Computer-aided Diagnostic Tool Research Articles

Cone-beam CT landmark detection for measuring basal bone width: a retrospective validation study

BackgroundAccurate assessment of basal bone width is essential for distinguishing individuals with normal occlusion from patients with maxillary transverse deficiency who may require maxillary expansion. Herein, we evaluated the effectiveness of a deep learning (DL) model in measuring landmarks of basal bone width and assessed the consistency of automated measurements compared to manual measurements.MethodsBased on the U-Net algorithm, a coarse-to-fine DL model was developed and trained using 80 cone-beam computed tomography (CBCT) images. The model’s prediction capabilities were validated on 10 CBCT scans and tested on an additional 34. To evaluate the performance of the DL model, its measurements were compared with those taken manually by one junior orthodontist using the concordance correlation coefficient (CCC).ResultsIt took approximately 1.5 s for the DL model to perform the measurement task in only CBCT images. This framework showed a mean radial error of 1.22 ± 1.93 mm and achieved successful detection rates of 71.34%, 81.37%, 86.77%, and 91.18% in the 2.0-, 2.5-, 3.0-, and 4.0-mm ranges, respectively. The CCCs (95% confidence interval) of the maxillary basal bone width and mandibular basal bone width distance between the DL model and manual measurement for the 34 cases were 0.96 (0.94–0.97) and 0.98 (0.97–0.99), respectively.ConclusionThe novel DL framework developed in this study improved the diagnostic accuracy of the individual assessment of maxillary width. These results emphasize the potential applicability of this framework as a computer-aided diagnostic tool in orthodontic practice.

Skin cancer classification leveraging multi-directional compact convolutional neural network ensembles and gabor wavelets

Skin cancer (SC) is an important medical condition that necessitates prompt identification to ensure timely treatment. Although visual evaluation by dermatologists is considered the most reliable method, its efficacy is subjective and laborious. Deep learning-based computer-aided diagnostic (CAD) platforms have become valuable tools for supporting dermatologists. Nevertheless, current CAD tools frequently depend on Convolutional Neural Networks (CNNs) with huge amounts of deep layers and hyperparameters, single CNN model methodologies, large feature space, and exclusively utilise spatial image information, which restricts their effectiveness. This study presents SCaLiNG, an innovative CAD tool specifically developed to address and surpass these constraints. SCaLiNG leverages a collection of three compact CNNs and Gabor Wavelets (GW) to acquire a comprehensive feature vector consisting of spatial–textural–frequency attributes. SCaLiNG gathers a wide range of image details by breaking down these photos into multiple directional sub-bands using GW, and then learning several CNNs using those sub-bands and the original picture. SCaLiNG also combines attributes taken from various CNNs trained with the actual images and subbands derived from GW. This fusion process correspondingly improves diagnostic accuracy due to the thorough representation of attributes. Furthermore, SCaLiNG applies a feature selection approach which further enhances the model’s performance by choosing the most distinguishing features. Experimental findings indicate that SCaLiNG maintains a classification accuracy of 0.9170 in categorising SC subcategories, surpassing conventional single-CNN models. The outstanding performance of SCaLiNG underlines its ability to aid dermatologists in swiftly and precisely recognising and classifying SC, thereby enhancing patient outcomes.

Gastric Cancer Detection with Ensemble Learning on Digital Pathology: Use Case of Gastric Cancer on GasHisSDB Dataset.

Gastric cancer has become a serious worldwide health concern, emphasizing the crucial importance of early diagnosis measures to improve patient outcomes. While traditional histological image analysis is regarded as the clinical gold standard, it is labour intensive and manual. In recognition of this problem, there has been a rise in interest in the use of computer-aided diagnostic tools to help pathologists with their diagnostic efforts. In particular, deep learning (DL) has emerged as a promising solution in this sector. However, current DL models are still restricted in their ability to extract extensive visual characteristics for correct categorization. To address this limitation, this study proposes the use of ensemble models, which incorporate the capabilities of several deep-learning architectures and use aggregate knowledge of many models to improve classification performance, allowing for more accurate and efficient gastric cancer detection. To determine how well these proposed models performed, this study compared them with other works, all of which were based on the Gastric Histopathology Sub-Size Images Database, a publicly available dataset for gastric cancer. This research demonstrates that the ensemble models achieved a high detection accuracy across all sub-databases, with an average accuracy exceeding 99%. Specifically, ResNet50, VGGNet, and ResNet34 performed better than EfficientNet and VitNet. For the 80 × 80-pixel sub-database, ResNet34 exhibited an accuracy of approximately 93%, VGGNet achieved 94%, and the ensemble model excelled with 99%. In the 120 × 120-pixel sub-database, the ensemble model showed 99% accuracy, VGGNet 97%, and ResNet50 approximately 97%. For the 160 × 160-pixel sub-database, the ensemble model again achieved 99% accuracy, VGGNet 98%, ResNet50 98%, and EfficientNet 92%, highlighting the ensemble model's superior performance across all resolutions. Overall, the ensemble model consistently provided an accuracy of 99% across the three sub-pixel categories. These findings show that ensemble models may successfully detect critical characteristics from smaller patches and achieve high performance. The findings will help pathologists diagnose gastric cancer using histopathological images, leading to earlier identification and higher patient survival rates.

Colon Cancer Disease Diagnosis Based on Convolutional Neural Network and Fishier Mantis Optimizer.

Colon cancer is a prevalent and potentially fatal disease that demands early and accurate diagnosis for effective treatment. Traditional diagnostic approaches for colon cancer often face limitations in accuracy and efficiency, leading to challenges in early detection and treatment. In response to these challenges, this paper introduces an innovative method that leverages artificial intelligence, specifically convolutional neural network (CNN) and Fishier Mantis Optimizer, for the automated detection of colon cancer. The utilization of deep learning techniques, specifically CNN, enables the extraction of intricate features from medical imaging data, providing a robust and efficient diagnostic model. Additionally, the Fishier Mantis Optimizer, a bio-inspired optimization algorithm inspired by the hunting behavior of the mantis shrimp, is employed to fine-tune the parameters of the CNN, enhancing its convergence speed and performance. This hybrid approach aims to address the limitations of traditional diagnostic methods by leveraging the strengths of both deep learning and nature-inspired optimization to enhance the accuracy and effectiveness of colon cancer diagnosis. The proposed method was evaluated on a comprehensive dataset comprising colon cancer images, and the results demonstrate its superiority over traditional diagnostic approaches. The CNN-Fishier Mantis Optimizer model exhibited high sensitivity, specificity, and overall accuracy in distinguishing between cancer and non-cancer colon tissues. The integration of bio-inspired optimization algorithms with deep learning techniques not only contributes to the advancement of computer-aided diagnostic tools for colon cancer but also holds promise for enhancing the early detection and diagnosis of this disease, thereby facilitating timely intervention and improved patient prognosis. Various CNN designs, such as GoogLeNet and ResNet-50, were employed to capture features associated with colon diseases. However, inaccuracies were introduced in both feature extraction and data classification due to the abundance of features. To address this issue, feature reduction techniques were implemented using Fishier Mantis Optimizer algorithms, outperforming alternative methods such as Genetic Algorithms and simulated annealing. Encouraging results were obtained in the evaluation of diverse metrics, including sensitivity, specificity, accuracy, and F1-Score, which were found to be 94.87%, 96.19%, 97.65%, and 96.76%, respectively.

Bridging the Clinical-Computational Transparency Gap in Digital Pathology.

Computational pathology combines clinical pathology with computational analysis, aiming to enhance diagnostic capabilities and improve clinical productivity. However, communication barriers between pathologists and developers often hinder the full realization of this potential. To propose a standardized framework that improves mutual understanding of clinical objectives and computational methodologies. The goal is to enhance the development and application of computer-aided diagnostic (CAD) tools. The article suggests pivotal roles for pathologists and computer scientists in the CAD development process. It calls for increased understanding of computational terminologies, processes, and limitations among pathologists. Similarly, it argues that computer scientists should better comprehend the true use cases of the developed algorithms to avoid clinically meaningless metrics. CAD tools improve pathology practice significantly. Some tools have even received US Food and Drug Administration approval. However, improved understanding of machine learning models among pathologists is essential to prevent misuse and misinterpretation. There is also a need for a more accurate representation of the algorithms' performance compared to that of pathologists. A comprehensive understanding of computational and clinical paradigms is crucial for overcoming the translational gap in computational pathology. This mutual comprehension will improve patient care through more accurate and efficient disease diagnosis.

DPML: Prior-guided multitask learning for dental object recognition on limited panoramic radiograph dataset

Addressing the growing need for advanced computer-aided diagnostic tools in dentistry, this study presents DPML (Dental Prior-guided Multitask Learning), a pioneering framework tailored for the analysis of dental panoramic radiographs in the face of data scarcity. DPML introduces a novel multitask learning strategy that effectively mitigates the limitations of sparse, well-annotated dental datasets. The methodology integrates four advanced techniques: targeted data augmentation for enhanced image diversity, segmentation structures for improved edge definition, polar-coordinate methods for precise tooth contour extraction, and a spatially informed classification strategy for accurate tooth type identification. Our comparative analysis demonstrates that DPML markedly surpasses traditional approaches in tooth segmentation, detection, and classification. Remarkably, it achieves a 95.06% mean Average Precision (mAP) and a 97.70% precision rate in dental object recognition tasks on a small-size dataset.

Application of Data Mining Techniques in Biopsy Interpretation and Staging of Carcinoma Cancer Disease: A Case Study of Northeastern Nigeria

This paper introduces an innovative framework tailored for carcinoma cancer staging in northeastern Nigeria, employing data mining techniques and machine learning algorithms, with analysis conducted using the WEKA toolkit. Given the challenges in healthcare systems, especially in cancer diagnosis and treatment, this study aims to enhance diagnostic precision thorough analysis of biopsy reports. By assessing Support Vector Machine (SVM) and decision tree algorithms using datasets from six tertiary health institutions, the framework demonstrates promising outcomes in accurately determining cancer malignancy stage. SVM exhibits high accuracies in staging breast and prostate cancer, while decision tree algorithms show notable accuracy rates in staging skin cancer. These results highlight the potential of machine learning in advancing cancer diagnosis and treatment planning, particularly in resource-constrained settings, and pave the way for the development of computer-aided diagnostic tools tailored for carcinoma cancer staging in northeastern Nigeria.

Abstract PO3-07-03: Development & Validation of an AI-supported Workflow for Clinical Scoring of HER2, ER, PR & Ki67 Immunohistochemistry in Breast Cancer Tissue

Abstract Background: Immunohistochemical evaluation of HER2 status, the hormone receptors ER and PR, and the proliferation marker Ki67 forms part of the routine clinical diagnostic pathway for invasive breast carcinomas, and is the cornerstone of treatment stratification, informing both prognosis and patient management. Pathologist scoring of immunohistochemistry (IHC) at the microscope is time-consuming and prone to significant inter- and intra-observer variability. We developed HALO Breast AI, a decision-support system designed to improve efficiency and diagnostic accuracy through automating whole slide image (WSI) scoring. Here, we present preliminary results of a validation study of HALO Breast AI. Methods: HALO Breast AI was developed with routine diagnostic cases sourced from three institutes. The algorithm was trained using 107,755 pathologist-reviewed annotations to identify and threshold DAB-positive tumor cells within automatically segmented tumor regions. Internal validation was conducted on 80 unseen WSI, using 60,012 pathologist-reviewed annotations to assess analytical performance. Comparison of the algorithm scores to the mode of 3 expert pathologists (where at least 2 out of 3 agreed) was used to assess consensus agreement. Clinical performance and generalizability were assessed by comparing the algorithm scores to clinical data from two independent external institutes across 200 unseen WSI (n=50 per marker) from institute one and 300 unseen WSI (n=100 per marker [ ER, PR & HER2, only]) from institute 2. Results: The median image F1-score for tumor classification was 0.91, while the median image F1-score for cell level validation was 0.96. Internal validation showed agreement between HALO Breast AI and the mode of 3 expert pathologists of 100% for ER, 90% for PR, 95% for Ki67 and 90% for HER2. High concordance was measured between the algorithm and the pathologists’ scores, with Light’s kappa of 0.91 for ER, 0.85 for PR, 0.79 for HER2 and 0.82 for Ki67. Performance on WSI obtained from external institute one, showed agreement between the clinical score obtained from HALO Breast AI and the clinical data of 96% for ER, 94% for PR, 84% for HER2 and 84% for Ki67. External institute two showed agreement of 90% for ER, 90% for PR and 83% for HER2. Of the ER & PR cases that were in disagreement at the 1% clinical cut-off, the algorithm percent-positive scores were within a 1-3% range in 5 out of the 5 from institute 1 and 14 out of the 20 from institute 2. So, although on a categorical scale the AI assigned category disagreed with the clinical category, the results were close on a continuous scale. Conclusions: HALO Breast AI accurately detects tumor regions and tumor cells within breast cancer tissue and demonstrates high clinical agreement when scoring routine diagnostic IHC. Additionally, HALO Breast AI shows good generalisability, with a consistently strong performance across inherent variability that exists between external, independent data sets. Computer-aided diagnostic tools such as HALO Breast AI have the potential to support pathologists in the diagnostic setting by improving workflow efficiency and standardising results. Citation Format: Meredith Lodge, Alastair Ironside, Ashley Graham, Antonio Polonia, Stefan Reinhard, Wiebke Solass, Inti Zlobec, Peter Caie. Development & Validation of an AI-supported Workflow for Clinical Scoring of HER2, ER, PR & Ki67 Immunohistochemistry in Breast Cancer Tissue [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-07-03.

Toward explainable AI in radiology: Ensemble-CAM for effective thoracic disease localization in chest X-ray images using weak supervised learning.

Chest X-ray (CXR) imaging is widely employed by radiologists to diagnose thoracic diseases. Recently, many deep learning techniques have been proposed as computer-aided diagnostic (CAD) tools to assist radiologists in minimizing the risk of incorrect diagnosis. From an application perspective, these models have exhibited two major challenges: (1) They require large volumes of annotated data at the training stage and (2) They lack explainable factors to justify their outcomes at the prediction stage. In the present study, we developed a class activation mapping (CAM)-based ensemble model, called Ensemble-CAM, to address both of these challenges via weakly supervised learning by employing explainable AI (XAI) functions. Ensemble-CAM utilizes class labels to predict the location of disease in association with interpretable features. The proposed work leverages ensemble and transfer learning with class activation functions to achieve three objectives: (1) minimizing the dependency on strongly annotated data when locating thoracic diseases, (2) enhancing confidence in predicted outcomes by visualizing their interpretable features, and (3) optimizing cumulative performance via fusion functions. Ensemble-CAM was trained on three CXR image datasets and evaluated through qualitative and quantitative measures via heatmaps and Jaccard indices. The results reflect the enhanced performance and reliability in comparison to existing standalone and ensembled models.

Detection of dental periapical lesions using retinex based image enhancement and lightweight deep learning model

Dental periapical lesions, commonly associated with inflammation around the tooth apex, pose a significant challenge in early diagnosis and treatment. This study introduces a novel approach for the detection of dental periapical lesions through the integration of Retinex-based image enhancement techniques and a lightweight deep learning model. The Retinex algorithm is employed to enhance the radiographic images, addressing issues related to inconsistent illumination and contrast. Subsequently, a tailored lightweight deep learning model is designed to efficiently extract relevant features from the enhanced images. Present methodology leverages a dataset of dental radiographs to train and evaluate the deep learning model, incorporating a diverse range of periapical lesion cases. The model is optimized for computational efficiency while maintaining high accuracy, making it suitable for deployment in resource-constrained environment. To enhance precision in lesion detection, the U-Net segmentation technique has been incorporated, providing a sophisticated approach to delineate and analyze specific areas of interest within the radiographic images. This addition further refines our diagnostic framework, contributing to the robustness of lesion identification. Experimental results demonstrate the effectiveness of the Retinex-based image enhancement in improving the visibility of periapical lesions. The lightweight deep learning model exhibits promising performance in accurately detecting and classifying dental periapical lesions, showcasing its potential for early and efficient diagnosis. The results obtained from the present model are compared with those from Convolutional Neural Network (CNN) as well as with diagnosis of expert practitioners and the model is observed to perform very well. The study contributes to the advancement of computer-aided diagnostic tools in dentistry, offering a scalable and accessible solution for the identification of dental periapical lesions through the fusion of image enhancement and lightweight deep learning techniques.

Improving cervical cancer classification in PAP smear images with enhanced segmentation and deep progressive learning-based techniques.

Cervical cancer, a prevalent and deadly disease among women, comes second only to breast cancer, with over 700 daily deaths. The Pap smear test is a widely utilized screening method for detecting cervical cancer in its early stages. However, this manual screening process is prone to a high rate of false-positive outcomes because of human errors. Researchers are using machine learning and deep learning in computer-aided diagnostic tools to address this issue. These tools automatically analyze and sort cervical cytology and colposcopy images, improving the precision of identifying various stages of cervical cancer. This article uses state-of-the-art deep learning methods, such as ResNet-50 for categorizing cervical cancer cells to assist medical professionals. The method includes three key steps: preprocessing, segmentation using k-means clustering, and classifying cancer cells. The model is assessed based on performance metrics viz; precision, accuracy, kappa score, precision, sensitivity, and specificity. In the end, the high success rate shows that the ResNet50 model is a valuable tool for timely detection of cervical cancer. In conclusion, the infected cervical region is pinpointed using spatial K-means clustering and preprocessing operations. This sequence of actions is followed by a progressive learning technique. The Progressive Learning technique then proceeded through several stages: Stage 1 with 64 × 64 images, Stage 2 with 224 × 224 images, Stage 3 with 512 × 512 images, and the final Stage 4 with 1024 × 1024 images. The outcomes show that the suggested model is effective for analyzing Pap smear tests, achieving 97.4% accuracy and approx. 98% kappa score.

Automated Detection of Rheumatic Heart Disease via Unsegmented Heart Sound Analysis

Rheumatic Heart Disease (RHD) remains a significant cause of morbidity and mortality in many parts of the world, particularly in low-resource settings. Early detection and intervention are critical to mitigating its progression and reducing associated complications. This paper presents an automated approach for the detection of RHD utilizing unsegmented heart sound analysis through deep learning techniques. By leveraging Spectro-temporal representations of raw heart sound signals, our proposed method aims to capture subtle yet discriminative patterns indicative of RHD pathology. The deep learning model is trained on a dataset comprising a diverse range of heart sounds, including those from both RHD-positive and RHD-negative individuals. Through rigorous evaluation of unseen data, our approach demonstrates promising performance in accurately distinguishing between RHD and healthy heart sounds. Furthermore, the proposed methodology offers the potential for scalable and cost-effective screening in resource-limited settings, thereby facilitating early identification and management of RHD cases. This work contributes to the advancement of computer-aided diagnostic tools for cardiovascular diseases, with implications for improving healthcare outcomes and reducing the burden of RHD globally. Key Words: Rheumatic Heart Disease, Heart Sound Analysis, Deep Learning, Automated Detection, Unsegmented Signals, Spectro-Temporal Representation, Cardiovascular Disease.

Evolutionary RNN framework for Precise Lung Nodule Detection from CT Scans

Radiologists find it challenging and time-consuming to recognize and evaluate nodules of lung using CT scans that are malignant. Because of this, early lung growth prediction is necessary for the inquiry technique, which raises the likelihood that the treatment will be successful. Computer-aided diagnostic (CAD) tools have been used to help with this issue. The primary goal of the work is to identify if the nodules are cancerous or not and to deliver more accurate results. The RNN [Recurrent] which is a type of neural network model that includes a feedback loop. In this paper, evolutionary algorithms are examined using the MATLAB Tool, including the Grey Wolf Optimization Algorithm and Recurrent Neural Network (RNN) Techniques. Additionally, statistical characteristics are generated and examined in comparison to other RNNs using Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) combinations. Comparing the suggested approach to other state-of-the-art techniques, it yielded results with extremely high accuracy, sensitivity, specificity, and precision. In the past few years, there has been a substantial increase for evolutionary algorithms in the field of feature selection due to their simplicity and potential for global search capabilities. The suggested solutions have outperformed classical approaches employed across various fields, showing excellent results. Determining whether lung nodules will become malignant or not will be made easier with early identification.

Quantification of Proteus syndrome-associated lung disease

BackgroundProteus syndrome is an ultra-rare mosaic overgrowth disorder. Individuals with Proteus syndrome can develop emphysematous and cystic changes of the lung that may lead to progressive respiratory symptoms and require surgical intervention. This retrospective study seeks to quantify the radiographic features of Proteus syndrome-associated lung disease using computed tomography (CT) of the chest. The first method derives a Cystic Lung Score (CLS) by using a computer-aided diagnostic tool to quantify the fraction of cystic involvement of the lung. The second method yields a Clinician Visual Score (CVS), an observer reported scale of severity based on multiple radiographic features. The aim of this study was to determine if these measurements are associated with clinical symptoms, pulmonary function test (PFT) measurements, and if they may be used to assess progression of pulmonary disease.ResultsOne hundred and thirteen imaging studies from 44 individuals with Proteus syndrome were included. Dyspnea and oxygen use were each associated with higher CLS (p = 0.001 and < 0.001, respectively) and higher CVS (p < 0.001 and < 0.001). Decreases in percent predicted FVC, FEV1, and DLCO each correlated with increased CLS and CVS. The annual increase of CLS in children, 5.6, was significantly greater than in adults, 1.6. (p = 0.03). The annual increase in CVS in children, 0.4, was similar to adults, 0.2 (p = 0.36).ConclusionsProteus syndrome-associated lung disease is progressive. The rate of cystic progression is increased in children. Increased scores in CLS and CVS were associated with clinical symptoms and decreased pulmonary function. Both methods were able to detect change over time and were associated with clinically meaningful outcomes which may enable their use in interventional studies.

A Laplacian Pyramid Based Generative H&E Stain Augmentation Network.

Hematoxylin and Eosin (H&E) staining is a widely used sample preparation procedure for enhancing the saturation of tissue sections and the contrast between nuclei and cytoplasm in histology images for medical diagnostics. However, various factors, such as the differences in the reagents used, result in high variability in the colors of the stains actually recorded. This variability poses a challenge in achieving generalization for machine-learning based computer-aided diagnostic tools. To desensitize the learned models to stain variations, we propose the Generative Stain Augmentation Network (G-SAN) - a GAN-based framework that augments a collection of cell images with simulated yet realistic stain variations. At its core, G-SAN uses a novel and highly computationally efficient Laplacian Pyramid (LP) based generator architecture, that is capable of disentangling stain from cell morphology. Through the task of patch classification and nucleus segmentation, we show that using G-SAN-augmented training data provides on average 15.7% improvement in F1 score and 7.3% improvement in panoptic quality, respectively. Our code is available at https://github.com/lifangda01/GSAN-Demo.

A Review of Deep Learning Approaches for Early Diagnosis of Alzheimer's Disease

Alzheimer's disease (AD), one of the major neurodegenerative diseases, has become the most common cause of dementia problems. Up to now, there is a lack of effective targeted therapeutic drugs and effective treatment modalities to stop the progression of the disease. With the continuous development of computer technology, the use of computer-aided diagnostic technology tools for AD early classification studies will provide clinicians with important assistance. Deep learning-based Alzheimer's disease (AD) imaging classification has become a current research hotspot. In this paper, we first describe the commonly used publicly available datasets in the AD imaging classification task; then introduce the commonly used deep learning classification models for AD diagnosis; secondly, we compare the studies that target different biomarkers of the subjects and the use of unimodal or a combination of different modalities for the early classification of AD; and finally, The challenges of AD classification are summarized and future research directions are proposed.

Impact of Voxel Normalization on a Machine Learning-Based Method: A Study on Pulmonary Nodule Malignancy Diagnosis Using Low-Dose Computed Tomography (LDCT).

Lung cancer (LC) stands as the foremost cause of cancer-related fatality rates worldwide. Early diagnosis significantly enhances patient survival rate. Nowadays, low-dose computed tomography (LDCT) is widely employed on the chest as a tool for large-scale lung cancer screening. Nonetheless, a large amount of chest radiographs creates an onerous burden for radiologists. Some computer-aided diagnostic (CAD) tools can provide insight to the use of medical images for diagnosis and can augment diagnostic speed. However, due to the variation in the parameter settings across different patients, substantial discrepancies in image voxels persist. We found that different voxel sizes can create a compromise between model generalization and diagnostic efficacy. This study investigates the performance disparities of diagnostic models trained on original images and LDCT images reconstructed to different voxel sizes while making isotropic. We examined the ability of our method to differentiate between benign and malignant nodules. Using 11 features, a support vector machine (SVM) was trained on LDCT images using an isotropic voxel with a side length of 1.5 mm for 225 patients in-house. The result yields a favorable model performance with an accuracy of 0.9596 and an area under the receiver operating characteristic curve (ROC/AUC) of 0.9855. In addition, to furnish CAD tools for clinical application, future research including LDCT images from multi-centers is encouraged.

One-Stop Automated Diagnostic System for Carpal Tunnel Syndrome in Ultrasound Images Using Deep Learning

Ultrasound (US) examination has unique advantages in diagnosing carpal tunnel syndrome (CTS), although identification of the median nerve (MN) and diagnosis of CTS depend heavily on the expertise of examiners. In the aim of alleviating this problem, we developed a one-stop automated CTS diagnosis system (OSA-CTSD) and evaluated its effectiveness as a computer-aided diagnostic tool. We combined real-time MN delineation, accurate biometric measurements and explainable CTS diagnosis into a unified framework, called OSA-CTSD. We then collected a total of 32,301 static images from US videos of 90 normal wrists and 40 CTS wrists for evaluation using a simplified scanning protocol. The proposed model exhibited better segmentation and measurement performance than competing methods, with a Hausdorff distance (95th percentile) score of 7.21 px, average symmetric surface distance score of 2.64 px, Dice score of 85.78% and intersection over union score of 76.00%. In the reader study, it exhibited performance comparable to the average performance of experienced radiologists in classifying CTS and outperformed inexperienced radiologists in terms of classification metrics (e.g., accuracy score 3.59% higher and F1 score 5.85% higher). Diagnostic performance of the OSA-CTSD was promising, with the advantages of real-time delineation, automation and clinical interpretability. The application of such a tool not only reduces reliance on the expertise of examiners but also can help to promote future standardization of the CTS diagnostic process, benefiting both patients and radiologists.

A novel deep learning framework for the identification of tortuous vessels in plus diseased infant retinal images

Retinopathy of prematurity (ROP), sometimes known as Terry syndrome, is an ophthalmic condition that affects premature babies. It is the main cause of childhood blindness and morbidity of vision throughout life. ROP frequently coexists with a disease stage known as Plus disease, which is marked by severe tortuosity and dilated retinal blood vessels. The goal of this research is to create a diagnostic technique that can discriminate between infants with Plus disease from healthy subjects. Blood vascular tortuosity is used as a prognostic indicator for the diagnosis. We examine the quantification of retinal blood vessel tortuosity and propose a computer-aided diagnosis system that can be used as a tool for ROP identification. Deep neural networks are used in the proposed approach to segment retinal blood vessels, which is followed by the prediction of tortuous vessel pixels in the segmented vessel map. Digital fundus images obtained from Retcam3TM is used for screening. We use a proprietary data set of 289 infant retinal images (89 with Plus disease and 200 healthy) from Narayana Nethralaya in Bangalore, India, to illustrate the efficacy of our methodology. The findings of this study demonstrate the reliability of the proposed method as a computer-aided diagnostic tool that can help medical professionals make an early diagnosis of ROP.

Automated Recognition of Congenital Heart Defects (CHD) from Ultrasonic Fetal Heart Biometry

The growth and development of the fetus is monitored widely using Ultrasound imaging modality and has become synonymous with prenatal scanning. Analysing the images helps in the detection of congenital defects at an early stage which help to plan the delivery appropriately or for taking some corrective measures. The inherent noise of ultrasound imaging makes analysis of the images largely dependent on the expertise of the clinician and is prone to errors. Computer Aided Diagnostic (CAD) tools used in medical diagnosis assist the clinicians in analysing the images and reducing the chances of missing the defect. AI tools, machine learning and deep learning algorithms have shown improved classification performance of medical images in recent times have resulting in widespread implementation of these technologies in CAD tools. The proposed work discusses the use of machine learning for detection of Congenital heart disease (CHD). Frames resulting from the B mode Ultrasound cineloop sequence have been used for the study. A total of 870 frames with 550 normal frames and 320 abnormal frames are used for the study. Law’s texture feature have been extracted from the pre processed frames which in turn is used for classification. The performance of each of these features is compared using different kernels of SVM classifier. Tuning parameters regularization and gamma have been used to optimise the classification performance using grid search algorithm. It is observed that Laws texture feature results in better classification performance with an average accuracy and precision of 95% and specificity, sensitivity of 98%. Precision Recall curves and AUC-ROC plots have been used for comparing the performance of different kernels.