Computer-aided Detection System Research Articles

ERetinaNet: An Efficient Neural Network Based on RetinaNet for Mammographic Breast Mass Detection.

Mammography is an effective method for diagnosing breast diseases, and computer-aided detection (CAD) systems play an important role in the detection of breast masses. However, low contrast and the interference of surrounding tissues make the detection of masses challenging. In this paper, an efficient RetinaNet network named ERetinaNet is proposed to improve the accuracy and inference speed of mammographic breast mass detection. Efficient modules are designed and introduced into the network to facilitate the extraction of comprehensive features, while the structure of the network is simplified to improve the inference speed. A Faster RepVGG (FRepVGG) architecture is first proposed as the backbone network that utilizes three effective strategies: 1) The multi-branch structure used during training enhances learning, and it is equivalently converted to a single-path structure during inference by re-parameterization technique to accelerate the detection speed. 2) The Extraction operation is proposed to condense the features of intermediate layers. 3) An effective Multi-spectral Channel Attention (eMCA) module is added in the last layer of each stage, enabling the network to pay more attention to the target region. In addition, Vision Transformer (ViT) is added to ERetinaNet, which enables ERetinaNet to learn global semantic information. The detection head is simplified to make ERetinaNet more efficient. The experimental results show that compared with the original RetinaNet, ERetinaNet improves the mean Average Precision (mAP) from 79.16% to 85.01% and significantly shortens the inference time. Moreover, the detection accuracy of ERetinaNet outperforms other excellent object detection networks, such as Faster R-CNN, SSD, YOLOv3 and YOLOv7.

The use of a real-time computer-aided detection system for visible lesions in the Barrett’s esophagus during live endoscopic procedures: a pilot study (with video)

This pilot study evaluated the performance of a recently developed computer-aided detection (CADe) system for Barrett's neoplasia during live endoscopic procedures. Fifteen patients with a visible lesion and 15 without were included in this study. A CAD-assisted workflow was used that included a slow pullback video recording of the entire Barrett's segment with live CADe assistance, followed by CADe-assisted level-based video recordings every 2 cm of the Barrett's segment. Outcomes were per-patient and per-level diagnostic accuracy of the CAD-assisted workflow, in which the primary outcome was per-patient invivo CADe sensitivity. In the per-patient analyses, the CADe system detected all visible lesions (sensitivity 100%). Per-patient CADe specificity was 53%. Per-level sensitivity and specificity of the CADe-assisted workflow were 100% and 73%, respectively. In this pilot study, detection by the CADe system of all potentially neoplastic lesions in Barrett's esophagus was comparable to that of an expert endoscopist. Continued refinement of the system may improve specificity. External validation in larger multicenter studies is planned. (Clinical trial registration number: NCT05628441.).

Assessing the Accuracy of a Computer-Aided Detection System for Suspected Malignant Breast Lesions Using Magnetic Resonance Imaging

Background: Mammograms often reveal breast microcalcifications, necessitating invasive procedures to ascertain whether they are cancerous or benign. Objectives: Although many microcalcifications are linked to noncancerous conditions, this study sought to investigate the efficacy of a computer-aided detection (CAD) system using breast MRI in distinguishing between benign and malignant breast anomalies. Methods: This cross-sectional study included forty patients with mammographically suspicious microcalcifications who underwent stereotactically-guided biopsies at our institution over two years. Prior to the biopsy, these patients received a breast MRI within eight weeks. Surgical interventions were carried out for cases identified as malignant or of uncertain malignant potential. The study aimed to determine diagnostic benchmarks by comparing the breast imaging reporting and database system (BI-RADS) category assignments from initial mammography screenings and breast MRI reports to the pathology findings. Results: Histopathology reports showed that of the total cases, 23 were benign, and 17 were malignant. Breast MRI exhibited a sensitivity of 88.8%, specificity of 54.5%, a positive predictive value of 58.5%, and a negative predictive value of 94.1%. Further analysis using CAD demonstrated sensitivity, specificity, positive predictive value, and negative predictive value of 100%, 50.0%, 59.0%, and 100%, respectively. Conclusions: Utilizing breast MRI with the support of CAD, radiologists could significantly enhance their capability to differentiate between benign and malignant mammographic microcalcifications. This innovative diagnostic approach has the potential to decrease the necessity for unnecessary breast biopsies.

Deep learning-based computer-aided diagnosis system for the automatic detection and classification of lateral cervical lymph nodes on original ultrasound images of papillary thyroid carcinoma: a prospective diagnostic study.

This study aims to develop a deep learning-based computer-aided diagnosis (CAD) system for the automatic detection and classification of lateral cervical lymph nodes (LNs) on original ultrasound images of papillary thyroid carcinoma (PTC) patients. A retrospective data set of 1801 cervical LN ultrasound images from 1675 patients with PTC and a prospective test set including 185 images from 160 patients were collected. Four different deep leaning models were trained and validated in the retrospective data set. The best model was selected for CAD system development and compared with three sonographers in the retrospective and prospective test sets. The Deformable Detection Transformer (DETR) model showed the highest diagnostic efficacy, with a mean average precision score of 86.3% in the retrospective test set, and was therefore used in constructing the CAD system. The detection performance of the CAD system was superior to the junior sonographer and intermediate sonographer with accuracies of 86.3% and 92.4% in the retrospective and prospective test sets, respectively. The classification performance of the CAD system was better than all sonographers with the areas under the curve (AUCs) of 94.4% and 95.2% in the retrospective and prospective test sets, respectively. This study developed a Deformable DETR model-based CAD system for automatically detecting and classifying lateral cervical LNs on original ultrasound images, which showed excellent diagnostic efficacy and clinical utility. It can be an important tool for assisting sonographers in the diagnosis process.

The application of mammography imaging in the diagnosis and prediction of breast diseases

Breast diseases pose a significant threat to women’s health, so early detection and treatment are extremely important. In this context, early disease identification has become crucial in the diagnosis and treatment of breast diseases. This paper begins by outlining the pivotal role of mammography in the early diagnosis of breast cancer, comparing the structural similarities and differences between normal and diseased breast tissues. This comparison underscores the primary role of mammography in the diagnosis and treatment of breast diseases. Additionally, our paper provides an overview of fundamental concepts related to breast cancer detection, diagnosis, and prediction systems. It delves into the latest research developments in auxiliary diagnostic detection, examination, and risk prediction systems associated with breast cancer. Our objective is to offer a comprehensive understanding of the role of computer-aided detection, diagnosis, and prediction systems in breast diseases, fostering further development and application. This work aims to explore and drive innovation in the field, enhance early detection rates of breast diseases, and guide readers towards novel directions, thus contributing to female healthcare management.

Robustness evaluation of deep neural networks for endoscopic image analysis: Insights and strategies

Computer-aided detection and diagnosis systems (CADe/CADx) in endoscopy are commonly trained using high-quality imagery, which is not representative for the heterogeneous input typically encountered in clinical practice. In endoscopy, the image quality heavily relies on both the skills and experience of the endoscopist and the specifications of the system used for screening. Factors such as poor illumination, motion blur, and specific post-processing settings can significantly alter the quality and general appearance of these images. This so-called domain gap between the data used for developing the system and the data it encounters after deployment, and the impact it has on the performance of deep neural networks (DNNs) supportive endoscopic CAD systems remains largely unexplored. As many of such systems, for e.g. polyp detection, are already being rolled out in clinical practice, this poses severe patient risks in particularly community hospitals, where both the imaging equipment and experience are subject to considerable variation. Therefore, this study aims to evaluate the impact of this domain gap on the clinical performance of CADe/CADx for various endoscopic applications. For this, we leverage two publicly available data sets (KVASIR-SEG and GIANA) and two in-house data sets. We investigate the performance of commonly-used DNN architectures under synthetic, clinically calibrated image degradations and on a prospectively collected dataset including 342 endoscopic images of lower subjective quality. Additionally, we assess the influence of DNN architecture and complexity, data augmentation, and pretraining techniques for improved robustness. The results reveal a considerable decline in performance of 11.6% (±1.5) as compared to the reference, within the clinically calibrated boundaries of image degradations. Nevertheless, employing more advanced DNN architectures and self-supervised in-domain pre-training effectively mitigate this drop to 7.7% (±2.03). Additionally, these enhancements yield the highest performance on the manually collected test set including images with lower subjective quality. By comprehensively assessing the robustness of popular DNN architectures and training strategies across multiple datasets, this study provides valuable insights into their performance and limitations for endoscopic applications. The findings highlight the importance of including robustness evaluation when developing DNNs for endoscopy applications and propose strategies to mitigate performance loss.

FCCS-Net: Breast cancer classification using Multi-Level fully Convolutional-Channel and spatial attention-based transfer learning approach

Breast cancer ranks among one of the most lethal cancer varieties among the many types that exist. Timely detection is paramount, as late diagnosis can exacerbate its severity. Computer-aided detection systems can complement the clinician in early decision-making. Therefore, in this study, a multi-level, complete convolution-driven attention-based transfer learning approach named ‘FCCS-Net’ has been proposed, for breast cancer classification. In contrast to the shared multi-layer perceptron (MLP)-based attention mechanism, the proposed approach employs a fully convolutional attention mechanism to focus the important cellular features in inter-channel and intra-channel feature space. This proposed attention is applied across multiple levels of a pre-trained ResNet18 model, supplemented with additional residual connections. The performance of the proposed FCCS-Net is tested on publicly available datasets such as ‘BreakHis’,‘IDC’ and ‘BACH’, containing breast cancer histopathology images. On the BreakHis dataset, the proposed method achieves accuracy rates of 99.25%, 98.32%, 99.50%, and 96.98% at 40X, 100X, 200X, and 400X optical zoom levels, respectively. In the case of the IDC dataset, a classification accuracy of 90.58% is attained at 40X magnifications, whereas with BACH dataset 91.25% average classification accuracy has been obtained. These findings establish the robustness and efficacy of the FCCS-Net in detecting breast cancer through histopathology images. The area focused by each attention layer has also been visually explained. The integration of multi-level, fully convolutional attention with supplementary residual connections holds the potential to advance breast cancer detection methodologies. The relevant PyTorch code for implementing the FCCS-Net model can be accessed at https://github.com/maurya123ritesh47/FCCS-Net.

The Impact of Expectation Management and Model Transparency on Radiologists’ Trust and Utilization of AI Recommendations for Lung Nodule Assessment on Computed Tomography: Simulated Use Study

Background Many promising artificial intelligence (AI) and computer-aided detection and diagnosis systems have been developed, but few have been successfully integrated into clinical practice. This is partially owing to a lack of user-centered design of AI-based computer-aided detection or diagnosis (AI-CAD) systems. Objective We aimed to assess the impact of different onboarding tutorials and levels of AI model explainability on radiologists’ trust in AI and the use of AI recommendations in lung nodule assessment on computed tomography (CT) scans. Methods In total, 20 radiologists from 7 Dutch medical centers performed lung nodule assessment on CT scans under different conditions in a simulated use study as part of a 2×2 repeated-measures quasi-experimental design. Two types of AI onboarding tutorials (reflective vs informative) and 2 levels of AI output (black box vs explainable) were designed. The radiologists first received an onboarding tutorial that was either informative or reflective. Subsequently, each radiologist assessed 7 CT scans, first without AI recommendations. AI recommendations were shown to the radiologist, and they could adjust their initial assessment. Half of the participants received the recommendations via black box AI output and half received explainable AI output. Mental model and psychological trust were measured before onboarding, after onboarding, and after assessing the 7 CT scans. We recorded whether radiologists changed their assessment on found nodules, malignancy prediction, and follow-up advice for each CT assessment. In addition, we analyzed whether radiologists’ trust in their assessments had changed based on the AI recommendations. Results Both variations of onboarding tutorials resulted in a significantly improved mental model of the AI-CAD system (informative P=.01 and reflective P=.01). After using AI-CAD, psychological trust significantly decreased for the group with explainable AI output (P=.02). On the basis of the AI recommendations, radiologists changed the number of reported nodules in 27 of 140 assessments, malignancy prediction in 32 of 140 assessments, and follow-up advice in 12 of 140 assessments. The changes were mostly an increased number of reported nodules, a higher estimated probability of malignancy, and earlier follow-up. The radiologists’ confidence in their found nodules changed in 82 of 140 assessments, in their estimated probability of malignancy in 50 of 140 assessments, and in their follow-up advice in 28 of 140 assessments. These changes were predominantly increases in confidence. The number of changed assessments and radiologists’ confidence did not significantly differ between the groups that received different onboarding tutorials and AI outputs. Conclusions Onboarding tutorials help radiologists gain a better understanding of AI-CAD and facilitate the formation of a correct mental model. If AI explanations do not consistently substantiate the probability of malignancy across patient cases, radiologists’ trust in the AI-CAD system can be impaired. Radiologists’ confidence in their assessments was improved by using the AI recommendations.

Revolutionizing gastric cancer diagnosis through advanced machine learning approaches

<p class="Keywords">Early detection of gastric cancer through a Computer-Aided Detection (CAD) system has the potential to significantly reduce the mortality rate associated with this disease. This study aims to investigate the effects of class imbalance on the performance of machine learning classifiers in this context. Using a dataset of 145,787 screening records from NHS Liverpool Hospital, we employed stratified sampling to create balanced and unbalanced datasets and evaluated the performance of four machine learning algorithms—Logistic Regression, Support Vector Machine, Naive Bayes, and Multilayer Perceptron—under five different test conditions. The study’s novelty lies in its detailed examination of class imbalance in gastric cancer diagnosis, emphasizing the crucial role of balanced datasets in machine learning-based early detection systems. For the MLP model under 10-fold cross-validation, the Class 0 sensitivity (non-cancer cases) of the unbalanced dataset was 0.968, higher than the balanced dataset’s 0.902. However, the Class 1 sensitivity (cancer cases) and Positive Predictive Value (PPV) of the unbalanced dataset were much lower (0.383 and 0.527) than those of the balanced dataset (0.959 and 0.907), indicating a significant improvement in identifying true positive cases when using a balanced dataset. These findings highlight the negative effect of class imbalance on prediction accuracy for positive cancer cases and underscore the importance of addressing this imbalance for more reliable and accurate predictions in medical diagnosis and screening. This approach has the potential to improve patient outcomes and may contribute to strategies aimed at reducing the mortality rate associated with gastric cancer.</p>

A computer-aided polyp detection system in screening and surveillance colonoscopy: an international, multicentre, randomised, tandem trial

Studies on the effect of computer-aided detection (CAD) in a daily clinical screening and surveillance colonoscopy population practice are scarce. The aim of this study was to evaluate a novel CAD system in a screening and surveillance colonoscopy population. This multicentre, randomised, controlled trial was done in ten hospitals in Europe, the USA, and Israel by 31 endoscopists. Patients referred for non-immunochemical faecal occult blood test (iFOBT) screening or surveillance colonoscopy were included. Patients were randomomly assigned to CAD-assisted colonoscopy or conventional colonoscopy; a subset was further randomly assigned to undergo tandem colonoscopy: CAD followed by conventional colonoscopy or conventional colonoscopy followed by CAD. Primary objectives included adenoma per colonoscopy (APC) and adenoma per extraction (APE). Secondary objectives included adenoma miss rate (AMR) in the tandem colonoscopies. The study was registered at ClinicalTrials.gov, NCT04640792. A total of 916 patients were included in the modified intention-to-treat analysis: 449 in the CAD group and 467 in the conventional colonoscopy group. APC was higher with CAD compared with conventional colonoscopy (0·70 vs 0·51, p=0·015; 314 adenomas per 449 colonoscopies vs 238 adenomas per 467 colonoscopies; poisson effect ratio 1·372 [95% CI 1·068-1·769]), while showing non-inferiority of APE compared with conventional colonoscopy (0·59 vs 0·66; p<0·001 for non-inferiority; 314 of 536 extractions vs 238 of 360 extractions). AMR in the 127 (61 with CAD first, 66 with conventional colonoscopy first) patients completing tandem colonoscopy was 19% (11 of 59 detected during the second pass) in the CAD first group and 36% (16 of 45 detected during the second pass) in the conventional colonoscopy first group (p=0·024). CAD increased adenoma detection in non-iFOBT screening and surveillance colonoscopies and reduced adenoma miss rates compared with conventional colonoscopy, without an increase in the resection of non-adenomatous lesions. Magentiq Eye.

Deep learning-based lung nodule detection: a review

CT scan acquisition is fast and cost-effective and has become the main lung imaging tool. However, the increase in large numbers of CT scans has placed a heavy burden on radiologists; therefore, automated lung nodule detection techniques are needed to reduce the workload of radiologists and computer-aided detection systems are proposed for further accurate diagnosis of the condition. This review provides a comprehensive overview of recent automated lung nodule detection techniques and challenges, etc., as well as a detailed overview and discussion of current research gaps, future developments, and research trends. Relevant articles published in databases such as IEEE Xplore, Science Direct, PubMed, and Web of Science cover research algorithms published from 2014 to 2023, mainly discussing deep learning-based techniques. The schemes presented in these articles, the databases used, the experimental results, and the performance of the algorithms are compared and discussed. This work aims to introduce researchers and readers to the latest techniques and their advances in the detection of lung nodules in the last decade, which will help researchers and radiologists to further understand the latest techniques in this field.

A136 EXPLORING ENDOSCOPIST PERCEPTIONS OF ARTIFICIAL INTELLIGENCE-AIDED COLONOSCOPY: A QUALITATIVE ANALYSIS

Abstract Background Artificial intelligence (AI) is gaining recognition as a promising adjunct in healthcare including gastrointestinal endoscopy. Randomized controlled trials have demonstrated improved polyp detection with computer-aided detection (CADe) systems in colonoscopy. Despite the foreseeable translation of CADe systems into the endoscopy suite, no study to date has explored how introducing this novel technology influences endoscopist perceptions and cognitive processes. Aims To explore how AI assistance impacts endoscopists’ perceptions of and cognitive processes during colonoscopy. Methods Faculty in gastroenterology and general surgery at the University Health Network (Toronto, Canada) were interviewed in April 2023 to explore their baseline perceptions of AI before the planned installation of the Medtronic GI GeniusTM Intelligent Endoscopy Module, an AI polyp detection tool, at Toronto Western Hospital in May 2023. After performing a minimum of 10 colonoscopies using GI GeniusTM, participants were re-interviewed to discuss their perceptions of AI-assisted colonoscopy and how it influenced their cognition during the procedure. Analysis was informed by constructivist grounded theory, whereby interview data were transcribed and coded iteratively using constant comparison to generate themes. Results In this interim constant comparative analysis, 9 participants (6 gastroenterologists and 3 general surgeons) were interviewed to generate 16 interview transcripts (9 pre-exposure and 7 post-exposure; 1 interview pending, 1 participant on leave). Participants held the view that: (1) AI will inevitably become the standard of care in colonoscopy, where clinicians and AI function symbiotically and in partnership; (2) CADe systems may facilitate standardization of practice and reduce inter-endoscopist variability by improving detection of specific types of lesions (e.g., sessile serrated polyps); (3) CADe assistance provides a “second set of eyes” that interacts with, but does not replace, endoscopists’ cognitive processes; and (4) trainees should gain familiarity with AI systems, despite mixed opinions regarding AI’s specific role in training. Conclusions We found that study participants conceptualize AI as an essential component of colonoscopy practice and training, with endoscopist thought processes and CADe systems appearing to function symbiotically. Elucidating the specific ways in which AI and endoscopist cognition interact will facilitate the future refinement of these tools. Funding Agencies None

TOWARD EFFECTIVE BREAST CANCER DETECTION IN THERMAL IMAGES USING EFFICIENT FEATURE SELECTION ALGORITHM AND FEATURE EXTRACTION METHODS

In this paper, an intelligent method is developed for improving the performance of the Computer-Aided Detection (CAD) system. The research objective is to improve the performance of the CAD system in Breast Cancer (BC) detection with high accuracy using thermal images. The research strategy is efficient using feature extraction, feature selection, classification and artificial intelligence methods. In the developed method, the features in the Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) are extracted from images. The features are selected using the firefly feature selection algorithm. These selected features are observed to be relevant for the abnormality detection in healthy and unhealthy breasts. The [Formula: see text]-Nearest Neighbors (kNN), Support Vector Machine (SVM), and Decision-Tree (D-Tree) classifiers are then applied to these features for the detection of malignancy in the breast. The breast thermograms of 200 subjects available at the Database for Mastology Research for breast research using InfraRed images, DMR-IR database, are considered for evaluation of our intelligent method. The results demonstrate that the accuracy is 98.8%, 81.5%, and 94.5%, the sensitivity is 99%, 83.15%, and 94%, and the specificity is 98.2%, 80%, and 93% when using SVM, kNN, and D-Tree classifier algorithm, respectively. This reveals the effectiveness of our intelligent method to improve the accuracy of the CAD system in the BC detection.

Performance Analysis in Children of Traditional and Deep Learning CT Lung Nodule Computer-Aided Detection Systems Trained on Adults.

BACKGROUND. Although primary lung cancer is rare in children, chest CT is commonly performed to assess for lung metastases in children with cancer. Lung nodule computer-aided detection (CAD) systems have been designed and studied primarily using adult training data, and the efficacy of such systems when applied to pediatric patients is poorly understood. OBJECTIVE. The purpose of this study was to evaluate in children the diagnostic performance of traditional and deep learning CAD systems trained with adult data for the detection of lung nodules on chest CT scans and to compare the ability of such systems to generalize to children versus to other adults. METHODS. This retrospective study included pediatric and adult chest CT test sets. The pediatric test set comprised 59 CT scans in 59 patients (30 boys, 29 girls; mean age, 13.1 years; age range, 4-17 years), which were obtained from November 30, 2018, to August 31, 2020; lung nodules were annotated by fellowship-trained pediatric radiologists as the reference standard. The adult test set was the publicly available adult Lung Nodule Analysis (LUNA) 2016 subset 0, which contained 89 deidentified scans with previously annotated nodules. The test sets were processed through the traditional FlyerScan (github.com/rhardie1/FlyerScanCT) and deep learning Medical Open Network for Artificial Intelligence (MONAI; github.com/Project-MONAI/model-zoo/releases) lung nodule CAD systems, which had been trained on separate sets of CT scans in adults. Sensitivity and false-positive (FP) frequency were calculated for nodules measuring 3-30 mm; nonoverlapping 95% CIs indicated significant differences. RESULTS. Operating at two FPs per scan, on pediatric testing data FlyerScan and MONAI showed significantly lower detection sensitivities of 68.4% (197/288; 95% CI, 65.1-73.0%) and 53.1% (153/288; 95% CI, 46.7-58.4%), respectively, than on adult LUNA 2016 subset 0 testing data (83.9% [94/112; 95% CI, 79.1-88.0%] and 95.5% [107/112; 95% CI, 90.0-98.4%], respectively). Mean nodule size was smaller (p < .001) in the pediatric testing data (5.4 ± 3.1 [SD] mm) than in the adult LUNA 2016 subset 0 testing data (11.0 ± 6.2 mm). CONCLUSION. Adult-trained traditional and deep learning-based lung nodule CAD systems had significantly lower sensitivity for detection on pediatric data than on adult data at a matching FP frequency. The performance difference may relate to the smaller size of pediatric lung nodules. CLINICAL IMPACT. The results indicate a need for pediatric-specific lung nodule CAD systems trained on data specific to pediatric patients.

Evaluating false-positive detection in a computer-aided detection system for colonoscopy.

Computer-aided detection (CADe) systems can efficiently detect polyps during colonoscopy. However, false-positive (FP) activation is a major limitation of CADe. We aimed to compare the rate and causes of FP using CADe before and after an update designed to reduce FP. We analyzed CADe-assisted colonoscopy videos recorded between July 2022 and October 2022. The number and causes of FPs and excessive time spent by the endoscopist on FP (ET) were compared pre- and post-update using 1:1 propensity score matching. During the study period, 191 colonoscopy videos (94 and 97 in the pre- and post-update groups, respectively) were recorded. Propensity score matching resulted in 146 videos (73 in each group). The mean number of FPs and median ET per colonoscopy were significantly lower in the post-update group than those in the pre-update group (4.2±3.7 vs 18.1±11.1; P<0.001 and 0 vs 16s; P<0.001, respectively). Mucosal tags, bubbles, and folds had the strongest association with decreased FP post-update (pre-update vs post-update: 4.3±3.6 vs 0.4±0.8, 0.32±0.70 vs 0.04±0.20, and 8.6±6.7 vs 1.6±1.7, respectively). There was no significant decrease in the true positive rate (post-update vs pre-update: 95.0% vs 99.2%; P=0.09) or the adenoma detection rate (post-update vs pre-update: 52.1% vs 49.3%; P=0.87). The updated CADe can reduce FP without impairing polyp detection. A reduction in FP may help relieve the burden on endoscopists.

Assessing clinical efficacy of polyp detection models using open-access datasets.

Ensuring accurate polyp detection during colonoscopy is essential for preventing colorectal cancer (CRC). Recent advances in deep learning-based computer-aided detection (CADe) systems have shown promise in enhancing endoscopists' performances. Effective CADe systems must achieve high polyp detection rates from the initial seconds of polyp appearance while maintaining low false positive (FP) detection rates throughout the procedure. We integrated four open-access datasets into a unified platform containing over 340,000 images from various centers, including 380 annotated polyps, with distinct data splits for comprehensive model development and benchmarking. The REAL-Colon dataset, comprising 60 full-procedure colonoscopy videos from six centers, is used as the fifth dataset of the platform to simulate clinical conditions for model evaluation on unseen center data. Performance assessment includes traditional object detection metrics and new metrics that better meet clinical needs. Specifically, by defining detection events as sequences of consecutive detections, we compute per-polyp recall at early detection stages and average per-patient FPs, enabling the generation of Free-Response Receiver Operating Characteristic (FROC) curves. Using YOLOv7, we trained and tested several models across the proposed data splits, showcasing the robustness of our open-access platform for CADe system development and benchmarking. The introduction of new metrics allows for the optimization of CADe operational parameters based on clinically relevant criteria, such as per-patient FPs and early polyp detection. Our findings also reveal that omitting full-procedure videos leads to non-realistic assessments and that detecting small polyp bounding boxes poses the greatest challenge. This study demonstrates how newly available open-access data supports ongoing research progress in environments that closely mimic clinical settings. The introduced metrics and FROC curves illustrate CADe clinical efficacy and can aid in tuning CADe hyperparameters.

Segmentation of Mass in Mammogram Using Gaze Search Patterns

It is very difficult for a radiologist to correctly detect small lesions and lesions hidden on dense breast tissue on a mammogram. Therefore, recently, computer-aided detection (CAD) systems have been widely used to assist radiologists in interpreting images. Thus, in this study, we aimed to segment mass on the mammogram with high accuracy by using focus images obtained from an eye-tracking device. We obtained focus images for two mammography expert radiologists and 19 mammography technologists on 8 abnormal and 8 normal mammograms published by the DDSM. Next, the auto-encoder, Pix2Pix, and UNIT learned the relationship between the actual mammogram and the focus image, and generated the focus image for the unknown mammogram. Finally, we segmented regions of mass on mammogram using the U-Net for each focus image generated by the auto-encoder, Pix2Pix, and UNIT. The dice coefficient in the UNIT was 0.64±0.14. The dice coefficient in the UNIT was higher than that in the auto-encoder and Pix2Pix, and there was a statistically significant difference (p<0.05). The dice coefficient of the proposed method, which combines the focus images generated by the UNIT and the original mammogram, was 0.66±0.15, which is equivalent to the method using the original mammogram. In the future, it will be necessary to increase the number of cases and further improve the segmentation.

The image segmentation and polyp detection in colonoscopy images using computer aided detection (CAD)

Segmentation is the process through which each pixel in an image is assigned a class. Segmentation may be used to medical image analysis to assist in image-guided surgery, radiation therapy, and better radiological diagnostics. This article summarises our study towards creating a computer-assisted detection (CAD) method for colonoscopy images containing polyps. Our method is a hybrid context-shape strategy that uses shape information to accurately detect polyps and context information to eliminate non-polyp structures. To begin, a basic edge map is generated from a colonoscopy picture. Second, we use our unique feature extraction and edge classification technique to remove non-polyp edges from the edge map. Third, we use our updated edge maps and newly developed voting mechanism to identify polyp candidates with probabilistic confidence ratings. The proposed CAD system was compared to two publicly accessible polyp databases: CVC-ColonDB, which includes 300 colonoscopy pictures with 300 polyp occurrences from 15 different polyps, and ASU-Mayo, which contains 19,400 colonoscopy frames with a total of 5,200 polyp instances from 10 distinct polyps.

Detection of fibrosing interstitial lung disease-suspected chest radiographs using a deep learning-based computer-aided detection system: a retrospective, observational study

ObjectivesTo investigate the effectiveness of BMAX, a deep learning-based computer-aided detection system for detecting fibrosing interstitial lung disease (ILD) on chest radiographs among non-expert and expert physicians in the real-world...

TOWARDS NOVEL DIAGNOSTIC MARKER BASED ON BREAST ARTERIAL CALCIFICATIONS

In recent years, medical image analysis has advanced significantly, mainly due to the development of computer vision tools. With AI becoming more and more prevalent, these technologies offer tremendous potential for the field of radiology. Novel computer-aided detection (CAD) methods can aid radiologists by reducing their workload, decreasing costs, speeding up diagnoses, and providing supporting tools for more accurate diagnostic assessments. Our latest study on Breast Arterial Calcification (BAC) demonstrates how these cutting-edge CAD systems can seamlessly integrate into clinical practice, ultimately enhancing patient care efficiency and effectiveness. Recent literature [1-3] shows that BAC could become a cardiovascular disease (CVD) biomarker. In collaboration with Radboud UMC (Nijmegen,The Netherlands), we evaluated a novel BAC detector capable of identifying BAC occurrence in screening mammograms. It is particularly challenging since even well-trained radiologists often differ in BAC grade assessments. Detecting and quantifying early BAC stages, which are not apparent in the mammograms, is crucial for accurate CVD assessment. Using advanced imaging techniques and statistical modelling, our research aims to determine BAC prevalence and severity concerning CVD outcomes, including stroke and coronary artery disease. We achieved a 70% accuracy rate in BAC detection from 500 mammograms annotated bytworadiologists via consensus. It is a noteworthy achievement given the heterogeneous, imbalanced data and weak labelling provided byradiologists. However, we are still looking for a sufficient pool of annotated parallel BAC-CVD cases, which is essential for further study. Although annotation is costly, and quality varies with radiologist expertise, we aim to expand our dataset to include data from other institutions to provide more generalizable results, preferably independent of demographic biases.