Computer-aided Detection Research Articles

Full dimensional dynamic 3D convolution and point cloud in pulmonary nodule detection

Lung cancer is a leading cause of death worldwide, making early and accurate diagnosis essential for improving patient outcomes. Recently, deep learning (DL) has proven to be a powerful tool, significantly enhancing the accuracy of computer-aided pulmonary nodule detection (PND). In this study, we introduce a novel approach called the Omni-dimension Dynamic Residual 3D Net (ODR3DNet) for PND, which utilizes full-dimensional dynamic 3D convolution, along with a specialized machine learning algorithm for detecting lung nodules in 3D point clouds. The primary goal of ODR3DNet is to overcome the limitations of conventional 3D Convolutional Neural Networks (CNNs), which often struggle with adaptability and have limited feature extraction capabilities. Our ODR3DNet algorithm achieves a high CPM (Competition Performance Metric) score of 0.885, outperforming existing mainstream PND algorithms and demonstrating its effectiveness. Through detailed ablation experiments, we confirm that the OD3D module plays a crucial role in this performance boost and identify the optimal configuration for the algorithm. Moreover, we developed a dedicated machine learning detection algorithm tailored for lung 3D point cloud data. We outline the key steps for reconstructing the lungs in 3D and establish a comprehensive process for building a lung point cloud dataset, including data preprocessing, 3D point cloud conversion, and 3D volumetric box annotation. Experimental results validate the feasibility and effectiveness of our proposed approach.

Corrigendum to: Computer-aided Detection and Diagnosis of Cancer Regions in Mammogram Images using Resource-Efficient CNN Architecture

<p>In the online version of the article, a change was made in the author’s affiliation section. The affiliation of Manikandan Thiyagarajanin in the online version of the article titled “Computer-aided Detection and Diagnosis of Cancer Regions in Mammogram Images using Resource-Efficient CNN Architecture” has been updated in “Current Medical Imaging Reviews”, 2024; 20: e15734056309483 [1]. </p> <p> The original article can be found online at: https://www.eurekaselect.com/article/142257 </p> <p> Original: </p> <p> Helan Vidhya Thankaraj<sup>1,*</sup> and Manikandan Thiyagarajan<sup>1,2</sup> </p> <p> <sup>1</sup>Assistant Professor (S.G) Electronics and Communication Engineering, Rajalakshmi Engineering College, Thandalam, Chennai, India </p> <p> <sup>2</sup>Professor, Electronics and Communication Engineering, Rajalakshmi Engineering College, Thandalam, Chennai, India </p> <p> Corrected: </p> <p> Helan Vidhya Thankaraj<sup>1,*</sup> and Manikandan Thiyagarajan<sup>2</sup> </p> <p> <sup>1</sup>Assistant Professor (S.G) Electronics and Communication Engineering, Rajalakshmi Engineering College, Thandalam, Chennai, India </p> <p> <sup>2</sup>Professor, Electronics and Communication Engineering, Rajalakshmi Engineering College, Thandalam, Chennai, India</p>

Interventions and implementation considerations for reducing pre-treatment loss to follow-up in adults with pulmonary tuberculosis: A scoping review

Background Tuberculosis (TB) is a leading cause of death worldwide with over 90% of reported cases occurring in low- and middle-income countries (LMICs). Pre-treatment loss to follow-up (PTLFU) is a key contributor to TB mortality and infection transmission. Objectives We performed a scoping review to map available evidence on interventions to reduce PTLFU in adults with pulmonary TB, identify gaps in existing knowledge, and develop a conceptual framework to guide intervention implementation. Methods We searched eight electronic databases up to February 6 2024, medRxiv for pre-prints, and reference lists of included studies. Two review authors independently selected studies and extracted data using a predesigned form. We analysed data descriptively, presented findings in a narrative summary and developed a conceptual framework based on the Practical, Robust Implementation, and Sustainability Model to map the factors for effective intervention implementation. Results We reviewed 1262 records and included 17 studies. Most studies were randomized controlled trials (8/17, 47%). Intervention barriers included stigma and inadequate resources; enablers included mobile phones and TB testing and results on the same day. We identified eight interventions that reduced PTLFU: treatment support groups; mobile notifications; community health workers; integrated HIV/TB services; Xpert MTB/RIF as the initial diagnostic test; computer-aided detection with chest radiography screening; active linkage to care; and multi-component strategies. Conclusion Given the variation of healthcare settings, TB programs should consider contextual factors such as user acceptability, political commitment, resources, and infrastructure before adopting an intervention. Future research should utilize qualitative study designs, be people-centred, and include social and economic factors affecting PTLFU.

Mandibular cortical morphology in patients with osteoporosis using computer‐aided detection with dental panoramic imaging: A pilot study of the relation to age and bone‐modifying agents

AbstractAimMandibular cortical morphology with panoramic radiography is considered to be a useful technique for identifying at risk of osteoporosis. The aim of the study was to investigate the mandibular cortical morphology in patients with osteoporosis using computer‐aided detection (CAD) with dental panoramic imaging, especially the relation to age and bone‐modifying agents.MethodsOne hundred and thirteen patients with osteoporosis were underwent panoramic radiography. The mandibular cortical morphology, such as mandibular cortical index (MCI) and degree of deformation of mandibular inferior cortex, was analyzed automatically with a CAD with dental panoramic imaging “PanoSCOPE.” Those were analyzed in relation to age and bone‐modifying agents.ResultsRegarding Class 3 (severely eroded) of the MCI classifications, the frequency in 61–70, 71–80, and 81–90 years were 22.2%, 38.6%, and 26.3%, respectively, and the MCI in bisphosphonate and denosumab were 29.8% and 42.1%, respectively. The degree of deformation in 61–70, 71–80, and 81–90 years were 45.9 ± 22.6, 58.8 ± 20.0, and 52.6 ± 21.5, respectively, and those in bisphosphonate and denosumab were 54.1 ± 20.6 and 57.5 ± 25.0, respectively. The MCI and degree of deformation in patients with osteoporosis were not significant differences for age and bone‐modifying agents.ConclusionThis study suggests that the CAD using PanoSCOPE can perform a quantitative analysis of mandibular cortical morphology in patients with osteoporosis.

Predicting the generalization of computer aided detection (CADe) models for colonoscopy

Generalizability of AI colonoscopy algorithms is important for wider adoption in clinical practice. However, current techniques for evaluating performance on unseen data require expensive and time-intensive labels. We show that a "Masked Siamese Network" (MSN), trained to predict masked out regions of polyp images without labels, can predict the performance of Computer Aided Detection (CADe) of polyps on colonoscopies, without labels. This holds on Japanese colonoscopies even when MSN is only trained on Israeli colonoscopies, which differ in scoping hardware, endoscope software, screening guidelines, bowel preparation, patient demographics, and the use of techniques such as narrow-band imaging (NBI) and chromoendoscopy (CE). Since our technique uses neither colonoscopy-specific information nor labels, it has the potential to apply to more medical imaging domains.

Deep Learning-Based Slice Thickness Reduction for Computer-Aided Detection of Lung Nodules in Thick-Slice CT.

Computer-aided detection (CAD) systems for lung nodule detection often face challenges with 5 mm computed tomography (CT) scans, leading to missed nodules. This study assessed the efficacy of a deep learning-based slice thickness reduction technique from 5 mm to 1 mm to enhance CAD performance. In this retrospective study, 687 chest CT scans were analyzed, including 355 with nodules and 332 without nodules. CAD performance was evaluated on nodules, to which all three radiologists agreed. The slice thickness reduction technique significantly improved the area under the receiver operating characteristic curve (AUC) for scan-level analysis from 0.867 to 0.902, with a p-value < 0.001, and nodule-level sensitivity from 0.826 to 0.916 at two false positives per scan. Notably, the performance showed greater improvements on smaller nodules than larger nodules. Qualitative analysis confirmed that nodules mistaken for ground glass on 5 mm scans could be correctly identified as part-solid on the refined 1 mm CT, thereby improving the diagnostic capability. Applying a deep learning-based slice thickness reduction technique significantly enhances CAD performance in lung nodule detection on chest CT scans, supporting the clinical adoption of refined 1 mm CT scans for more accurate diagnoses.

Enhancing detection of various pancreatic lesions on endoscopic ultrasound through artificial intelligence: a basis for computer-aided detection systems.

Endoscopic ultrasound (EUS) is the most sensitive method for evaluation of pancreatic lesions but is limited by significant operator dependency. Artificial intelligence (AI), in the form of computer-aided detection (CADe) systems, has shown potential in increasing accuracy and bridging operator dependency in several endoscopic domains. However, the complexity of integrating AI into EUS is far more challenging. This aims to develop and test the basis for a CADe system for real-time detection and segmentation of all pancreatic lesions. In this single-center study EUS studies of pancreatic findings were included. Lesions were outlined by two expert (>5years performing EUS) endoscopists, and the two leading types of models were benchmarked. The models' performance was evaluated through per-pixel intersection over union (IoU). A total of 1497 EUS images from 165 patients were evaluated. The dataset included malignancies, neuroendocrine tumors, benign cysts, chronic and acute pancreatitis, normal fatty pancreas, and benign lesions. The best model demonstrated detection and segmentation on the test set with a mean IoU of 0.73, achieving a PPV, NPV, total accuracy, and ROC of 0.82, 0.96, 0.95, and 0.95, respectively. The algorithm is adaptable for real-time processing. We developed and tested deep learning models for real-time detection and segmentation of pancreatic lesions during EUS with promising results. This constitutes the basis for a CADe system for EUS, which could be valuable in future detection and evaluation of pancreatic lesions. Further studies for validation and generalization are underway.

Clinical consequences of computer-aided colorectal polyp detection

Background and aimRandomised trials show improved polyp detection with computer-aided detection (CADe), mostly of small lesions. However, operator and selection bias may affect CADe’s true benefit. Clinical outcomes of increased...

Efficacy of Mammographic Artificial Intelligence-Based Computer-Aided Detection in Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy.

This study evaluates the potential of an AI-based computer-aided detection (AI-CAD) system in digital mammography for predicting pathologic complete response (pCR) in breast cancer patients after neoadjuvant chemotherapy (NAC). A retrospective analysis of 132 patients who underwent NAC and surgery between January 2020 and December 2022 was performed. Pre- and post-NAC mammograms were analyzed using conventional CAD and AI-CAD systems, with negative exams defined by the absence of marked abnormalities. Two radiologists reviewed mammography, ultrasound, MRI, and diffusion-weighted imaging (DWI). Concordance rates between CAD and AI-CAD were calculated, and the diagnostic performance, including the area under the receiver operating characteristics curve (AUC), was assessed. The pre-NAC concordance rates were 90.9% for CAD and 97% for AI-CAD, while post-NAC rates were 88.6% for CAD and 89.4% for AI-CAD. The MRI had the highest diagnostic performance for pCR prediction, with AI-CAD performing comparably to other modalities. Univariate analysis identified significant predictors of pCR, including AI-CAD, mammography, ultrasound, MRI, histologic grade, ER, PR, HER2, and Ki-67. In multivariable analysis, negative MRI, histologic grade 3, and HER2 positivity remained significant predictors. In conclusion, this study demonstrates that AI-CAD in digital mammography shows the potential to examine the pCR of breast cancer patients following NAC.

Diagnostic accuracy of TB screening tests in a prospective multinational cohort: Chest-X-ray with computer-aided detection, Xpert TB host response, and C-reactive protein.

Accessible, accurate screening tests are necessary to advance tuberculosis (TB) case finding and early detection in high-burden countries. We prospectively screened adults with ≥2 weeks of cough at primary health centers in the Philippines, Vietnam, South Africa, Uganda, and India. Participants received chest-X-ray, Cepheid Xpert TB Host Response (Xpert HR) testing, and point-of-care C-reactive protein (CRP) testing (Boditech). Chest-X-ray images were processed using CAD4TB v7, a computer-aided detection algorithm. We assessed diagnostic accuracy against a microbiologic reference standard (sputum Xpert Ultra, culture). Optimal cut-points were chosen to maximize specificity at 90% sensitivity. Two-test screening algorithms were considered, using 1) sequential negative serial screening (positive defined as positive on either test) and 2) sequential positive serial screening (positive defined as positive on both tests). Between July 2021 and August 2022, 1,392 participants with presumptive TB had valid index tests and reference standard results, and 303 (22%) had confirmed TB. In head-to-head comparisons, CAD4TB v7 showed the highest specificity at 90% sensitivity (70.3% vs. 65.1% for Xpert HR, difference 95% CI 1.6 to 8.9; 49.7% for CRP, difference 95% CI 17.0 to 24.3). Three two-test screening algorithms met WHO target product profile (TPP) minimum accuracy thresholds and had higher accuracy than any test alone. At 90% sensitivity, the specificity was 79.6% for Xpert HR-CAD4TB [sequential negative], 75.9% for CRP-CAD4TB [sequential negative], and 73.7% for Xpert HR-CAD4TB [sequential positive]. CAD4TB achieves TPP targets and outperforms Xpert HR and CRP. Combining screening tests further increased accuracy. NCT04923958.

Creating a Standardized Tool for the Evaluation and Comparison of Artificial Intelligence-Based Computer-Aided Detection Programs in Colonoscopy: a Modified Delphi Approach

BACKGROUND AND AIMMultiple computer-aided detection (CADe) software have now achieved regulatory approval in the US, Europe, and Asia and are being used in routine clinical practice to support colorectal cancer screening. There is uncertainty regarding how different CADe algorithms may perform. No objective methodology exists for comparing different algorithms. We aimed to identify priority scoring metrics for CADe evaluation and comparison. METHODSA modified Delphi approach was used. Twenty-five global leaders in CADe in colonoscopy, including endoscopists, researchers, and industry representatives, participated in an online survey over the course of 8 months. Participants generated 121 scoring criteria, 54 of which were deemed within the study scope and distributed for review and asynchronous email-based open comment. Participants then scored criteria in order of priority on a 5-point Likert scale during ranking round one. The top eleven highest-priority criteria were re-distributed, with another opportunity for open-comment, followed by a final round of priority scoring to identify the final 6 criteria. RESULTSMean priority scores for the 54 criteria ranged from 2.25 to 4.38 following the first ranking round. The top eleven criteria following ranking round one yielded mean priority scores ranging from 3.04 to 4.16. The final six highest priority criteria were 1) sensitivity (average = 4.16) and separate & independent validation of the CADe algorithm (4.16), 3) adenoma detection rate (4.08), 4) false positive rate (4.00), 5) latency (3.84), and 6) adenoma miss rate (3.68). CONCLUSIONSThis is the first reported international consensus statement of priority scoring metrics for CADe in colonoscopy. These scoring criteria should inform CADe software development and refinement. Future research should validate these metrics on a benchmark video data set to develop a validated scoring instrument.

Impact of Deep Learning-Based Computer-Aided Detection and Electronic Notification System for Pneumothorax on Time to Treatment: Clinical Implementation

ObjectiveTo assess whether the implementation of deep learning (DL) computer-aided detection (CAD) that screens for suspected pneumothorax (PTX) on chest radiography (CXR) combined with an electronic notification system (ENS) that simultaneously alerts both the radiologist and the referring clinician would affect time to treatment (TTT) in a real-world clinical practice. MethodsIn May 2022, a commercial deep learning-based CAD and ENS (DL-CAD-ENS) was introduced for all CXRs at an 818-bed general hospital, with 33 attending doctors and their residents using ENS, while 155 others used only CAD. We used difference-in-differences estimates to compare TTT between the CAD and ENS group and the CAD-only group for the period from January 2018 to April 2022 and from May 2022 to April 2023. ResultsA total of 603,028 CXRs from 140,841 unique patients were included, with a PTX prevalence of 2.0%. There was a significant reduction in TTT for supplemental oxygen therapy for the CAD and ENS group compared with the CAD-only group in the post-implementation period (-143.8 min; 95% confidence interval [CI], -277.8 to -9.9; P = .035). However, there was no significant difference in TTT for other treatments, including aspiration or tube-thoracostomy (14.4 min; 95% CI, -35.0 to 63.9) and consultation with the thoracic and cardiovascular surgery department (86.3 min; 95% CI, -175.1 to 347.6). ConclusionThe introduction of a DL-CAD-ENS reduced the time to initiate oxygen supplementation for patients with PTX.

A Data Augmentation Method and the Embedding Mechanism for Detection of Pulmonary Nodules on Small Samples

Abstract Lung Computed Tomography (CT) screening for pulmonary nodules provides an effective method for early diagnosis. The deep-learning-based computer-aided detection (CAD) system effectively identifies and precisely localizes suspicious pulmonary nodules in CT images, thereby significantly enhancing the accuracy and efficiency of CT diagnosis. In the medical field, the availability of medical data is limited, and research using small samples is of practical significance. By studying the data augmentation technology based on the generative model under the condition of small samples, and refining the model structure through the embedding mechanism, the accuracy and robustness of the deep learning model are explored. A 3D pixel-level statistical algorithm is proposed for the generation of pulmonary nodules. By combining simulated pulmonary nodules with healthy lung tissue, we can generate new samples of pulmonary nodules. The embedding mechanism is designed to enhance the comprehension of pixel meanings in pulmonary nodule samples by introducing latent variables. The results of the 3DVNET model with the augmentation method for pulmonary nodule detection under small sample conditions demonstrate that the proposed data augmentation method outperforms the method based on a generative adversarial network (GAN) framework, and the embedding mechanism for pulmonary nodules detection shows a significant improvement in accuracy. Conclusion: the proposed data augmentation method and embedding mechanism demonstrate significant potential in enhancing the accuracy and robustness of the model, thereby facilitating their application to various common imaging diagnostic tasks, and research using small samples is of practical significance.

IncARMAG: A convolutional neural network with multi-level autoregressive moving average graph convolutional processing framework for medical image classification

Effective computer-aided detection and diagnosis algorithms are being sought to carry out complex pattern recognition with great precision while increasing efficiency and robustness through automation. In this study, we leveraged self-constructing graph (SCG) and autoregressive moving average (ARMA) graph convolution to postprocess multi-level feature maps obtained from an Inception V3 network for medical image classification. The adopted SCG learns latent embeddings from the feature maps, which are subsequently used to construct graph topology based on embedding similarity. The ARMA convolution then learns to encode more flexible spectral filter responses that simultaneously takes into account long-range dependencies from the SCG graph. The proposed IncARMAG model was evaluated on 2D MedMNIST datasets, which consist of several imaging modalities and classification tasks in the medical domain. Results from our experiments on MedMNIST datasets showed that IncARMAG outperformed state-of-the-art models on many cases. IncARMAG has also achieved the highest accuracy based on evaluation of various benchmark CNN, transformers, and hybrid models on four selected datasets. These results highlight how IncARMAG is a strong candidate in medical image analysis involving classification tasks.

Gastric neoplasm detection of computer-aided detection-assisted esophagogastroduodenoscopy changes with implement scenarios: a real-world study.

The implementation of computer-aided detection (CAD) devices in esophagogastroduodenoscopy (EGD) could autonomously identify gastric precancerous lesions and neoplasms and reduce the miss rate of gastric neoplasms in prospective trials. However, there is still insufficient evidence of their use in real-life clinical practice. A real-world, two-center study was conducted at Wenzhou Central Hospital (WCH) and Renmin Hospital of Wuhan University (RHWU). High biopsy rate and low biopsy rate strategies were adopted, and CAD devices were applied in 2019 and 2021 at WCH and RHWU, respectively. We compared differences in gastric precancerous and neoplasm detection of EGD before and after the use of CAD devices in the first half of the year. A total of 33 885 patients were included and 32 886 patients were ultimately analyzed. In WCH of which biopsy rate >95%, with the implementation of CAD, more the number of early gastric cancer divided by all gastric neoplasm (EGC/GN) (0.35% vs 0.59%, P=0.028, OR [95% CI]=1.65 [1.0-2.60]) was found, while gastric neoplasm detection rate (1.39% vs 1.36%, P=0.897, OR [95% CI]=0.98 [0.76-1.26]) remained stable. In RHWU of which biopsy rate <20%, the gastric neoplasm detection rate (1.78% vs 3.23%, P<0.001, OR [95% CI]=1.84 [1.33-2.54]) nearly doubled after the implementation of CAD, while there was no significant change in the EGC/GN. The application of CAD devices devoted to distinct increases in gastric neoplasm detection according to different biopsy strategies, which implied that CAD devices demonstrated assistance on gastric neoplasm detection while varied effectiveness according to different implementation scenarios.

Lung nodule localization and size estimation on chest tomosynthesis.

We aim to investigate the localization, visibility, and measurement of lung nodules in digital chest tomosynthesis (DTS). Computed tomography (CT), maximum intensity projections (CT-MIP) (transaxial versus coronal orientation), and computer-aided detection (CAD) were used as location reference, and inter- and intra-observer agreement regarding lung nodule size was assessed. Five radiologists analyzed DTS and CT images from 24 participants with lung , focusing on lung nodule localization, visibility, and measurement on DTS. Visual grading was used to compare if coronal or transaxial CT-MIP better facilitated the localization of lung nodules in DTS. The majority of the lung nodules (79%) were rated as visible in DTS, although less clearly in comparison with CT. Coronal CT-MIP was the preferred orientation in the task of locating nodules on DTS. On DTS, area-based lung nodule size estimates resulted in significantly less measurement variability when compared with nodule size estimated based on mean diameter (mD) ( ). Also, on DTS, area-based lung nodule size estimates were more accurate ( ) than lung nodule size estimates based on mean diameter ( ). Coronal CT-MIP images are superior to transaxial CT-MIP images in facilitating lung nodule localization in DTS. Most found on CT can be visualized, correctly localized, and measured in DTS, and area-based measurement may be the key to more precise and less variable nodule measurements on DTS.

Improving Prostate Image Segmentation Based on Equilibrium Optimizer and Cross-Entropy

Over the past decade, the development of computer-aided detection tools for medical image analysis has seen significant advancements. However, tasks such as the automatic differentiation of tissues or regions in medical images remain challenging. Magnetic resonance imaging (MRI) has proven valuable for early diagnosis, particularly in conditions like prostate cancer, yet it often struggles to produce high-resolution images with clearly defined boundaries. In this article, we propose a novel segmentation approach based on minimum cross-entropy thresholding using the equilibrium optimizer (MCE-EO) to enhance the visual differentiation of tissues in prostate MRI scans. To validate our method, we conducted two experiments. The first evaluated the overall performance of MCE-EO using standard grayscale benchmark images, while the second focused on a set of transaxial-cut prostate MRI scans. MCE-EO’s performance was compared against six stochastic optimization techniques. Statistical analysis of the results demonstrates that MCE-EO offers superior performance for prostate MRI segmentation, providing a more effective tool for distinguishing between various tissue types.

Increasing transparency of computer-aided detection impairs decision-making in visual search.

Recent developments in artificial intelligence (AI) have led to changes in healthcare. Government and regulatory bodies have advocated the need for transparency in AI systems with recommendations to provide users with more details about AI accuracy and how AI systems work. However, increased transparency could lead to negative outcomes if humans become overreliant on the technology. This study investigated how changes in AI transparency affected human decision-making in a medical-screening visual search task. Transparency was manipulated by either giving or withholding knowledge about the accuracy of an 'AI system'. We tested performance in seven simulated lab mammography tasks, in which observers searched for a cancer which could be correctly or incorrectly flagged by computer-aided detection (CAD) 'AI prompts'. Across tasks, the CAD systems varied in accuracy. In the 'transparent' condition, participants were told the accuracy of the CAD system, in the 'not transparent' condition, they were not. The results showed that increasing CAD transparency impaired task performance, producing an increase in false alarms, decreased sensitivity, an increase in recall rate, and a decrease in positive predictive value. Along with increasing investment in AI, this research shows that it is important to investigate how transparency of AI systems affect human decision-making. Increased transparency may lead to overtrust in AI systems, which can impact clinical outcomes.

Systematic Meta-Analysis of Computer-Aided Detection of Breast Cancer Using Hyperspectral Imaging.

The most commonly occurring cancer in the world is breast cancer with more than 500,000 cases across the world. The detection mechanism for breast cancer is endoscopist-dependent and necessitates a skilled pathologist. However, in recent years many computer-aided diagnoses (CADs) have been used to diagnose and classify breast cancer using traditional RGB images that analyze the images only in three-color channels. Nevertheless, hyperspectral imaging (HSI) is a pioneering non-destructive testing (NDT) image-processing technique that can overcome the disadvantages of traditional image processing which analyzes the images in a wide-spectrum band. Eight studies were selected for systematic diagnostic test accuracy (DTA) analysis based on the results of the Quadas-2 tool. Each of these studies' techniques is categorized according to the ethnicity of the data, the methodology employed, the wavelength that was used, the type of cancer diagnosed, and the year of publication. A Deeks' funnel chart, forest charts, and accuracy plots were created. The results were statistically insignificant, and there was no heterogeneity among these studies. The methods and wavelength bands that were used with HSI technology to detect breast cancer provided high sensitivity, specificity, and accuracy. The meta-analysis of eight studies on breast cancer diagnosis using HSI methods reported average sensitivity, specificity, and accuracy of 78%, 89%, and 87%, respectively. The highest sensitivity and accuracy were achieved with SVM (95%), while CNN methods were the most commonly used but had lower sensitivity (65.43%). Statistical analyses, including meta-regression and Deeks' funnel plots, showed no heterogeneity among the studies and highlighted the evolving performance of HSI techniques, especially after 2019.

Artificial Intelligence-Assisted Colonoscopy for Polyp Detection : A Systematic Review and Meta-analysis.

Randomized clinical trials (RCTs) of computer-aided detection (CADe) system-enhanced colonoscopy compared with conventional colonoscopy suggest increased adenoma detection rate (ADR) and decreased adenoma miss rate (AMR), but the effect on detection of advanced colorectal neoplasia (ACN) is unclear. To conduct a systematic review to compare performance of CADe-enhanced and conventional colonoscopy. Cochrane Library, Google Scholar, Ovid EMBASE, Ovid MEDLINE, PubMed, Scopus, and Web of Science Core Collection databases were searched through February 2024. Published RCTs comparing CADe-enhanced and conventional colonoscopy. Average adenoma per colonoscopy (APC) and ACN per colonoscopy were primary outcomes. Adenoma detection rate, AMR, and ACN detection rate (ACN DR) were secondary outcomes. Balancing outcomes included withdrawal time and resection of nonneoplastic polyps (NNPs). Subgroup analyses were done by neural network architecture. Forty-four RCTs with 36 201 cases were included. Computer-aided detection-enhanced colonoscopies have higher average APC (12 090 of 12 279 [0.98] vs. 9690 of 12 292 [0.78], incidence rate difference [IRD] = 0.22 [95% CI, 0.16 to 0.28]) and higher ADR (7098 of 16 253 [44.7%] vs. 5825 of 15 855 [36.7%], rate ratio [RR] = 1.21 [CI, 1.15 to 1.28]). Average ACN per colonoscopy was similar (1512 of 9296 [0.16] vs. 1392 of 9121 [0.15], IRD = 0.01 [CI, -0.01 to 0.02]), but ACN DR was higher with CADe system use (1260 of 9899 [12.7%] vs. 1119 of 9746 [11.5%], RR = 1.16 [CI, 1.02 to 1.32]). Using CADe systems resulted in resection of almost 2 extra NNPs per 10 colonoscopies and longer total withdrawal time (0.53 minutes [CI, 0.30 to 0.77]). Statistically significant heterogeneity in quality and sample size and inability to blind endoscopists to the intervention in included studies may affect the performance estimates. Computer-aided detection-enhanced colonoscopies have increased APC and detection rate but no difference in ACN per colonoscopy and a small increase in ACN DR. There is minimal increase in procedure time and no difference in performance across neural network architectures. None. (PROSPERO: CRD42023422835).