Imaging Of Lesions Research Articles

Primary mesenchymal tumors of the prostate:18F FDG PET/CT findings.

Primary mesenchymal tumors of the prostate are very rare, and there is no systematic report on fluorine-18-fluorodeoxyglucose positron emission tomography (18F-FDG PET/CT). The aim of this research was to characterize mesenchymal tumors of the prostate on 18F-FDG PET/CT. We included 13 patients with pathologically confirmed mesenchymal neoplasms of the prostate. The location, size, maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), total glycolysis of the lesion (TLG), pathologic findings, and available imaging study of the tumors were reviewed. Of the 13 patients (median age, 38 years; ranged from 13 to 74 years), the mean size of the tumors was 8.8cm in diameter ranged from 5 to 16.1cm. SUVmax ranged from 3.1 to 17.6 (mean 9.5), MTV ranged from 4.9 to 398 cm3 (mean 109 cm3) and TLG ranged from 21.3 to 1216g (mean 96g). The seminal vesicles, rectum, and bladder are the most commonly affected sites of invasion, while metastasis typically occurs in the bones, lungs, and lymph nodes. Mesenchymal tumors of the prostate revealed large sizes and high SUVmax, MTV and TLG levels on 18F-FDG PET/CT. 18F-FDG PET/CT may be useful for accurate diagnosis, staging, and restaging, which plays an important role in the management of subsequent plans.

Enhanced skin cancer diagnosis: a deep feature extraction-based framework for the multi-classification of skin cancer utilizing dermoscopy images

Skin cancer is one of the most common, deadly, and widespread cancers worldwide. Early detection of skin cancer can lead to reduced death rates. A dermatologist or primary care physician can use a dermatoscope to inspect a patient to diagnose skin disorders visually. Early detection of skin cancer is essential, and in order to confirm the diagnosis and determine the most appropriate course of therapy, patients should undergo a biopsy and a histological evaluation. Significant advancements have been made recently as the accuracy of skin cancer categorization by automated deep learning systems matches that of dermatologists. Though progress has been made, there is still a lack of a widely accepted, clinically reliable method for diagnosing skin cancer. This article presented four variants of the Convolutional Neural Network (CNN) model (i.e., original CNN, no batch normalization CNN, few filters CNN, and strided CNN) for the classification and prediction of skin cancer in lesion images with the aim of helping physicians in their diagnosis. Further, it presents the hybrid models CNN-Support Vector Machine (CNNSVM), CNN-Random Forest (CNNRF), and CNN-Logistic Regression (CNNLR), using a grid search for the best parameters. Exploratory Data Analysis (EDA) and random oversampling are performed to normalize and balance the data. The CNN models (original CNN, strided, and CNNSVM) obtained an accuracy rate of 98%. In contrast, CNNRF and CNNLR obtained an accuracy rate of 99% for skin cancer prediction on a HAM10000 dataset of 10,015 dermoscopic images. The encouraging outcomes demonstrate the effectiveness of the proposed method and show that improving the performance of skin cancer diagnosis requires including the patient's metadata with the lesion image.

Automatic Watershed Segmentation of Cancerous Lesions in Unsupervised Breast Histology Images

Segmentation of nuclei in histology images is key in analyzing and quantifying morphology changes of nuclei features and tissue structures. Conventional diagnosis, segmenting, and detection methods have relied heavily on the manual-visual inspection of histology images. These methods are only effective on clearly visible cancerous lesions on histology images thus limited in their performance due to the complexity of tissue structures in histology images. Hence, early detection of breast cancer is key for treatment and profits from Computer-Aided-Diagnostic (CAD) systems introduced to efficiently and automatically segment and detect nuclei cells in pathology. This paper proposes, an automatic watershed segmentation method of cancerous lesions in unsupervised human breast histology images. Firstly, this approach pre-processes data through various augmentation methods to increase the size of dataset images, then a stain normalization technique is applied to these augmented images to isolate nuclei features from tissue structures. Secondly, data enhancement techniques namely; erosion, dilation, and distance transform are used to highlight foreground and background pixels while removing unwanted regions from the highlighted nuclei objects on the image. Consequently, the connected components method groups these highlighted pixel components with similar intensity values and, assigns them to their relevant labeled component binary mask. Once all binary masked groups have been determined, a deep-learning recurrent neural network from the Keras architecture uses this information to automatically segment nuclei objects with cancerous lesions and their edges on the image via watershed filling. This segmentation method is evaluated on an unsupervised, augmented human breast cancer histology dataset of 11,151 images. This proposed method produced a significant evaluation result of 98% F1-accuracy score.

Double-Condensing Attention Condenser: Leveraging Attention in Deep Learning to Detect Skin Cancer from Skin Lesion Images

Skin cancer is the most common type of cancer in the United States and is estimated to affect one in five Americans. Recent advances have demonstrated strong performance on skin cancer detection, as exemplified by state of the art performance in the SIIM-ISIC Melanoma Classification Challenge; however, these solutions leverage ensembles of complex deep neural architectures requiring immense storage and computation costs, and therefore may not be tractable. A recent movement for TinyML applications is integrating Double-Condensing Attention Condensers (DC-AC) into a self-attention neural network backbone architecture to allow for faster and more efficient computation. This paper explores leveraging an efficient self-attention structure to detect skin cancer in skin lesion images and introduces a deep neural network design with DC-AC customized for skin cancer detection from skin lesion images. We demonstrate that our approach with only 1.6 million parameters and 0.32 GFLOPs achieves better performance compared to traditional architecture designs as it obtains an area under the ROC curve of 0.90 on the public ISIC 2020 test set and 0.89 on the private ISIC test set, over 0.13 above the best Cancer-Net SCa network architecture design. The final model is publicly available as a part of a global open-source initiative dedicated to accelerating advancement in machine learning to aid clinicians in the fight against cancer. Future work of this research includes iterating on the design of the selected network architecture and refining the approach to generalize to other forms of cancer.

ScSEETV‐Net: Spatial and Channel Squeeze‐Excitation and Edge Attention Guidance V‐Shaped Network for Skin Lesion Segmentation

Early detection of skin cancer ensures the survival of many cases. There are still challenges in segmenting dermoscopic skin lesion images. Artifacts in the lesion images, such as various dirt, hairs, low contrast, and unclear boundaries, are challenges that affect segmentation accuracy. Convolutional neural networks have brought success in skin lesion segmentation. U‐shaped and V‐shaped deep learning‐based segmentation architectures learn boundary information in the first layers. However, this information becomes weaker in the following layers. Herein, the Edge‐aTtention module is added to the V‐Net architecture to move edge information to the last layer, and the spatial and channel squeeze‐excitation module is added to emphasize high‐level features by recalibrating the channel information to learn lesion boundaries better. The scSEETV‐Net is supported by fusing the binary cross‐entropy, which calculates the loss on a pixel‐based, and the focal Twersky loss function, which has significant success in class imbalances. The proposed architecture achieves 0.9212 Jaccard and 0.9552 Dice in the ISIC2016 dataset, 0.8273 Jaccard and 0.8949 Dice in the ISIC2017 dataset, and 0.8070 Jaccard and 0.8831 Dice in the ISIC2018 dataset. Comparative analyses show that the proposed methodology outperforms the state‐of‐the‐art techniques in the literature.

Evaluation of a novel forward-looking optical coherence tomography probe for endoscopic applications: an ex vivo feasibility study.

As a result of recent advances in the development of small microelectromechanical system mirrors, a novel forward-looking optical coherence tomography (OCT) probe with a uniquely large field of view is being commercially developed. The aim of this study is to prospectively assess the feasibility of this advanced OCT probe in interpreting ex vivo images of colorectal polyp tissue and to identify necessary steps for further development. A total of 13 colorectal lesions from 9 patients, removed during endoscopic resection, were imaged ex vivo with the OCT device and compared with histopathological images that served as the gold standard for diagnostics. Normal tissue from one patient, removed during the endoscopic procedure, was imaged as a negative control. We assessed the presence of features indicative for polyp type and degree of dysplasia, by comparing OCT images to histopathological images and by evaluating the presence of OCT-specific features identified by previous studies, such as effacement (loss of layered tissue structure), a hyperreflective epithelial layer, and irregularity of the surface. As verified by corresponding histological images, tissue structures such as blood vessels and tissue layers could be distinguished in OCT images of the normal tissue sample. Detailed structures on histological images such as crypts and cell nuclei could not be identified in the OCT images. However, we did identify OCT features specific for colorectal lesions, such as effacement and a hyperreflective epithelial layer. In general, the imaging depth was about 1mm. Some relevant tissue structures could be observed in OCT images of the novel device. However, some adaptations, such as increasing imaging depth using a laser with a longer central wavelength, are required to improve its clinical value for the imaging of colorectal lesions.

Application of machine learning for the differentiation of thymomas and thymic cysts using deep transfer learning: A multi-center comparison of diagnostic performance based on different dimensional models.

This study aimed to evaluate the feasibility and performance of deep transfer learning (DTL) networks with different types and dimensions in differentiating thymomas from thymic cysts in a retrospective cohort. Based on chest-enhanced computed tomography (CT), the region of interest was delineated, and the maximum cross section of the lesion was selected as the input image. Five convolutional neural networks (CNNs) and Vision Transformer (ViT) were used to construct a 2D DTL model. The 2D model constructed by the maximum section (n) and the upper and lower layers (n - 1, n + 1) of the lesion was used for feature extraction, and the features were selected. The remaining features were pre-fused to construct a 2.5D model. The whole lesion image was selected for input and constructing a 3D model. In the 2D model, the area under curve (AUC) of Resnet50 was 0.950 in the training cohort and 0.907 in the internal validation cohort. In the 2.5D model, the AUCs of Vgg11 in the internal validation cohort and external validation cohort 1 were 0.937 and 0.965, respectively. The AUCs of Inception_v3 in the training cohort and external validation cohort 2 were 0.981 and 0.950, respectively. The AUC values of 3D_Resnet50 in the four cohorts were 0.987, 0.937, 0.938, and 0.905. The DTL model based on multiple different dimensions can be used as a highly sensitive and specific tool for the non-invasive differential diagnosis of thymomas and thymic cysts to assist clinicians in decision-making.

The Evolution of Artificial Intelligence in Medical Imaging: From Computer Science to Machine and Deep Learning.

Artificial intelligence (AI), the wide spectrum of technologies aiming to give machines or computers the ability to perform human-like cognitive functions, began in the 1940s with the first abstract models of intelligent machines. Soon after, in the 1950s and 1960s, machine learning algorithms such as neural networks and decision trees ignited significant enthusiasm. More recent advancements include the refinement of learning algorithms, the development of convolutional neural networks to efficiently analyze images, and methods to synthesize new images. This renewed enthusiasm was also due to the increase in computational power with graphical processing units and the availability of large digital databases to be mined by neural networks. AI soon began to be applied in medicine, first through expert systems designed to support the clinician's decision and later with neural networks for the detection, classification, or segmentation of malignant lesions in medical images. A recent prospective clinical trial demonstrated the non-inferiority of AI alone compared with a double reading by two radiologists on screening mammography. Natural language processing, recurrent neural networks, transformers, and generative models have both improved the capabilities of making an automated reading of medical images and moved AI to new domains, including the text analysis of electronic health records, image self-labeling, and self-reporting. The availability of open-source and free libraries, as well as powerful computing resources, has greatly facilitated the adoption of deep learning by researchers and clinicians. Key concerns surrounding AI in healthcare include the need for clinical trials to demonstrate efficacy, the perception of AI tools as 'black boxes' that require greater interpretability and explainability, and ethical issues related to ensuring fairness and trustworthiness in AI systems. Thanks to its versatility and impressive results, AI is one of the most promising resources for frontier research and applications in medicine, in particular for oncological applications.

Image-Based Generative Artificial Intelligence in Radiology: Comprehensive Updates.

Generative artificial intelligence (AI) has been applied to images for image quality enhancement, domain transfer, and augmentation of training data for AI modeling in various medical fields. Image-generative AI can produce large amounts of unannotated imaging data, which facilitates multiple downstream deep-learning tasks. However, their evaluation methods and clinical utility have not been thoroughly reviewed. This article summarizes commonly used generative adversarial networks and diffusion models. In addition, it summarizes their utility in clinical tasks in the field of radiology, such as direct image utilization, lesion detection, segmentation, and diagnosis. This article aims to guide readers regarding radiology practice and research using image-generative AI by 1) reviewing basic theories of image-generative AI, 2) discussing the methods used to evaluate the generated images, 3) outlining the clinical and research utility of generated images, and 4) discussing the issue of hallucinations.

Exploring the feasibility of FOCUS DWI with deep learning reconstruction for breast cancer diagnosis: A comparative study with conventional DWI.

This study compared field-of-view (FOV) optimized and constrained undistorted single-shot diffusion-weighted imaging (FOCUS DWI) with deep-learning-based reconstruction (DLR) to conventional DWI for breast imaging. This study prospectively enrolled 49 female patients suspected of breast cancer from July to December 2023. The patients underwent conventional and FOCUS breast DWI and data were reconstructed with and without DLR. Two radiologists independently evaluated three images per patient using a 5-point Likert scale. Objective evaluations, including signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and apparent diffusion coefficient (ADC), were conducted using manual region of interest-based analysis. The subjective and objective evaluations were compared using the Friedman test. The scores for the overall image quality, anatomical details, lesion conspicuity, artifacts, and distortion in FOCUS-DLR DWI were higher than in conventional DWI (all P < 0.001). The SNR of FOCUS-DLR DWI was higher than that of conventional and FOCUS DWI (both P < 0.001), while FOCUS and conventional DWI were similar (P = 0.096). Conventional, FOCUS, and FOCUS-DLR DWI had similar CNR and ADC values. Our findings indicate that images produced by FOCUS-DLR DWI were superior to conventional DWI, supporting the applicability of this technique in clinical practice. DLR provides a new approach to optimize breast DWI.

Enhancing Skin Cancer Diagnosis through the Integration of Deep Learning and Machine Learning Approaches

Skin cancer is a disease characterized by the uncontrolled proliferation of skin cells, typically manifesting as lesions or abnormal growths. Early diagnosis is critical for improving treatment outcomes. This study proposes an innovative approach to skin cancer diagnosis by integrating modern deep learning models with traditional machine learning algorithms. A three-phase methodology was developed. In the first phase, meaningful features were extracted from skin lesion images using various transfer learning models, including Xception, VGG16, ResNet152V2, InceptionV3, InceptionResNetV2, MobileNetV2, EfficientNetB2, and DenseNet201. In the second phase, dimensionality reduction was performed using Principal Component Analysis (PCA). In the final phase, the reduced feature sets were classified using K-Nearest Neighbors (KNN) and Random Forest (RF) algorithms. Experimental results demonstrated that the highest accuracy of 91.28% was achieved through the combination of DenseNet201 for feature extraction, PCA for dimensionality reduction, and Random Forest for classification. These findings highlight the effectiveness of integrating transfer learning models, dimensionality reduction techniques, and machine learning algorithms in enhancing the accuracy of skin cancer diagnosis.

Expert opinion elicitation for assisting deep learning based Lyme disease classifier with patient data

BackgroundDiagnosing erythema migrans (EM) skin lesion, the most common early symptom of Lyme disease, using deep learning techniques can be effective to prevent long-term complications. Existing works on deep learning based EM recognition only utilizes lesion image due to the lack of a dataset of Lyme disease related images with associated patient data. Doctors rely on patient information about the background of the skin lesion to confirm their diagnosis. To assist deep learning model with a probability score calculated from patient data, this study elicited opinions from fifteen expert doctors. To the best of our knowledge, this is the first expert elicitation work to calculate Lyme disease probability from patient data. MethodsFor the elicitation process, a questionnaire with questions and possible answers related to EM was prepared. Doctors provided relative weights to different answers to the questions. We converted doctors' evaluations to probability scores using Gaussian mixture based density estimation. We exploited formal concept analysis and decision tree for elicited model validation and explanation. We also proposed an algorithm for combining independent probability estimates from multiple modalities, such as merging the EM probability score from a deep learning image classifier with the elicited score from patient data. ResultsWe successfully elicited opinions from fifteen expert doctors to create a model for obtaining EM probability scores from patient data. ConclusionsThe elicited probability score and the proposed algorithm can be utilized to make image based deep learning Lyme disease pre-scanners robust. The proposed elicitation and validation process is easy for doctors to follow and can help address related medical diagnosis problems where it is challenging to collect patient data.

Patient Ability to Take Dermoscopic Follow-Up Images of Atypical Melanocytic Lesions With Smartphones: A Pilot Study

Introduction: Short-term teledermoscopic monitoring helps to distinguish early melanomas from nevi. As the incidence of melanoma is increasing, there are several benefits of patients taking their own dermoscopic images, but only a few previous studies have investigated the feasibility of this approach. Objectives: We sought to examine patient ability to take evaluable dermoscopic images of atypical melanocytic lesions in need of short-term monitoring. Methods: Patients were asked to take follow-up images in their homes using a borrowed dermoscope and their own smartphone. We investigated if the management decision differed when assessing follow-up images taken by patients compared to follow-up images taken by hospital staff. Lesions were rated as either changed, unchanged, or in need of further monitoring. In addition, image quality and patients’ attitudes towards taking dermoscopic follow-up images were studied. Results: Ninety-five patients with 132 lesions completed the study. Images taken by hospital staff were of better quality than images taken by patients (P&lt;0.001). A total of 24 dermoscopic images taken by patients (18.2%) were of poor quality and considered unsuitable for assessment at follow-up. In the remaining 108 lesions, the management decision was concordant in 95 cases (88.0%). Most patients found the procedure to be easy to perform and 76.0% of patients answered that they preferred self-photography. Conclusions: Self-photography for teledermoscopic evaluation of atypical melanocytic lesions is feasible, but it results in worse image quality, which may lead to discordant evaluations. Dermoscopes used for this purpose need to be more user-friendly and maintain a higher technical standard.

A Multi-label Artificial Intelligence Approach for Improving Breast Cancer Detection With Mammographic Image Analysis.

Breast cancer remains a major global health concern. This study aimed to develop a deep-learning-based artificial intelligence (AI) model that predicts the malignancy of mammographic lesions and reduces unnecessary biopsies in patients with breast cancer. In this retrospective study, we used deep-learning-based AI to predict whether lesions in mammographic images are malignant. The AI model learned the malignancy as well as margins and shapes of mass lesions through multi-label training, similar to the diagnostic process of a radiologist. We used the Curated Breast Imaging Subset of Digital Database for Screening Mammography. This dataset includes annotations for mass lesions, and we developed an algorithm to determine the exact location of the lesions for accurate classification. A multi-label classification approach enabled the model to recognize malignancy and lesion attributes. Our multi-label classification model, trained on both lesion shape and margin, demonstrated superior performance compared with models trained solely on malignancy. Gradient-weighted class activation mapping analysis revealed that by considering the margin and shape, the model assigned higher importance to border areas and analyzed pixels more uniformly when classifying malignant lesions. This approach improved diagnostic accuracy, particularly in challenging cases, such as American College of Radiology Breast Imaging-Reporting and Data System categories 3 and 4, where the breast density exceeded 50%. This study highlights the potential of AI in improving the diagnosis of breast cancer. By integrating advanced techniques and modern neural network designs, we developed an AI model with enhanced accuracy for mammographic image analysis.

Comparison of image quality and lesion conspicuity between conventional and deep learning reconstruction in gadoxetic acid-enhanced liver MRI

ObjectiveTo compare the image quality and lesion conspicuity of conventional vs deep learning (DL)-based reconstructed three-dimensional T1-weighted images in gadoxetic acid-enhanced liver magnetic resonance imaging (MRI).MethodsThis prospective study (NCT05182099) enrolled participants scheduled for gadoxetic acid-enhanced liver MRI due to suspected focal liver lesions (FLLs) who provided signed informed consent. A liver MRI was conducted using a 3-T scanner. T1-weighted images were reconstructed using both conventional and DL-based (AIRTM Recon DL 3D) reconstruction algorithms. Three radiologists independently reviewed the image quality and lesion conspicuity on a 5-point scale.ResultsFifty participants (male = 36, mean age 62 ± 11 years) were included for image analysis. The DL-based reconstruction showed significantly higher image quality than conventional images in all phases (3.71–4.40 vs 3.37–3.99, p < 0.001 for all), as well as significantly less noise and ringing artifacts than conventional images (p < 0.05 for all), while also showing significantly altered image texture (p < 0.001 for all). Lesion conspicuity was significantly higher in DL-reconstructed images than in conventional images in the arterial phase (2.15 [95% confidence interval: 1.78, 2.52] vs 2.03 [1.65, 2.40], p = 0.036), but no significant difference was observed in the portal venous phase and hepatobiliary phase (p > 0.05 for all). There was no significant difference in the figure-of-merit (0.728 in DL vs 0.709 in conventional image, p = 0.474).ConclusionDL reconstruction provided higher-quality three-dimensional T1-weighted imaging than conventional reconstruction in gadoxetic acid-enhanced liver MRI.Critical relevance statementDL reconstruction of 3D T1-weighted images improves image quality and arterial phase lesion conspicuity in gadoxetic acid-enhanced liver MRI compared to conventional reconstruction.Key PointsDL reconstruction is feasible for 3D T1-weighted images across different spatial resolutions and phases.DL reconstruction showed superior image quality with reduced noise and ringing artifacts.Hepatic anatomic structures were more conspicuous on DL-reconstructed images.Graphical

Plaque phenotyping using deep learning in optical coherence tomography imaging

Abstract Background Artificial intelligence (AI) techniques have emerged as powerful tools in various computer vision tasks, particularly in classification, detection, and segmentation. Purpose This study aimed to develop a deep learning model for categorizing optical coherence tomography (OCT) frames into different plaque phenotypes, including thin-cap fibroatheroma (TCFA) and fibroatheroma. Methods A prospective multicenter observational study enrolled 168 patients who underwent percutaneous coronary intervention (PCI) and received OCT imaging for non-target lesions with angiographically estimated diameter stenosis greater than 30% (177 vessels). Among them, 129 patients (135 vessels) underwent follow-up OCT imaging at 1 year. A total of 45,072 OCT frames were randomly divided into training and validation sets in a 4:1 ratio. An EfficientNet-B3 convolutional neural network model was developed to classify OCT frames into TCFA, thick-cap fibroatheroma, non-fibroatheroma, adaptive intimal thickening, and normal vessel categories. Results A confusion matrix comparing manually assessed plaque phenotypes with AI-predicted phenotypes was presented in Table. In the validation datasets (9,015 frames), the overall accuracy of the model was 85.1%. For predicting the presence of TCFA, sensitivity was 81.1%, specificity was 98.8%, positive predictive value was 63.9%, and negative predictive value was 99.5%. Regarding the presence of fibroatheroma, sensitivity was 90.3%, specificity was 89.4%, positive predictive value was 93.1%, and negative predictive value was 85.4%. Conclusions The deep learning algorithm demonstrated high accuracy in detecting TCFA and fibroatheroma with excellent reproducibility. This automated process enables operators to promptly identify high-risk lesions during PCI, potentially leading to improved clinical outcomes.

Neuronal Intranuclear Inclusion Disease Presenting with Acute-Onset Dementia and Cortical Edema: A Case Report.

Neuronal Intranuclear Inclusion Disease (NIID) is a neurodegenerative disorder characterized by the formation of eosinophilic inclusions in the neurons, visceral and skin cells. The cause is associated with the GGC nucleotide repeat expansion in the NOTCH2NLC gene. The imaging hallmark of NIID is hyperintensities on diffusion-weighted imaging (DWI) at the corticomedullary junction. Clinical manifestations of NIID are highly heterogeneous. Here, we report a case of NIID presenting with acute-onset dementia and cortical edema. We describe an elderly male patient who presented with sudden dementia within a day. Considering the abrupt onset and the stroke history, we initially diagnosed vascular disease. However, further imaging revealed cortical edema in the temporo-parieto-occipital lobes. Blood and cerebrospinal fluid tests ruled out immunological, metabolic, infectious, or neoplastic etiologies. Genetic testing ultimately confirmed the diagnosis of NIID. Intravenous immunoglobulin (IVIG) therapy did not improve the patient's symptoms; However, about 1 month after treatment, spontaneous improvement was observed. It is noteworthy that 22 months before the onset of cognitive impairment, the patient's MRI for headaches already exhibited the typical imaging lesions of this disease in the cerebellum paravermal region. Patients with encephalopathy syndrome exhibiting imaging features resembling mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS) syndrome or Creutzfeldt-Jakob disease should consider the NIID as differential diagnosis. Chronic headaches and symmetric lesions in the cerebellar paravermal region on MRI may be noteworthy indicators of NIID during non-episodic phases.

Anomaly detection scheme for lung CT images using vector quantized variational auto-encoder with support vector data description.

This study aims to develop an anomaly-detection scheme for lesions in CT images. Our database consists of lung CT images obtained from 1500 examinees. It includes 1200 normal and 300 abnormal cases. In this study, SVDD (Support Vector Data Description) mapping the normal latent variables into a hypersphere as small as possible on the latent space is introduced to VQ-VAE (Vector Quantized-Variational Auto-Encoder). VQ-VAE with SVDD is constructed from two encoders, two decoders, and an embedding space. The first encoder compresses the input image into the latent-variable map, whereas the second encoder maps the normal latent variables into a hypersphere as small as possible. The first decoder then up-samples the mapped latent variables into a latent-variable map with the original size. The second decoder finally reconstructs the input image from the latent-variable map replaced by the embedding representations. The data of each examinee is classified as abnormal or normal based on the anomaly score defined as the combination of the difference between the input image and the reconstructed image and the distance between the latent variables and the center of the hypersphere. The area under the ROC curve for VQ-VAE with SVDD was 0.76, showing an improvement when compared with the conventional VAE (0.63, p < .001). VQ-VAE with SVDD developed in this study can yield higher anomaly-detection accuracy than the conventional VAE. The proposed method is expected to be useful for identifying examinees with lesions and reducing interpretation time in CT screening.

DWI lesions after intracranial aneurysm treatment with contour or WEB-does the device matter?

In this study, we report our experience with the occurrence of diffusion-weighted imaging (DWI) lesions following aneurysm treatment with Woven EndoBridge (WEB) device and Contour Neurovascular System in elective settings. We compared both techniques in a retrospective single-center analysis, to investigate whether there are significant differences in the incidence of microemboli depending on the device used. Sixty-two cases treated with Contour were retrospectively compared with 84 WEB cases. Postinterventional MRI was performed in all patients within five days after the angiographic procedure. Only DWI lesions in the dependent vascular territory of the treated aneurysm were considered. Data on occurrence of DWI lesions, device implantation times as well as device change or repositions maneuvers were compared. A total of 146 patients were included in this study. Postinterventional DWI lesions were detected in 33 of the 62 (53.2%) Contour cases and in 36 of the 84 (42.8%) WEB cases with an average lesion number of 1.70 ± 3.45 in our WEB group and 1.68 ± 2.53 in our Contour cohort. Neither periinterventional device changes nor device reposition maneuvers had a significant impact on the occurrence of DWI lesions. Although four patients experienced transient neurological deterioration, the DWI lesions were not clinically relevant at the time of discharge, as the patients' NIHSS remained stable compared to the preintervention NIHSS in the entire cohort. There was no statistically significant difference between the implantation times of WEB and Contour, although Contour could be implanted slightly faster. Aneurysm treatment with WEB or Contour results in a comparable numbers of procedure-related DWI lesions. This effect was consistent after adjustment for other demographic or technical variables.

Small-scale cross-layer fusion network for classification of diabetic retinopathy

Deep learning-based automatic classification of diabetic retinopathy (DR) helps to enhance the accuracy and efficiency of auxiliary diagnosis. This paper presents an improved residual network model for classifying DR into five different severity levels. First, the convolution in the first layer of the residual network was replaced with three smaller convolutions to reduce the computational load of the network. Second, to address the issue of inaccurate classification due to minimal differences between different severity levels, a mixed attention mechanism was introduced to make the model focus more on the crucial features of the lesions. Finally, to better extract the morphological features of the lesions in DR images, cross-layer fusion convolutions were used instead of the conventional residual structure. To validate the effectiveness of the improved model, it was applied to the Kaggle Blindness Detection competition dataset APTOS2019. The experimental results demonstrated that the proposed model achieved a classification accuracy of 97.75% and a Kappa value of 0.971 7 for the five DR severity levels. Compared to some existing models, this approach shows significant advantages in classification accuracy and performance.