv2 algorithm, the proposed method achieved improvements in DCP of 1.3% and 1.1% respectively. Additionally, comparative tests with multiple instance segmentation networks have shown that the proposed improved network can achieve a better balance between segmentation accuracy and efficiency. The experimental results demonstrate the effectiveness of the proposed network improvements and the potential of single-stage instance segmentation networks in overlapping cell image segmentation.

Objective: We propose a modular stretchable coil design using conductive threads and commercially available embroidery machines. The coil design increases customizability of coil arrays for individual patients and each body part. Methods: Eight rectangular coils were constructed with custom-fabricated stretchable tinsel copper threads incorporated onto textile. Tune, match, and detune circuits were incorporated on the coil. A hook-and-loop mechanism was used to attach and decouple the modular coils. Phantom and in vivo scans at various anatomical flexion angles were acquired to highlight performance, and a temperature test was performed to verify safety. Results: In vivo MRI experiments demonstrate high sensitivity and coverage of each anatomy. As the coils are stretched, the sensitive volume increases at a rate of 10.93 mL/cm2. The SNR reduction of a single coil was greater during compression than when stretched, but this did not affect image quality for the array. The modularity of the array allows for adaptability for any anatomy with simple on-demand adjustment to the number and position of coil elements. Conclusion: The images demonstrated high sensitivity and coverage of the stretchable array for various anatomies and flexion angles. Stretching the coils increases the sensitive volume, allowing for a larger region to be effectively imaged. The resonance shift and SNR decrease during stretch and compression support further investigation of methods to reduce frequency shift in stretchable coils. Significance: The proposed array design allows for highly stretchable, flexible, modular, and conformal patient-centered coils that allow for increased imaging quality, greater comfort, and rapid production.

Stress is a mental health problem that is hazardous if not recognized early. A promising approach for noninvasive stress detection involves leveraging camera technology; however, there are notable challenges involved in this method, particularly regarding signal accuracy and quality, which are primarily caused by motion artifacts. This study aims to improve stress classification accuracy by using web camera-based imaging photoplethysmography signals. We introduce a frame alignment method that can significantly correct noise to mitigate motion artifacts. We use heart rate (HR) and HR variability metrics to monitor stress and classify the measurements into stress and no-stress conditions. The sample in this study comprises students who were stressed using the arithmetic task method. Our findings showed that the mean HR and low-frequency component increased under stressful conditions while the high-frequency component decreased. The primary contribution of this study involves refining the accuracy and reducing the time required for stress classification time. Notably, the proposed approach markedly improves the accuracy, with substantial noise correction in the resultant signal. We evaluate various classification methods, including logistic regression, support vector machine, naïve Bayes, and random forest. Our results demonstrate that logistic regression achieved the highest accuracy of 98.5%, with a receiver operating curve value of 98%, an F1 score of 98.4%, and a computation time of 4.7 s. Overall, the proposed methodology holds considerable promise for camera-based noninvasive human stress assessment.

In Electrical Impedance Tomography, the ill-posed nature of the inverse problem and the interaction between the electric current and objects cause poor spatial resolution and blurring in the reconstructed images. This study proposes a numerical method, called the voltage-dependent point-spread function, which depends on the initial setting of the environment, such as voltage, conductivity, permeability, and frequency, to overcome the above problems. When the object layout is convoluted with a point spread function, the Electrical Impedance Tomography image can be reconstructed regardless of the conditions in forward and inverse problems, which is suitable for applications requiring big data, such as deep learning or machine learning. The object’s size in the Electrical Impedance Tomography image can be reconstructed to approximately the real size by deconvolution with a point-spread function. The proposed method was validated using conventional methods through simulations and experimental experiments. The results demonstrated the feasibility and applicability of this method in clinical practice.

This study proposes a wave-weighted modulation method to improve the distinction between areas with small conductivity differences in Electrical Impedance Tomography. The modulated signal is generated by modulating the probe and carrier signals, which are adjusted through modulation indexes, including the amplitude, frequency, and phase. The proposed method was compared with commonly applied methods in Electrical Impedance Tomography and validated through simulations and experimental results. The simulation results showed significant improvements over commonly applied methods, including an increase in the current flow density and a tendency for the current to pass through the different conductivity layers, thereby enhancing the ability to distinguish between layers with small conductivity differences within the object. This study demonstrates the feasibility of the wave-weighted modulation method for enhancing the visualization of biological tissue regions with small conductivity differences in heterogeneous environments, such as hemorrhage, swelling, cancerous and normal tissue layers, or changes in physiological function.

Recognizing the increasing relevance of entrepreneurship skillsets to engineers, CASS launched an initiative to educate on entrepreneurship with technology solutions aligned with CAS Society’s visions. In this special issue, this article introduces the motivation, course setup, and the results of an intensive six-week hybrid course stimulating entrepreneurial thinking. The course was based on a novel iterative and agile innovation framework designed to lead up to the BIOCASS 2023 conference in Toronto, inviting the best proposals for an in-person presentation. The learners confirmed a changed mindset towards innovation generation. [ABSTRACT FROM AUTHOR]

Deep brain stimulation (DBS) is an effective treatment for neurological disorders, and accurately reconstructing the DBS lead trajectories is crucial for MRI compatibility assessment and surgical planning. This paper presents a novel fully automated framework for reconstructing DBS lead trajectories from postoperative CT images. The leads were first segmented by thresholding, but would be fused together somewhere. Mean curvature analysis of multi-layer CT number isosurfaces was introduced to effectively address lead fusion, due to the different topological characteristics of the isosurfaces in and out of the fusion regions. The position of electrode contacts was determined through morphological analysis to get the starting point and the initial direction for trajectory tracking. The next trajectory point was derived by calculating the weighted average coordinates of the candidate points, using the distance from the current estimated trajectory and the CT number as weights. This method has demonstrated high accuracy and efficiency, successfully and automatically reconstructing complex bilateral trajectories for 13 patient cases in less than 10 minutes with errors less than 1 mm. This work overcomes the limitations of existing semi-automatic techniques that require extensive manual intervention. It paves the way for optimizing DBS lead trajectory to reduce tissue heating and image artifacts, which will contribute to neuroimaging studies and improve clinical outcomes. Code for our proposed algorithm is publicly available on Github. [ABSTRACT FROM AUTHOR]

Motor imagery (MI) is widely employed in stroke rehabilitation due to the event-related desynchronization (ERD) phenomenon in sensorimotor cortex induced by MI is similar to actual movement. However, the traditional BCI paradigm, in which the patient imagines the movement of affected hand (AH-MI) with a weak ERD caused by the damaged brain regions, retards motor relearning process. In this work, we applied a novel MI paradigm based on the “sixth-finger” (SF-MI) in stroke patients and systematically uncovered the ERD pattern enhancement of novel MI paradigm compared to traditional MI paradigm. Twenty stroke patients were recruited for this experiment. Event-related spectral perturbation was adopted to supply details about ERD. Brain activation region, intensity and functional connectivity were compared between SF-MI and AH-MI to reveal the ERD enhancement performance of novel MI paradigm. A “wider range, stronger intensity, greater connection” ERD activation pattern was induced in stroke patients by novel SF-MI paradigm compared to traditional AH-MI paradigm. The bilateral sensorimotor and prefrontal modulation was found in SF-MI, which was different in AH-MI only weak sensorimotor modulation was exhibited. The ERD enhancement is mainly concentrated in mu rhythm. More synchronized and intimate neural activity between different brain regions was found during SF-MI tasks compared to AH-MI tasks. Classification results (>80% in SF-MI vs. REST) also indicated the feasibility of applying novel MI paradigm to clinical stroke rehabilitation. This work provides a novel MI paradigm and demonstrates its neural activation-enhancing performance, helping to develop more effective MI-based BCI system for stroke rehabilitation. [ABSTRACT FROM AUTHOR]

Clinical studies have proved significant improvements in hand motor function in stroke patients when assisted by robotic devices. However, there were few studies on neural activity changes in the brain during execution. This study aimed to investigate the brain activation pattern caused by soft rehabilitation glove and virtual reality scenes. Twenty healthy subjects and twenty stroke patients were recruited to complete three controlled trials: grasping passively with robotic glove assistance (RA), watching grasping movement video in virtual reality (VR), and the joint use of robotic glove and virtual reality (VRA). Neural activity in the prefrontal cortex, motor cortex and occipital lobe was synchronously collected by the functional near-infrared spectroscopy (fNIRS) device. Activation level and functional connectivity of these brain regions were subsequently calculated and statistically analyzed. For both groups, the VR and VRA tasks induced activation of larger cortical areas. Stroke group had higher average cortical activation in all three tasks compared to healthy group, especially in the prefrontal cortex (${P} \lt 0.05$). Functional connectivity was weaker in the stroke group than in the healthy group across most regions, but was significantly stronger across some regions of the right hemisphere. These findings suggest significant differences in activation patterns across three tasks. In addition, multi-sensory stimulation can promote functional communication between more brain regions in patients. It has potential for neuromodulation in rehabilitation training by setting up different sensory stimulation modalities. [ABSTRACT FROM AUTHOR]

Cochlear implant (CI) is a neural prosthesis that can restore hearing for patients with severe to profound hearing loss. Observed variability in auditory rehabilitation outcomes following cochlear implantation may be due to cerebral reorganization. Electroencephalography (EEG), favored for its CI compatibility and non-invasiveness, has become a staple in clinical objective assessments of cerebral plasticity post-implantation. However, the electrical activity of CI distorts neural responses, and EEG susceptibility to these artifacts presents significant challenges in obtaining reliable neural responses. Despite the use of various artifact removal techniques in previous studies, the automatic identification and reduction of CI artifacts while minimizing information loss or damage remains a pressing issue in objectively assessing advanced auditory functions in CI recipients. To address this problem, we propose an approach that combines machine learning algorithms—specifically, Support Vector Machines (SVM)—along with Independent Component Analysis (ICA) and Ensemble Empirical Mode Decomposition (EEMD) to automatically detect and minimize electrical artifacts in EEG data. The innovation of this research is the automatic detection of CI artifacts using the temporal properties of EEG signals. By applying EEMD and ICA, we can process and remove the identified CI artifacts from the affected EEG channels, yielding a refined signal. Comparative analysis in the temporal, frequency, and spatial domains suggests that the corrected EEG recordings of CI recipients closely align with those of peers with normal hearing, signifying the restoration of reliable neural responses across the entire scalp while eliminating CI artifacts. [ABSTRACT FROM AUTHOR]

High-speed videoendoscopy is a major tool for quantitative laryngology. Glottis segmentation and glottal midline detection are crucial for computing vocal fold-specific, quantitative parameters. However, fully automated solutions show limited clinical applicability. Especially unbiased glottal midline detection remains a challenging problem. We developed a multitask deep neural network for glottis segmentation and glottal midline detection. We used techniques from pose estimation to estimate the anterior and posterior points in endoscopy images. Neural networks were set up in TensorFlow/Keras and trained and evaluated with the BAGLS dataset. We found that a dual decoder deep neural network termed GlottisNetV2 outperforms the previously proposed GlottisNet in terms of MAPE on the test dataset (1.85% to 6.3%) while converging faster. Using various hyperparameter tunings, we allow fast and directed training. Using temporal variant data on an additional data set designed for this task, we can improve the median prediction accuracy from 2.1% to 1.76% when using 12 consecutive frames and additional temporal filtering. We found that temporal glottal midline detection using a dual decoder architecture together with keypoint estimation allows accurate midline prediction. We show that our proposed architecture allows stable and reliable glottal midline predictions ready for clinical use and analysis of symmetry measures.

Deep learning-based methods have become an active research area in medical imaging. Malaria is diagnosed by testing red blood cells. Deep learning methods can be used to distinguish malaria infected cell images from non-infected cell images. The small number of malaria dataset may limit the application of deep learning. Moreover, the infected area in the cell images is generally vague and small, requiring more complex models and a larger dataset to train on. Motivated by the tendency of humans to highlight important words when reading, we propose a simple neural network training strategy for highlighting the infected pixel regions that are mainly responsible for malaria cell classification. In our experiments on the NIH(National Institutes of Health) malaria dataset available in public domain, the proposed method significantly improved classification accuracy for our four different sized models, ranging from simple to complex including Resnet and Mobilenet. Our proposed method significantly improved classification accuracy. The result indicate that approach achieves a classification accuracy of 97.2%, compared to 94.49% for a baseline model. In addition, we show the superiority of the proposed strategy by providing an analysis on the magnitude of weight parameters in terms of regularization.

Automatic analysis of colocalizing biological structures in multi-channel fluorescence microscopy images is an important task to quantify and understand biological processes at high spatial-temporal resolution. Here, we introduce a software suite for colocalization analysis of spot-like objects in multi-channel fluorescence microscopy images. The software suite consists of ColocQuant and ColocJ, and is easy to use for biologists. ColocQuant is a Python-based software with graphical user interface to quantify colocalization of particles in two or three channels. Object-based colocalization is performed by an efficient multi-dimensional graph-based $k$ -d-tree approach, which determines nearest neighbors involved in double or triple colocalization. ColocJ enables efficient and intuitive visualization of the color composition of colocalizations by a Maxwell color triangle and a color ribbon. Colocalization information can be visualized for an entire image or a selected region-of-interest. In addition, global statistics of the particle intensity, particle size, and the number of colocalizations over time are provided. The colocalization analysis results can be exported and used in other software. We illustrate the application of our software suite for multi-channel live cell fluorescence microscopy image sequences of viral proteins in hepatitis C virus infected cells. We performed two-channel and three-channel colocalization analysis.

Content-based medical image retrieval is an important diagnostic tool that improves the explainability of computer-aided diagnosis systems and provides decision-making support to healthcare professionals. A common approach to content-based image retrieval is learning a distance metric by transforming images into a feature space where the distance between samples is a similarity measure. Proxy metric learning methods are effective at learning this transformation due to the use of proxy feature vectors that enable efficient learning. Training with a distance-based classification loss enables a single proxy model to be suitable for both retrieval and classification. However, these methods are designed only for single-label data, making them unsuitable for multimorbidity medical images. Addressing this, we propose a novel multi-label proxy metric learning method for content-based image retrieval and classification. Unlike existing proxy-based methods, training samples assign to multiple proxies that span multiple class labels. This results in a feature space that encodes the complex relationships between diseases. We introduce negative proxies to better encode the relationships between samples without detected diseases. The efficacy of our approach is demonstrated experimentally on two multimorbidity radiology datasets. Results show that our method outperforms state-of-the-art image retrieval systems and baseline approaches. Our method is clinically significant as it improves on two key factors shown to affect medical professionals’ willingness to use computer-aided diagnosis systems: accuracy and interpretability.

This work presents a systematic evaluation of the effectiveness of an air-operated microwave imaging (MWI) system for the detection of arbitrarily oriented thin fractures in superficial bones, like the tibia. This includes the proposal of a new compact, portable setup where a single Vivaldi antenna performs a semi-cylindrical scan of the limb. The antenna is operated in monostatic radar mode, near the skin but without contact, thus ensuring hygiene and patient comfort during the exam. The image is reconstructed using a wave-migration algorithm in the frequency domain combined with an adaptative algorithm based on singular value decomposition which removes the skin artifact taking into account the non-uniform bone profile and tissue cover. The study investigates the system resolution, the robustness of the method to the uncertainty of the permittivity and thickness of the involved tested tissues, as well as the robustness to involuntary patient movement. The experimental validation was performed for the first time on an integral ex-vivo animal leg, with all tissues present, including skin and fur. It confirmed both the effectiveness of the method, and the feasibility of the setup.

Alzheimer’s disease (AD) is the most frequent method of dementia and ranks the fifth-leading disease. Brain imaging or neuroimaging like magnetic resonance imaging (MRI) was utilized in the medicinal analysis of brain condition to allow visualization of infrastructure and brain functionality. Machine learning (ML) approaches on MRI are utilized in the analysis of AD to accelerate the analysis procedure and assist doctors. But, in typical ML approaches utilizing handcrafted extracting feature systems on MRI are difficult, and requires the contribution of expert users. So, executing deep learning (DL) as an automatic extracting feature system is minimizing require for extracting features and automate the procedure. This article introduces a novel Biomedical Alzheimer’s Disease Diagnosis using Enhanced Manta ray Foraging Optimization based Deep Learning (ADD-EMRFODL) technique on brain MRI. The presented ADD-EMRFODL technique exploits Gabor filtering (GF) technique as a pre-processing step. Also, the presented ADD-EMRFODL technique utilizes densely connected network (DenseNet-121) model to derive feature vectors. Finally, the ADD process is performed via EMRFO with back propagation neural network (BPNN) classifier. The parameter tuning of the BPNN occurs utilizing the EMRFO algorithm with the intention of enhanced classification accuracy. To depict the improvised ADD performance of the ADD-EMRFODL methodology, a comprehensive set of simulations was effectuated. The results stated the improved outcomes of the ADD-EMRFODL algorithm over other current methodologies.

The domain of Artificial Intelligence (AI) is made important strides recently, leading to developments in several domains comprising biomedical diagnostics and research. The procedure of AI-based systems in biomedical analytics takes opened up novel avenues for the progress of disease analysis, drug discovery, and treatment. Cancer is the second major reason of death worldwide; around one in every six people pass away suffering from it. Among several kinds of cancers, the colon and lung variations are the most frequent and deadliest ones. Initial detection of conditions on both fronts significantly reduces the probability of mortality. Deep learning (DL) and Machine learning (ML) systems are exploited to speed up such cancer detection, permitting researchers to analyze a huge count of patients in a lesser time count and at a minimal cost. This study develops a new Biomedical Image Analysis for Colon and Lung Cancer Detection using Tuna Swarm Algorithm with Deep Learning (BICLCD-TSADL) model. The presented BICLCD-TSADL technique examines the biomedical images for the identification and classification of colon and lung cancer. To accomplish this, the BICLCD-TSADL technique applies Gabor filtering (GF) to preprocess the input images. In addition, the BICLCD-TSADL technique employs a GhostNet feature extractor to create a collection of feature vectors. Moreover, AFAO was executed to adjust the hyperparameters of the GhostNet technique. Furthermore, the TSA with echo state network (ESN) classifier is utilized for detecting lung and colon cancer. To demonstrate the more incredible outcome of the BICLCD-TSADL system, an extensive experimental outcome is carried out. The comprehensive comparative analysis highlighted the greater efficiency of the BICLCD-TSADL technique with other approaches with maximum accuracy of 99.33%.

Chronic otitis media is characterized by recurrent infections, leading to serious complications, such as meningitis, facial palsy, and skull base osteomyelitis. Therefore, active treatment based on early diagnosis is essential. This study developed a multi-modal multi-fusion (MMMF) model that automatically diagnoses ear diseases by applying endoscopic images of the tympanic membrane (TM) and pure-tone audiometry (PTA) data to a deep learning model. The primary aim of the proposed MMMF model is adding “normal with hearing loss” as a category, and improving the diagnostic accuracy of the conventional four ear diseases: normal, TM perforation, retraction, and cholesteatoma. To this end, the MMMF model was trained on 1,480 endoscopic images of the TM and PTA data to distinguish five ear disease states: normal, TM perforation, retraction, cholesteatoma, and normal (hearing loss). It employs a feature fusion strategy of cross-attention, concatenation, and gated multi-modal units in a multi-modal architecture encompassing a convolutional neural network (CNN) and multi-layer perceptron. We expanded the classification capability to include an additional category, normal (hearing loss), thereby enhancing the diagnostic performance of extant ear disease classification. The MMMF model demonstrated superior performance when implemented with EfficientNet-B7, achieving 92.9% accuracy and 90.9% recall, thereby outpacing the existing feature fusion methods. In addition, five-fold cross-validation experiments were conducted, in which the model consistently demonstrated robust performance when endoscopic images of the TM and PTA data were applied to the deep learning model across all datasets. The proposed MMMF model is the first to include a category of normal ear disease state with hearing loss. The developed model demonstrated superior performance compared to existing CNN models and feature fusion methods. Consequently, this study substantiates the utility of simultaneously applying PTA data and endoscopic images of the TM for the automated diagnosis of ear diseases in clinical settings and validates the usefulness of the multi-fusion method.

The current pandemic has necessitated rapid and automatic detection of coronavirus disease (COVID-19) infections. Various artificial intelligence functionalities coupled with biomedical images can be utilized to efficiently detect these infections and recommend a prompt response (curative intervention) to limit the virus’s spread. In particular, biomedical imaging could help to visualize the internal organs of the human body and disorders that affect them. One of them is chest X-rays (CXRs) which has widely been used for preventive medicine or disease screening. However, when it comes to detecting COVID-19 from CXR images, most of the approaches rely on standard image classification algorithms, which have limitations with low identification accuracy and improper extraction of key features. As a result, a convolutional neural network (CNN)-based fusion network has been developed for automated COVID-19 screening in this study. First, using attention networks and multiple fine-tuned CNN models, we extract key features that are resistant to overfitting. We then employ a locally connected layer to create a weighted combination of these models for final COVID-19 detection. Using a publicly available dataset of CXR images from healthy subjects as well as COVID-19 and pneumonia cases, we evaluated the predictive capabilities of our proposed model. Test results demonstrate that the proposed fusion model performs favorably compared to individual CNN models.

Deformable image registration plays a critical role in various tasks of medical image analysis. A successful registration algorithm, either derived from conventional energy optimization or deep networks, requires tremendous efforts from computer experts to well design registration energy or to carefully tune network architectures with respect to medical data available for a given registration task/scenario. This paper proposes an automated learning registration algorithm (AutoReg) that cooperatively optimizes both architectures and their corresponding training objectives, enabling non-computer experts to conveniently find off-the-shelf registration algorithms for various registration scenarios. Specifically, we establish a triple-level framework to embrace the searching for both network architectures and objectives with a cooperating optimization. Extensive experiments on multiple volumetric datasets and various registration scenarios demonstrate that AutoReg can automatically learn an optimal deep registration network for given volumes and achieve state-of-the-art performance. The automatically learned network also improves computational efficiency over the mainstream UNet architecture from 0.558 to 0.270 seconds for a volume pair on the same configuration. [ABSTRACT FROM AUTHOR]

In image processing, images are usually composed of partial views due to the uncertainty of collection and how to efficiently process these images, which is called incomplete multi-view learning, has attracted widespread attention. The incompleteness and diversity of multi-view data enlarges the difficulty of annotation, resulting in the divergence of label distribution between the training and testing data, named as label shift. However, existing incomplete multi-view methods generally assume that the label distribution is consistent and rarely consider the label shift scenario. To address this new but important challenge, we propose a novel framework termed as Incomplete Multi-view Learning under Label Shift (IMLLS). In this framework, we first give the formal definitions of IMLLS and the bidirectional complete representation which describes the intrinsic and common structure. Then, a multilayer perceptron which combines the reconstruction and classification loss is employed to learn the latent representation, whose existence, consistency and universality are proved with the theoretical satisfaction of label shift assumption. After that, to align the label distribution, the learned representation and trained source classifier are used to estimate the importance weight by designing a new estimation scheme which balances the error generated by finite samples in theory. Finally, the trained classifier reweighted by the estimated weight is fine-tuned to reduce the gap between the source and target representations. Extensive experimental results validate the effectiveness of our algorithm over existing state-of-the-arts methods in various aspects, together with its effectiveness in discriminating schizophrenic patients from healthy controls. [ABSTRACT FROM AUTHOR]

The treatment of patients with balance disorders is an urgent problem to be solved by the medical community. The causes of balance disorders are diverse. An aging population, traffic accidents, stroke, genetic diseases and so on are all possible factors. It has brought great pain and inconvenience to patients and their families. At present, there are two main types of assisted rehabilitation training robots for patients with balance disorders: exoskeleton robots and end robots. The exoskeleton robot is generally installed on the outside of the patient’s body to follow their movement, which can support the weight of the body and provide power support to help the patient train and recover lower limb ability. The use of end robots is usually to secure the patient’s foot to the motion platform and control the pedal to drive the lower limbs to conduct gait training. Such passive training is more suitable for patients with severe disorders. The patient has low awareness of active participation. This paper focuses on research on end rehabilitation training robots for balance disorders. In this paper, a robotic system for rehabilitation training of patients with balance disorders is invented. The robot body is a 9 degree of freedom (DOF) redundant series-parallel hybrid motion platform. Two sets of motion platforms with symmetrical mirror images are used together to simulate different motion modes of the human body and drive the human body to move. Each set of motion platforms is composed of a 6-DOF vestibular parallel device and a 3-DOF proprioception parallel device. It has the advantages of DOF decoupling and fast response, proposing a new structural form for the design of proprioceptive and vestibular simulation platforms. The robot’s functional level can be divided into a vestibular sense module and a proprioception module according to the structure. The two modules can work independently to achieve different functions or work together to achieve complex motion and multisensory fusion. This robot is a redundant mechanism device with 9 DOFs. Through a reasonable distribution of DOF and motion, the robot’s working space can be increased, and the robot’s flexibility and motion performance can be improved. In this paper, a trajectory tracking control algorithm for vestibular and proprioceptive simulation is proposed, which can provide unlimited body sense training for patients within the robot’s limited motion range. [ABSTRACT FROM AUTHOR]

Tinnitus is associated with abnormal functional connectivity of multiple regions of the brain. However, previous analytic methods have disregarded information on the direction of functional connectivity, leading to only a moderate efficacy of pretreatment planning. We hypothesized that the pattern of directional functional connectivity can provide key information on treatment outcomes. Sixty-four participants were enrolled in this study: eighteen patients with tinnitus were categorized into the effective group, twenty-two patients into the ineffective group, and twenty-four healthy participants into the healthy control group. We acquired resting-state functional magnetic resonance images prior to sound therapy and constructed an effective connectivity network of the three groups using an artificial bee colony algorithm and transfer entropy. The key feature of patients with tinnitus was the significantly increased signal output of the sensory network, including the auditory, visual, and somatosensory networks, and parts of the motor network. This provided critical insights into the gain theory of tinnitus development. The altered pattern of functional information orchestration, represented by a higher degree of hypervigilance-driven attention and enhanced multisensory integration, may explain poor clinical outcomes. The activated gating function of the thalamus is one of the key factors for a good prognosis in tinnitus treatment. We developed a novel method for analyzing effective connectivity, facilitating an understanding of the tinnitus mechanism and treatment outcome expectation based on the direction of information flow. [ABSTRACT FROM AUTHOR]

The inverse synthetic aperture radar (ISAR) images are often afflicted by boundary-blurring, discontinuity, sidelobe effects of strong scattering points, a large dynamic range of gray values, and azimuth defocus, which pose significant challenges to image segmentation. This letter proposes a novel semantic segmentation method for ISAR images of space targets. The method is based on contrastive learning (CL) and nonlocal Unet (NL-Unet). First, the method roughly segments the target contour using binary semantic tags to remove sidelobe interference and image noise. Then, the nonlocal (NL) self-attentive mechanism with a global perceptual field is used to exploit the structural symmetry of the ISAR image. Finally, to improve the segmentation ability of target small parts, the method adopts a training method based on CL to overcome the relatively weak problem of the supervised learning model. The proposed method outperforms existing methods on the simulated ISAR dataset. Moreover, it is practical and can be directly generalized to the real-measured dataset without retraining. [ABSTRACT FROM AUTHOR]

This article proposes a novel fuzzy-inspired nanobiosensing framework for soft classification of tumor by using multiple features of the tumor, where interrelationships between different features are accounted for. For illustration purpose, two features of breast cancer, namely, tumor tissue abnormality and tumor foci separation, are utilized to determine the disease state. The sensing performance is then evaluated through the fuzzy relation transfer analysis. This analysis enables an intuitive yet systematic way to characterize the variation of classification fuzziness occurred during the sensing process facilitated by nanoscale contrast agents. Subsequently, the mean absolute error of the pre- and post-contrast membership functions is employed to evaluate the sensing integrity. Finally, numerical results are presented to demonstrate the principles of the proposed framework. [ABSTRACT FROM AUTHOR]

Microbubbles and ultrasound are no longer teaming up only as a way to enhance images. New technologies are now using the two to create physical pathways into cells for easier drug delivery, even into the cells of highly drug-resistant cancerous tumors and across the blood–brain barrier. Going further, a Texas research group has developed drug-carrying microbubbles that can complete targeted delivery themselves, and another group in Missouri has shown that two-way traffic in channels across the blood–brain barrier also allow biomarkers to flow out, which provides a new window into the brain as well as brain diseases. [ABSTRACT FROM AUTHOR]

Interventional therapy is popular in modern surgical procedures for the treatment of vascular diseases. However, it remains challenging to smoothly steer the guidewire/catheter into distal tortuous lumen environments. In this article, a wired magnetic microrobot (WMM) is proposed for approaching hard-to-reach regions. The WMM consists of a commercial guidewire and an assembled tip module, which has two working modalities with a magnetically triggered switch. The tethered mode has high efficiency and reliability, where the forward–backward motion is controlled by the feeding device and the steering motion is actuated by the directional external magnetic field. The untethered mode has enhanced flexibility, where the ejected helical bullet is wirelessly propelled by the rotating external magnetic field. A homemade actuation system is adopted for large-workspace magnetic control and medical imaging-based feedback. Targeted scanning is conducted based on real-time segmentation of the vessel region in ultrasound (US) images and estimation of the vascular distribution. Both transverse and longitudinal views are used for visual tracking under different modes. With the proposed system, the WMM can be navigated in a 3-D tubular structure over a distance of 1000 mm, and the whole procedure can be performed under US imaging monitoring. [ABSTRACT FROM AUTHOR]

Over the recent years, volume of medical images and related digital records, called electronic medical records, generated, shared, and stored by different intelligent devices, sensors, and Internet of medical things networks, to name a few, has drastically increased. Such records are shared by cloud providers for storage and further processing. However, an increasingly serious concern is the illegal copying, modification, and forgery of medical records. This article presents a visible and imperceptible medical data hiding technique, namely ViMDH, which can prevent to intellectual property theft of medical records. The carrier image is visibly marked with logo mark, which is suitable for owner identification and avoid illegal duplication, and then an imperceptible data hiding based on nonsubsampled shearlet transform (NSST), redundant discrete wavelet transform (RDWT), and multiresolution singular value decomposition is introduced. Finally, key-based encryption scheme designed by RDWT-RSVD ensure the security of the watermarking system. Under the experimental evaluation, our ViMDH is not only visible and imperceptible, but also has a satisfactory advantage in robustness and security compared with the traditional watermarking schemes. [ABSTRACT FROM AUTHOR]

Open-set recognition (OSR) has achieved significant importance in recent years. For a robust recognition system, we need to identify the right class from a myriad of knowns and unknowns. In this work, we build and compare OSR systems for patient activity recognition (PAR) using compact radar sensors in a hospital setting. Radar sensors are an important part of a privacy-preserving monitoring system. Specifically, the proposed approach is based on a deep discriminative representation network (DDRN) trained using the large margin cosine loss (LMCL) and triplet loss (TL). A probability of an inclusion model in the embedding space based on the Weibull distribution is able to separate knowns from unknowns. This overall approach limits the risk of open space and enables us to easily identify any unknown activities. Our experiments show that the proposed approach is significantly better for open-set human activity recognition (HAR) with radar when compared with the state-of-the-art open-set approaches. [ABSTRACT FROM AUTHOR]

In this article, we introduce a new neonatal facial expression database for pain analysis. This database, called facial expression of neonatal pain (FENP), contains 11,000 neonatal facial expression images associated with 106 Chinese neonates from two children's hospitals, i.e., the Children's Hospital Affiliated to Nanjing Medical University and Second Affiliated Hospital Affiliated to Nanjing Medical University in China. The facial expression images cover four categories of facial expressions, i.e., severe pain expression, mild pain expression, crying expression and calmness expression, where each category contains 2750 neonatal facial expression images. Based on this database, we also investigate the pain facial expression recognition problem using several state-of-the-art facial expression features and expression recognition methods, such as Gabor+SVM, LBP+SVM, HOG+SVM, LBP+HOG+SVM, and several Convolutional Neural Network (CNN) methods (including AlexNet, VGGNet, GoogLeNet, ResNet and DenseNet). The experimental results indicate that the proposed neonatal pain facial expression database is very suitable for the study of both neonatal pain and facial expression recognition. Moreover, the FENP database is publicly available after signing a license agreement (the users can contact Jingjie Yan (yanjingjie@njupt.edu.cn), Guanming Lu (lugm@njupt.edu.cn)) or Xiaonan Li (xnli@njmu.edu.cn). [ABSTRACT FROM AUTHOR]

Self-supervised representation learning has been extremely successful in medical image analysis, as it requires no human annotations to provide transferable representations for downstream tasks. Recent self-supervised learning methods are dominated by noise-contrastive estimation (NCE, also known as contrastive learning), which aims to learn invariant visual representations by contrasting one homogeneous image pair with a large number of heterogeneous image pairs in each training step. Nonetheless, NCE-based approaches still suffer from one major problem that is one homogeneous pair is not enough to extract robust and invariant semantic information. Inspired by the archetypical triplet loss, we propose GraVIS, which is specifically optimized for learning self-supervised features from dermatology images, to group homogeneous dermatology images while separating heterogeneous ones. In addition, a hardness-aware attention is introduced and incorporated to address the importance of homogeneous image views with similar appearance instead of those dissimilar homogeneous ones. GraVIS significantly outperforms its transfer learning and self-supervised learning counterparts in both lesion segmentation and disease classification tasks, sometimes by 5 percents under extremely limited supervision. More importantly, when equipped with the pre-trained weights provided by GraVIS, a single model could achieve better results than winners that heavily rely on ensemble strategies in the well-known ISIC 2017 challenge. Code is available at https://bit.ly/3xiFyjx. [ABSTRACT FROM AUTHOR]

Cerebrovascular segmentation in time-of-flight magnetic resonance angiography (TOF-MRA) volumes is essential for a variety of diagnostic and analytical applications. However, accurate cerebrovascular segmentation in 3D TOF-MRA is faced with multiple issues, including vast variations in cerebrovascular morphology and intensity, noisy background, and severe class imbalance between foreground cerebral vessels and background. In this work, a 3D adversarial network model called A-SegAN is proposed to segment cerebral vessels in TOF-MRA volumes. The proposed model is composed of a segmentation network A-SegS to predict segmentation maps, and a critic network A-SegC to discriminate predictions from ground truth. Based on this model, the aforementioned issues are addressed by the prevailing visual attention mechanism. First, A-SegS is incorporated with feature-attention blocks to filter out discriminative feature maps, though the cerebrovascular has varied appearances. Second, a hard-example-attention loss is exploited to boost the training of A-SegS on hard samples. Further, A-SegC is combined with an input-attention layer to attach importance to foreground cerebrovascular class. The proposed methods were evaluated on a self-constructed voxel-wise annotated cerebrovascular TOF-MRA segmentation dataset, and experimental results indicate that A-SegAN achieves competitive or better cerebrovascular segmentation results compared to other deep learning methods, effectively alleviating the above issues. [ABSTRACT FROM AUTHOR]

The recent success of learning-based algorithms can be greatly attributed to the immense amount of annotated data used for training. Yet, many datasets lack annotations due to the high costs associated with labeling, resulting in degraded performances of deep learning methods. Self-supervised learning is frequently adopted to mitigate the reliance on massive labeled datasets since it exploits unlabeled data to learn relevant feature representations. In this work, we propose SS-StyleGAN, a self-supervised approach for image annotation and classification suitable for extremely small annotated datasets. This novel framework adds self-supervision to the StyleGAN architecture by integrating an encoder that learns the embedding to the StyleGAN latent space, which is well-known for its disentangled properties. The learned latent space enables the smart selection of representatives from the data to be labeled for improved classification performance. We show that the proposed method attains strong classification results using small labeled datasets of sizes 50 and even 10. We demonstrate the superiority of our approach for the tasks of COVID-19 and liver tumor pathology identification. [ABSTRACT FROM AUTHOR]

In this article, the concept of cryptographic algorithms is used as an efficient access control mechanism for Internet of Medical Things-based health care system. The algorithms, such as Rivest Cipher (RC6), are used to generate the key value, and elliptic curve digital signature algorithm will encrypt the key value from RC6 and the encrypted output is send to secure hash algorithm (SHA256) for hashing process based on cipher value which improves data integrity. Furthermore, these high-security algorithms are used to provide availability and confidentiality to protect sensitive information from implantable devices and strengthen the health care systems through enhanced services. Comprehensive experimental analysis and simulation results indicate that the proposed scheme is more secure against various known attacks, such as denial of service, router attack, and sensor attacks. This proposed system has better resistance protocols in analyzing the safety of patients. [ABSTRACT FROM AUTHOR]

The exponential growth of deep learning networks has allowed us to tackle complex tasks, even in fields as complicated as medicine. However, using these models requires a large corpus of data for the networks to be highly generalizable and with high performance. In this sense, data augmentation methods are widely used strategies to train networks with small data sets, being vital in medicine due to the limited access to data. A clear example of this is magnetic resonance imaging in pathology scans associated with cancer. In this vein, we compare the effect of several conventional data augmentation schemes on the ResNet50 network for brain tumor detection. In addition, we included our strategy based on principal component analysis. The training was performed with the network trained from zeros and transfer-learning, obtained from the ImageNet dataset. The investigation allowed us to achieve an F1 detection score of 92.34%. The score was achieved with the ResNet50 network through the proposed method and implementing the learning transfer. In addition, it was also concluded that the proposed method is different from the other conventional methods with a significance level of 0.05 through the Kruskal Wallis test statistic.

Background and objectives: Endovascular prosthesis placement is a predominant surgical procedure to reduce the risk of aneurysm rupture in abdominal aortic aneurysm patients. The formation of an endoleak is a major complication of stent-graft placement. Therefore, the aim of this study was to determine the influence of stent-graft’s spatial configuration on the risk of leakage under realistic flow conditions. Materials and Methods: We analyzed data collected from 10 male patients 55 ± 3 years old after CTA who had undergone endovascular treatment. Computational Fluid Dynamics technique was applied for the reconstruction of blood hemodynamic. Two cases of the stent-graft were analyzed each time, with and without the endoleak, reconstructed from one patient data with the endoleak. Endoleak-free geometries were prepared by virtual closure of the opening. Results: It was observed that high value of the blood velocity around the endoleak may provoke its slowly rupture, which may increase the blood flow to the aneurysm sack in the case of type II endoleaks. Moreover, the appearance of endoleak reduces the average blood velocity in the entire stent-graft which in some cases may contribute to blood coagulation. Furthermore, no clot can form at the site of the endoleak due to the high value of wall shear stress near of the endoleak which prevents thrombosis.

Objective: To identify radiomic and clinical features associated with post-ablation recurrence of AF, given that cardiac morphologic changes are associated with persistent atrial fibrillation (AF), and initiating triggers of AF often arise from the pulmonary veins which are targeted in ablation. Methods: Subjects with pre-ablation contrast CT scans prior to first-time catheter ablation for AF between 2014–2016 were retrospectively identified. A training dataset (D1) was constructed from left atrial and pulmonary vein morphometric features extracted from equal numbers of consecutively included subjects with and without AF recurrence determined at 1 year. The top-performing combination of feature selection and classifier methods based on C-statistic was evaluated on a validation dataset (D2), composed of subjects retrospectively identified between 2005–2010. Clinical models ( $\text{M}_{\mathrm {C}}$ ) were similarly evaluated and compared to radiomic ( $\text{M}_{\mathrm {R}}$ ) and radiomic-clinical models ( $\text{M}_{\mathrm {RC}}$ ), each independently validated on D2. Results: Of 150 subjects in D1, 108 received radiofrequency ablation and 42 received cryoballoon. Radiomic features of recurrence included greater right carina angle, reduced anterior-posterior atrial diameter, greater atrial volume normalized to height, and steeper right inferior pulmonary vein angle. Clinical features predicting recurrence included older age, greater BMI, hypertension, and warfarin use; apixaban use was associated with reduced recurrence. AF recurrence was predicted with radio-frequency ablation models on D2 subjects with C-statistics of 0.68, 0.63, and 0.70 for radiomic, clinical, and combined feature models, though these were not prognostic in patients treated with cryoballoon. Conclusions: Pulmonary vein morphology associated with increased likelihood of AF recurrence within 1 year of catheter ablation was identified on cardiac CT. Significance: Radiomic and clinical features-based predictive models may assist in identifying atrial fibrillation ablation candidates with greatest likelihood of successful outcome.

The x-ray computed tomography (CT) images with low dose are noisy and may contain photon starvation artifacts. The artifacts are location and direction dependent. Therefore, the common shift-invariant denoising filters do not work well. The state-of-the-art methods to process the low-dose CT images are image reconstruction based; they require the raw projection data. In many situations, the raw CT projections are not accessible. This paper suggests a method to denoise the low-dose CT image using the pseudo projections generated by the application of a forward projector on the low-dose CT image. The feasibility of the proposed method is demonstrated by real clinical data.

This paper studies the feasibility of Microwave Imaging (MWI) for detection of fractures in superficial bones like the tibia, using a simple and practical setup. First-responders could use it for fast preliminary diagnosis in emergency locations, where X-Rays are not available. It may prove valuable also for cases where X-ray are not recommended, e.g., length pregnant women or children. The method is inspired on the synthetic aperture radar technique. A single Vivaldi antenna is used to linearly scan the bone in the 8.3-11.1 GHz frequency range and collect the scattered fields. The system is operated in air, without the need for impractical impedance-matching immersion liquids. The image is reconstructed using a Kirchhoff migration algorithm. A Singular Value Decomposition (SVD) strategy is used to remove skin and background artifacts. To test this technique, a set of full-wave simulations and experiments were conducted on a multilayer phantom and on an ex-vivo animal bone. Results show that the system can detect and locate bone transverse fractures as small as 1 mm width and 13 mm deep, even when the bone is wrapped by 2 mm thick skin.

Advances in deep-learning-based pipelines have led to breakthroughs in a variety of microscopy image diagnostics. However, a sufficiently big training data set is usually difficult to obtain due to high annotation costs. In the case of banded chromosome images, the creation of big enough libraries is difficult for multiple pathologies due to the rarity of certain genetic disorders. Generative Adversarial Networks (GANs) have proven to be effective in generating synthetic images and extending training data sets. In our work, we implement a conditional GAN (cGAN) that allows generation of realistic single chromosome images following user-defined banding patterns. To this end, an image-to-image translation approach based on automatically created 2D chromosome segmentation label maps is used. Our validation shows promising results when synthesizing chromosomes with seen as well as unseen banding patterns. We believe that this approach can be exploited for data augmentation of chromosome data sets with structural abnormalities. Therefore, the proposed method could help to tackle medical image analysis problems such as data simulation, segmentation, detection, or classification in the field of cytogenetics.

Diabetes mellitus (DM) is a metabolic disorder characterized by increased blood glucose. The pathology can manifest itself with different conditions, including neuropathy, the main consequence of diabetic disease. Statistics show worrying figures worldwide, diagnosed an estimated 1.6 million people with DM by 2025. In this sense, alternative and automated methods are necessary to detect DM, allowing it to take the pertinent measures in its treatment and avoid critical complications, such as the diabetic foot. On the other hand, foot thermography is a promising tool that allows visualization of thermal patterns, patterns that are altered as a consequence of shear and friction associated with lower limb neuropathy. Based on these considerations, we explored different strategies to detect patients with DM from foot thermography in this research. Initially, the study focused on finding a classification index like Thermal Change Index (TCI). Subsequently, we used the deep convolutional neural networks paradigm, implementing 12 different data augmentation methods, of which four are conventional, and 8 are newly proposed methods. The results showed that the proposed and the conventional methods increased the network’s performance, where a 100% detection was achieved by weighting the DM probability percentages for both images of the feet. Finally, it was also possible to demonstrate the importance of transfer learning, which does not depend on the type of database, but on the data corpus with which the transfer was trained.

Coronavirus disease (COVID-19) was confirmed as a pandemic disease on February 11, 2020. The pandemic has already caused thousands of victims and infected several million people around the world. The aim of this work is to provide a Covid-19 infection screening tool. Currently, the most widely used clinical tool for detecting the presence of infection is the reverse transcription polymerase chain reaction (RT-PCR), which is expensive, less sensitive and requires the resource of specialized medical personnel. The use of X-ray images represents one of the latest challenges for the rapid diagnosis of the Covid-19 infection. This work involves the use of advanced artificial intelligence techniques for diagnosis using algorithms for classification purposes. The goal is to provide an automatic infection detection method while maximizing detection accuracy. A public database was used which includes images of COVID-19 patients, patients with viral pneumonia, patients with pulmonary opacity, and healthy patients. The methodology used in this study is based on transfer learning of pre-trained networks to alleviate the complexity of calculation. In particular, three different types of convolutional neural networks, namely, InceptionV3, ResNet50 and Xception, and the Vision Transformer are implemented. Experimental results show that the Vision Transformer outperforms convolutional architectures with a test accuracy of 99.3% vs 85.58% for ResNet50 (best among CNNs). Moreover, it is able to correctly distinguish among four different classes of chest X-ray images, whereas similar works only stop at three categories at most. The high accuracy of this computer-assisted diagnostic tool can significantly improve the speed and accuracy of COVID-19 diagnosis.

Ultrasound medical imaging is an entrenched and powerful tool for medical diagnosis. Image quality in ultrasound is mainly dependent on performance of piezoelectric transducer elements, which is further related to the electromechanical performance of the constituent piezoelectric materials. With rising need for piezoelectric materials with better performance and low cost, a highly $\langle 001\rangle $ textured piezo ceramic, Pb(Mg1/3Nb2/3)O3-Pb(Zr, Ti)O3, has been developed. Recently, textured ceramic materials can be produced at low cost and exhibit high piezoelectric strain constants and large electromechanical coupling coefficients. In this work, 15-MHz ultrasonic transducers with an effective aperture of 2.5 mm in diameter based on these highly $\langle 001\rangle $ textured ceramics have been successfully fabricated. The fabricated transducers achieved a central frequency of 15 MHz, a fractional bandwidth of 67% (at −6 dB), a high effective electromechanical coupling coefficient ${k}_{\text {eff}}$ of 0.55, and a low insertion loss (IL) of 21 dB. Ex vivo ultrasonic imaging of a porcine eyeball was used to assess the tomography quality of the transducer. The results show that utilized textured ceramic has a great potential in developing ultrasonic devices for biomedical imaging purposes. [ABSTRACT FROM AUTHOR]

Optical coherence tomography (OCT) is a widely-used modality in clinical imaging, which suffers from the speckle noise inevitably. Deep learning has proven its superior capability in OCT image denoising, while the difficulty of acquiring a large number of well-registered OCT image pairs limits the developments of paired learning methods. To solve this problem, some unpaired learning methods have been proposed, where the denoising networks can be trained with unpaired OCT data. However, majority of them are modified from the cycleGAN framework. These cycleGAN-based methods train at least two generators and two discriminators, while only one generator is needed for the inference. The dual-generator and dual-discriminator structures of cycleGAN-based methods demand a large amount of computing resource, which may be redundant for OCT denoising tasks. In this work, we propose a novel triplet cross-fusion learning (TCFL) strategy for unpaired OCT image denoising. The model complexity of our strategy is much lower than those of the cycleGAN-based methods. During training, the clean components and the noise components from the triplet of three unpaired images are cross-fused, helping the network extract more speckle noise information to improve the denoising accuracy. Furthermore, the TCFL-based network which is trained with triplets can deal with limited training data scenarios. The results demonstrate that the TCFL strategy outperforms state-of-the-art unpaired methods both qualitatively and quantitatively, and even achieves denoising performance comparable with paired methods. Code is available at: https://github.com/gengmufeng/TCFL-OCT. [ABSTRACT FROM AUTHOR]

Knee osteoarthritis (KOA) as a disabling joint disease has doubled in prevalence since the mid-20th century. Early diagnosis for the longitudinal KOA grades has been increasingly important for effective monitoring and intervention. Although recent studies have achieved promising performance for baseline KOA grading, longitudinal KOA grading has been seldom studied and the KOA domain knowledge has not been well explored yet. In this paper, a novel deep learning architecture, namely adversarial evolving neural network (A-ENN), is proposed for longitudinal grading of KOA severity. As the disease progresses from mild to severe level, ENN involves the progression patterns for accurately characterizing the disease by comparing an input image it to the template images of different KL grades using convolution and deconvolution computations. In addition, an adversarial training scheme with a discriminator is developed to obtain the evolution traces. Thus, the evolution traces as fine-grained domain knowledge are further fused with the general convolutional image representations for longitudinal grading. Note that ENN can be applied to other learning tasks together with existing deep architectures, in which the responses characterize progressive representations. Comprehensive experiments on the Osteoarthritis Initiative (OAI) dataset were conducted to evaluate the proposed method. An overall accuracy was achieved as 62.7%, with the baseline, 12-month, 24-month, 36-month, and 48-month accuracy as 64.6%, 63.9%, 63.2%, 61.8% and 60.2%, respectively. [ABSTRACT FROM AUTHOR]