Search Results (155)

In the last two decades, there has been a considerable surge in the development of artificial intelligence. Imaging is most frequently employed for the diagnostic evaluation of patients, as it is regarded as one of the most precise methods for identifying the presence of a disease. However, a study indicates that approximately 800,000 individuals in the USA die or incur permanent disability because of misdiagnosis. The present study is based on the use of computer-aided diagnosis of colorectal cancer. The objective of this study is to develop a practical, low-cost, AI-based decision-support tool that integrates clinical test data (blood/stool) and, if needed, colonoscopy images to help reduce misdiagnosis and improve early detection of colorectal cancer for clinicians. Convolutional neural networks (CNNs) and artificial neural networks (ANNs) are utilized in conjunction with a graphical user interface (GUI), which caters to individuals lacking programming expertise. The performance of the artificial neural network (ANN) is measured using the mean squared error (MSE) metric, and the obtained performance is 7.38. For CNN, two distinct cases are under consideration: one with two outputs and one with three outputs. The precision of the models is 97.2% for RGB and 96.7% for grayscale, respectively, in the first instance, and 83% for RGB and 82% for grayscale in the second instance. However, using a pretrained network yielded superior performance with 99.5% for 2-output models and 93% for 3-output models. The GUI is composed of two panels, with the best ANN model and the best CNN model being utilized in each. The primary function of the tool is to assist medical personnel in reducing the time required to make decisions and the probability of misdiagnosis. Full article

Background: Colorectal cancer (CRC) is a major contributor to cancer mortality on a global scale, with polyps being critical precursors. The accurate classification of colorectal polyps (CRPs) from colonoscopy images is essential for the timely diagnosis and treatment of CRC. Method: This research proposes a novel hybrid model, CRP-ViT, integrating ResNet50 with Vision Transformers (ViTs) to enhance feature extraction and improve classification performance. This study conducted a comprehensive comparison of the CRP-ViT model against traditional convolutional neural networks (CNNs) and emerging quantum neural networks (QNNs). Experiments were conducted for binary classification to predict the presence of polyps and multi-classification to predict specific polyp types (hyperplastic, adenomatous, and serrated). Results: The results demonstrate that CRP_QNN-ViT achieved superior classification performance while maintaining computational efficiency. CRP_QNN-ViT achieved an accuracy of 98.18% for training and 97.73% for validation on binary classification and 98.13% during training and 97.92% for validation on multi-classification tasks. In addition to the key metrics, computational parameters were compared, where CRP_QNN-ViT excelled in computational time. Conclusions: This comparative analysis reveals the potential of integrating quantum computing into medical image analysis and underscores the effectiveness of transformer-based architectures for CRP classification. Full article

Colorectal carcinoma remains one of the primary contributors to cancer deaths; however, it is also considered a preventable type of cancer, because the prognosis of the disease is directly dependent on its timely detection. Developing accurate risk prediction models for colorectal cancer is crucial for identifying individuals at both low and high risk, as risk stratification determines the need for additional interventions, which carry their own risks. The development of new non-invasive diagnostic methods based on biomaterial analysis, alongside standard diagnostic techniques such as colonoscopy with biopsy, computed tomography scanning, and magnetic resonance imaging, can address multiple objectives: improving screening accuracy, providing a comprehensive assessment of minimal residual disease, identifying patients at a high risk of colorectal cancer, and evaluating the effectiveness of ongoing treatments. The lack of sensitive diagnostic methods drives contemporary research toward the discovery of new tools for detecting tumor cells, particularly through the examination of biological materials, including blood, exhaled air, and tumor tissue itself. In this article, we analyze current studies regarding biomarkers in colorectal cancer and prognostic significance. Full article

Background/Objectives: White light imaging (WLI) of colonoscopy has a 26% adenoma miss rate. We aimed to evaluate the effectiveness of an additional 30 s (Add-30s) observation of the right-sided colon using a novel system (EVIS X1; Olympus Co.) with texture and color enhancement imaging (TXI). Methods: We reviewed 515 patients who underwent colonoscopy with Add-30s TXI between February 2021 and December 2023 at three affiliated hospitals. After initial right-sided colon observation with WLI, the colonoscope was reinserted into the cecum, and the right-sided colon was re-observed with Add-30s TXI. Adenoma and sessile serrated lesion (SSL) detection rate (ASDR) and adenoma detection rate (ADR) were examined. Multivariate analysis identified factors influencing lesion detection using the Add-30s TXI. The difference in WLI and TXI between the novel and previous scopes was performed using propensity score matching (PSM). The efficacy of WLI with the novel system was compared to that of the previous system. Results: Among the 515 cases, Add-30s TXI observation increased right-sided ADR and ASDR by 7.4% and 9.5%, respectively. The multivariate analysis showed novel scope as an independent factor for adenoma and SSL detection (odds ratio: 2.41, p < 0.01). Right-sided ADR and ASDR for Add-30s TXI were significantly higher in the novel scope than the previous scope (ADR, 25.2% vs. 15.3%; p = 0.04; ASDR, 32.4% vs. 18.9%; p = 0.02). ASDR for WLI observation was significantly higher in the novel system than the previous system (34.8% vs. 25.9%; p < 0.01). Conclusions: Add-30s TXI significantly improved the detection of missed adenomas and SSLs in the right-sided colon. Full article

Background/Objectives: A 2-year-old male presented to the emergency department (ED) with vomiting and abdominal discomfort following ingestion of multiple magnets from a sibling’s bracelet. This case highlights the risks associated with magnet ingestion and the need for coordinated multidisciplinary care and public health intervention. Methods: Radiographs revealed magnets in the oropharynx, stomach, and small bowel. Emergency physicians coordinated care with otolaryngology, gastroenterology, and general surgery. Results: Laryngoscopy successfully removed two magnets from the uvula, and endoscopy retrieved 30 magnets from the stomach. General surgery performed a diagnostic laparoscopy, identifying residual magnets in the colon. Gastroenterology attempted a colonoscopy but was unable to retrieve magnets due to formed stool, leading to bowel preparation and serial imaging. The patient eventually passed 12 magnets per rectum without surgical intervention. Conclusions: This case emphasizes the importance of multidisciplinary collaboration in managing magnet ingestion, a preventable cause of serious gastrointestinal injury. Recent studies highlight the increasing incidence and severity of such cases due to accessibility and inadequate regulation. These findings underscore the need for public awareness and adherence to management protocols to mitigate morbidity and mortality in pediatric patients. Full article

(1) Background and Aims: Patients with inflammatory bowel disease (IBD) and primary sclerosing cholangitis (PSC) face a significantly increased risk of malignancies, including a 10-fold higher risk for colorectal cancer (CRC) and a lifetime risk for cholangiocarcinoma (CCA) exceeding 20%. The mechanisms underlying this elevated risk remain elusive. This review consolidates recent findings on cancer risk in PSC-IBD patients, focusing on molecular pathways, diagnostic innovations, and prevention strategies. (2) Methods: A comprehensive PubMed search was performed to identify studies published through to March 2025 on oncogenic processes, molecular mechanisms, and advancements in diagnostic and preventive strategies for CRC and CCA in PSC-IBD patients. (3) Results: Surveillance guidelines recommend an annual colonoscopy for CRC and imaging combined with CA 19-9 monitoring for CCA. Recent studies highlight the role of molecular alterations, including epigenetic modifications, in tumorigenesis. Advances in molecular diagnostics, imaging, and endoscopic technologies are improving the accuracy and timeliness of cancer detection. (4) Conclusions: PSC-IBD patients remain at high risk for CRC and CCA, emphasizing the need for vigilant surveillance and advanced prevention strategies. Advances in early detection and precision diagnostics offer new opportunities to reduce the cancer burden in this high-risk population. Full article

Background: Synthetic and pseudosynthetic images can be used to extend colonoscopy datasets, which, in turn, are used to train AI-detection models, yet their clinical acceptability depends on whether medical professionals can still recognize non-real content. Aim: To quantify the ability of practicing gastroenterologists to discriminate real, pseudosynthetic, and synthetic polyp images and to determine how training level and synthesis method impact detection. Materials and Methods: A total of 32 Romanian gastroenterologists (18 residents and 14 seniors) reviewed 24 images (8 real, 8 augmented, 4 CycleGAN, and 4 diffusion) via an online form. Classification accuracy, 95% confidence intervals (CI), class sensitivity and precision, 3 × 3 confusion matrices, and Fleiss’ κ were calculated. Resident vs. senior differences were tested with Pearson χ²; CycleGAN versus diffusion detectability was analyzed with the Wilcoxon signed-rank test (α = 0.05). Results: Overall accuracy was 61.2% (95% CI 57.7–64.6). Residents and seniors performed similarly (62.3% vs. 59.8%; χ²₁ = 0.38, p = 0.54). Sensitivity/precision were 70.7%/62.2% for real, 51.6%/58.9% for augmented, and 61.3%/62.1% for synthetic images. Collapsing to “real vs. non-real” yielded 70.7% sensitivity and 78.5% specificity for real images. CycleGAN images were always recognized as synthetic (128/128; 97.1–100% CI), whereas diffusion images were correctly classified only 22.7% of the time (16.3–30.6%; Wilcoxon p < 0.001). The training level did not impact detection performance (χ²₂ < 1.2, p > 0.5). Inter-rater agreement was fair (κ = 0.30, 95% CI 0.15–0.43). Conclusions: Clinicians detect non-real colonoscopy images only slightly above chance, irrespective of experience. The diffusion synthesis method creates images that escape human scrutiny, suggesting the need for automated authenticity safeguards before synthetic datasets are applied in clinical or AI-validation contexts. Full article

Rectal cancer management necessitates a rigorous multidisciplinary strategy, emphasizing precise staging and detailed risk stratification to inform optimal therapeutic decision-making. Obtaining an accurate histological diagnosis before initiating treatment is essential. Comprehensive staging integrates clinical evaluation, thorough medical history analysis, assessment of carcinoembryonic antigen (CEA) levels, and computed tomography (CT) imaging of the abdomen and thorax. High-resolution pelvic magnetic resonance imaging (MRI), utilizing dedicated rectal protocols, is critical for identifying recurrence risks and delineating precise anatomical relationships. Endoscopic ultrasound further refines staging accuracy by determining the tumor infiltration depth in early-stage cancers, while preoperative colonoscopy effectively identifies synchronous colorectal lesions. In early-stage rectal cancers (T1–T2, N0, and M0), radical surgical resection remains the standard of care, although transanal local excision can be selectively indicated for certain T1N0 tumors. In contrast, locally advanced rectal cancers (T3, T4, and N+) characterized by microsatellite stability or proficient mismatch repair are optimally managed with total neoadjuvant therapy (TNT), which combines chemoradiotherapy with oxaliplatin-based systemic chemotherapy. Additionally, tumors exhibiting high microsatellite instability or mismatch repair deficiency respond favorably to immune checkpoint inhibitors (ICIs). The evaluation of tumor response following neoadjuvant therapy, utilizing MRI and endoscopic assessments, facilitates individualized treatment planning, including non-operative approaches for patients with confirmed complete clinical responses who comply with rigorous follow-up. Recent advancements in molecular characterization, targeted therapies, and immunotherapy highlight a significant evolution towards personalized medicine. The effective integration of these innovations requires enhanced interdisciplinary collaboration to improve patient prognosis and quality of life. Full article

Background/Objectives: A new LED endoscopy system featuring advanced noise-reduction technology, the EP-8000 with the EC-860ZP colonoscope (Fujifilm), was introduced in 2024. We evaluated the improvements in colonoscopic image quality of this system, comparing it with a previous system/scope (VP-7000/EC-760ZP). Methods: This is a multicenter, observational study. From January 2024 to February 2025, 150 patients undergoing colonoscopy at two institutions were enrolled. Images of the cecum and lesions were captured using white light imaging (WLI), blue light imaging (BLI), and linked color imaging (LCI) under similar conditions. Participants were divided into three groups: Group 1 (EP-8000+EC-860ZP; 50 cases), Group 2 (EP-8000+EC-760ZP; 50 cases), and Group 3 (VP-7000+EC-760ZP; 50 cases). Cecal and lesion images were evaluated for brightness, halation, and visibility using a four-point scale (1 = poor to 4 = excellent) by endoscopists and original values by image-analysis software. Results: In cecal images, the endoscopists’ scores in Group 1 were significantly better than in Group 3 for brightness (WLI: 3.71 ± 0.55 vs. 3.51 ± 0.58, p < 0.001, BLI: 3.15 ± 0.85 vs. 2.23 ± 0.92, p < 0.001; LCI: 3.83 ± 0.42 vs. 3.54 ± 0.58, p < 0.001) and for halation (WLI: 3.60 ± 0.51 vs. 3.18 ± 0.59, p < 0.001, BLI: 2.99 ± 0.69 vs. 2.71 ± 0.78, p < 0.001; LCI: 3.33 ± 0.60 vs. 3.10 ± 0.58, p < 0.001). Software analysis confirmed that Group 1 had superior brightness values compared with Group 3 for WLI, BLI, and LCI, as well as lower halation values for WLI and LCI. Regarding lesion images, brightness, halation, and visibility for WLI, BLI, and LCI were superior in Group 1 than in Group 3. Conclusions: The new LED system provided improvements in brightness, halation, and lesion visibility. Full article

Inflammatory bowel diseases are chronic diseases of the gastrointestinal tract with a growing prevalence worldwide, requiring precise diagnostic and monitoring methods to guide their appropriate treatment. In this context, MRE (Magnetic Resonance Enterography) has become an essential imaging technique as a non-invasive option for the diagnosis of Crohn’s disease and ulcerative colitis in recent years. This method provides detailed information about intestinal inflammation, disease activity, complications, and response to therapy, without the need to expose the patient to ionizing radiation. This study analyzes the advantages of MRE over other imaging methods, as well as its clinical applicability and current challenges. We also discuss future perspectives, including the integration of artificial intelligence and the optimization of protocols for better diagnostic accuracy. Full article

Artificial intelligence (AI) is rapidly transforming imaging modalities in inflammatory bowel disease (IBD), particularly in endoscopy and ultrasound. Despite their critical roles, both modalities are challenged by interobserver variability, subjectivity, and accessibility issues. AI offers significant potential to address these limitations by enhancing lesion detection, standardizing disease activity scoring, and supporting clinical decision-making. In endoscopy, deep convolutional neural networks have achieved high accuracy in detecting mucosal abnormalities and grading disease severity, reducing observer dependency and improving diagnostic consistency. AI-assisted colonoscopy systems have also demonstrated improvements in procedural quality metrics, including adenoma detection rates and withdrawal times. Similarly, AI applications in intestinal ultrasound show promise in automating measurements of bowel wall thickness, assessing vascularity, and distinguishing between inflammatory and fibrotic strictures, which are critical for tailored therapy decisions. Video capsule endoscopy has likewise benefited from AI, reducing interpretation times and enhancing the detection of subtle lesions. Despite these advancements, implementation challenges, including dataset quality, standardization, AI interpretability, clinician acceptance, and regulatory and ethical considerations, must be carefully addressed. The current review focuses on the most recent developments in the integration of AI into experimental designs, medical devices, and clinical workflows for optimizing diagnostic accuracy, treatment strategies, and patient outcomes in IBD management. Full article

Objective: Although prior research developed an artificial intelligence (AI)-based classification system predicting colorectal lesion histology, the heavy computational demands limited its practical application. Recent advancements in medical AI emphasize decentralized architectures using edge computing devices, enhancing accessibility and real-time performance. This study aims to construct and evaluate a deep learning-based colonoscopy image classification model for automatic histologic categorization for real-time use on edge computing hardware. Design: We retrospectively collected 2418 colonoscopic images, subsequently dividing them into training, validation, and internal test datasets at a ratio of 8:1:1. Primary evaluation metrics included (1) classification accuracy across four histologic categories (advanced colorectal cancer, early cancer/high-grade dysplasia, tubular adenoma, and nonneoplasm) and (2) binary classification accuracy differentiating neoplastic from nonneoplastic lesions. Additionally, an external test was conducted using an independent dataset of 269 colonoscopic images. Results: For the internal-test dataset, the model achieved an accuracy of 83.5% (95% confidence interval: 78.8–88.2%) for the four-category classification. In binary classification (neoplasm vs. nonneoplasm), accuracy improved significantly to 94.6% (91.8–97.4%). The external test demonstrated an accuracy of 82.9% (78.4–87.4%) in the four-category task and a notably higher accuracy of 95.5% (93.0–98.0%) for binary classification. The inference speed of lesion classification was notably rapid, ranging from 2–3 ms/frame in GPU mode to 5–6 ms/frame in CPU mode. During real-time colonoscopy examinations, expert endoscopists reported no noticeable latency or interference from AI model integration. Conclusions: This study successfully demonstrates the feasibility of a deep learning-powered colonoscopy image classification system designed for the rapid, real-time histologic categorization of colorectal lesions on edge computing platforms. This study highlights how nature-inspired frameworks can improve the diagnostic capacities of medical AI systems by aligning technological improvements with biomimetic concepts. Full article

Perivascular epithelioid cell tumors (PEComas) are rare mesenchymal neoplasms characterized by perivascular epithelioid cell proliferation. They can occur in various organs, but colonic PEComas are exceptionally rare, showing diagnostic challenges due to their nonspecific clinical presentation and similar features to those of other colorectal tumors. We present a case of a 61-year-old female with defecation accompanied by blood clots, initially diagnosed with a suspected tumor in the sigmoid colon. Despite initial biopsy yielding non-informative material, repeat colonoscopy and imaging studies revealed a malignant tumor with multinucleated giant (osteoclast-like) cells and probable p53 mutation, most likely of mesenchymal origin. Robotic surgical resection was performed, and ultimately pathological examination refined the diagnosis as a malignant PEComa of the colon. This case demonstrates the importance of considering PEComa in the differential diagnosis of colonic tumors. Further research is needed to ascertain the clinical behavior and optimal treatment for colonic PEComas. Full article

Background/Objectives: Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality worldwide, emphasizing the critical need for the accurate classification of precancerous polyps. This research presents an extensive analysis of the multiclassification framework leveraging various deep learning (DL) architectures for the automated classification of colorectal polyps from colonoscopy images. Methods: The proposed methodology integrates real-time data for training and utilizes a publicly available dataset for testing, ensuring generalizability. The real-time images were cautiously annotated and verified by a panel of experts, including post-graduate medical doctors and gastroenterology specialists. The DL models were designed to categorize the preprocessed colonoscopy images into four clinically significant classes: hyperplastic, serrated, adenoma, and normal. A suite of state-of-the-art models, including VGG16, VGG19, ResNet50, DenseNet121, EfficientNetV2, InceptionNetV3, Vision Transformer (ViT), and the custom-developed CRP-ViT, were trained and rigorously evaluated for this task. Results: Notably, the CRP-ViT model exhibited superior capability in capturing intricate features, achieving an impressive accuracy of 97.28% during training and 96.02% during validation with real-time images. Furthermore, the model demonstrated remarkable performance during testing on the public dataset, attaining an accuracy of 95.69%. To facilitate real-time interaction and clinical applicability, a user-friendly interface was developed using Gradio, allowing healthcare professionals to upload colonoscopy images and receive instant classification results. Conclusions: The CRP-ViT model effectively predicts and categorizes colonoscopy images into clinically relevant classes, aiding gastroenterologists in decision-making. This study highlights the potential of integrating AI-driven models into routine clinical practice to improve colorectal cancer screening outcomes and reduce diagnostic variability. Full article

The automatic segmentation technique for colorectal polyps in colonoscopy is considered critical for aiding physicians in real-time lesion identification and minimizing diagnostic errors such as false positives and missed lesions. Despite significant progress in existing research, accurate segmentation of colorectal polyps remains technically challenging due to persistent issues such as low contrast between polyps and mucosa, significant morphological heterogeneity, and susceptibility to imaging artifacts caused by bubbles in the colorectal lumen and poor lighting conditions. To address these limitations, this study proposed a novel pyramid vision transformer-based hierarchical path aggregation network (HPANet) for polyp segmentation. Specifically, firstly, the backward multi-scale feature fusion module (BMFM) was developed to enhance the ability of processing polyps with different scales. Secondly, the forward noise reduction module (FNRM) was designed to learn the texture features of the upper and lower layers to reduce the influence of noise such as bubbles. Finally, in order to solve the problem of boundary ambiguity caused by repeated up and down sampling, the boundary feature refinement module (BFRM) was developed to further refine the boundary. The proposed network was compared with several representative networks on five public polyp datasets. Experimental results show that the proposed network achieves better segmentation performance, especially on the Kvasir SEG dataset, where the mDice and mIoU coefficients reach 0.9204 and 0.8655. Full article