Search Results (105)

Background: Gastrointestinal (GI) disorders present significant healthcare challenges, requiring rapid, accurate, and effective diagnostic methods to improve treatment outcomes and prevent complications. Wireless capsule endoscopy (WCE) is an effective tool for diagnosing GI abnormalities; however, precisely identifying diverse lesions with similar visual patterns remains difficult. Methods: Many existing computer-aided diagnostic (CAD) systems rely on manually crafted features or single deep learning (DL) models, which often fail to capture the complex and varied characteristics of GI diseases. In this study, we proposed “EndoNet,” a multi-stage hybrid DL framework for eight-class GI disease classification using WCE images. Features were extracted from two different layers of three pre-trained convolutional neural networks (CNNs) (Inception, Xception, ResNet101), with both inter-layer and inter-model feature fusion performed. Dimensionality reduction was achieved using Non-Negative Matrix Factorization (NNMF), followed by selection of the most informative features via the Minimum Redundancy Maximum Relevance (mRMR) method. Results: Two datasets were used to evaluate the performance of EndoNer, including Kvasir v2 and HyperKvasir. Classification using seven different Machine Learning algorithms achieved a maximum accuracy of 97.8% and 98.4% for Kvasir v2 and HyperKvasir datasets, respectively. Conclusions: By integrating transfer learning with feature engineering, dimensionality reduction, and feature selection, EndoNet provides high accuracy, flexibility, and interpretability. This framework offers a powerful and generalizable artificial intelligence solution suitable for clinical decision support systems. Full article

Ulcerative colitis, polyps, esophagitis, and other gastrointestinal (GI) diseases significantly impact health, making early detection crucial for reducing mortality rates and improving patient outcomes. Traditional white light imaging (WLI) is commonly used during endoscopy to identify abnormalities in the gastrointestinal tract. However, insufficient contrast often limits its effectiveness, making it challenging to distinguish between healthy and unhealthy tissues, particularly when identifying subtle mucosal and vascular abnormalities. These limitations have prompted the need for more advanced imaging techniques that enhance pathological visualization and facilitate early diagnosis. Therefore, this study investigates the integration of the Spectrum-Aided Vision Enhancer (SAVE) mechanism to improve WLI images and increase disease detection accuracy. This approach transforms standard WLI images into hyperspectral imaging (HSI) representations, creating narrow-band imaging (NBI-like) visuals with enhanced contrast and tissue differentiation, thereby improving the visualization of vascular and mucosal structures critical for diagnosing GI disorders. This transformation allows for a clearer representation of blood vessels and membrane formations, which is essential for determining the presence of GI diseases. The dataset for this study comprises WLI images alongside SAVE-enhanced images, including four categories: ulcerative colitis, polyps, esophagitis, and healthy GI tissue. These images are organized into training, validation, and test sets to develop a deep learning-based classification model. Utilizing principal component analysis (PCA) and multiple regression analysis for spectral standardization ensures that the improved images retain spectral characteristics that are vital for clinical applications. By merging deep learning techniques with advanced imaging enhancements, this study aims to create an artificial intelligence (AI)–driven diagnostic system capable of early and accurate detection of GI diseases. InceptionV3 attained an overall accuracy of 94% in both scenarios; SAVE produced a modest enhancement in the ulcerative colitis F1-score from 92% to 93%, while the F1-scores for other classes exceeded 96%. SAVE resulted in a 10% increase in YOLOv8x accuracy, reaching 89%, with ulcerative colitis F1 improving to 82% and polyp F1 rising to 76%. VGG16 enhanced accuracy from 85% to 91%, and the F1-score for polyps improved from 68% to 81%. These findings confirm that SAVE enhancement consistently improves disease classification across diverse architectures, offers a practical, hardware-independent approach to hyperspectral-quality images, and enhances the accuracy of gastrointestinal screening. Furthermore, this research seeks to provide a practical and effective solution for clinical applications, improving diagnostic accuracy and facilitating superior patient care. Full article

Timely and accurate diagnosis of gastrointestinal diseases (GIDs) remains a critical bottleneck in clinical endoscopy, particularly due to the limited contrast and sensitivity of conventional white light imaging (WLI) in detecting early-stage mucosal abnormalities. To overcome this, this research presents Spectrum Aided Vision Enhancer (SAVE), an innovative, software-driven framework that transforms standard WLI into high-fidelity hyperspectral imaging (HSI) and simulated narrow-band imaging (NBI) without any hardware modification. SAVE leverages advanced spectral reconstruction techniques, including Macbeth Color Checker-based calibration, principal component analysis (PCA), and multivariate polynomial regression, achieving a root mean square error (RMSE) of 0.056 and structural similarity index (SSIM) exceeding 90%. Trained and validated on the Kvasir v2 dataset (n = 6490) using deep learning models like ResNet-50, ResNet-101, EfficientNet-B2, both EfficientNet-B5 and EfficientNetV2-B0 were used to assess diagnostic performance across six key GI conditions. Results demonstrated that SAVE enhanced imagery and consistently outperformed raw WLI across precision, recall, and F1-score metrics, with EfficientNet-B2 and EfficientNetV2-B0 achieving the highest classification accuracy. Notably, this performance gain was achieved without the need for specialized imaging hardware. These findings highlight SAVE as a transformative solution for augmenting GI diagnostics, with the potential to significantly improve early detection, streamline clinical workflows, and broaden access to advanced imaging especially in resource constrained settings. Full article

The neural network-based classification of endoscopy images plays a key role in diagnosing gastrointestinal diseases. However, current models for estimating ulcerative colitis (UC) severity still lack high performance, highlighting the need for more advanced and accurate solutions. This study aims to apply a state-of-the-art hybrid neural network architecture—combining convolutional neural networks (CNNs) and transformer models—to classify intestinal endoscopy images, utilizing the largest publicly available annotated UC dataset. A 10-fold cross-validation is performed on the LIMUC dataset using CoAtNet models, combined with the Class Distance Weighted Cross-Entropy (CDW-CE) loss function. The best model is compared against pure CNN and transformer baselines by evaluating performance metrics, including quadratically weighted kappa (QWK) and macro F1, for full Mayo score classification, and kappa and F1 scores for remission classification. The CoAtNet models outperformed both pure CNN and transformer models. The most effective model, CoAtNet_2, improved classification accuracy by $1.76\%$ and QWK by $1.46\%$ over the previous state-of-the-art models on the LIMUC dataset. Other metrics, including F1 score, also showed clear improvements. Experiments show that the CoAtNet model, which integrates convolutional and transformer components, improves UC assessment from endoscopic images, enhancing AI’s role in computer-aided diagnosis. Full article

Background/Objectives: Gastrointestinal diseases (GID), such as oesophagitis, polyps, and ulcerative colitis, contribute significantly to global morbidity and mortality. Conventional diagnostic methods employing white light imaging (WLI) in wireless capsule endoscopy (WCE) provide limited spectrum information, therefore influencing classification performance. Methods: A new technique called Spectrum Aided Vision Enhancer (SAVE), which converts traditional WLI images into hyperspectral imaging (HSI)-like representations, hence improving diagnostic accuracy. HSI involves the acquisition of image data across numerous wavelengths of light, extending beyond the visible spectrum, to deliver comprehensive information regarding the material composition and attributes of the imaged objects. This technique facilitates improved tissue characterisation, rendering it especially effective for identifying abnormalities in medical imaging. Using a carefully selected dataset consisting of 6000 annotated photos taken from the KVASIR and ETIS-Larib Polyp Database, this work classifies normal, ulcers, polyps, and oesophagitis. The performance of both the original WLI and SAVE transformed images was assessed using advanced deep learning architectures. The principal outcome was the overall classification accuracy for normal, ulcer, polyp, and oesophagitis categories, contrasting SAVE-enhanced images with standard WLI across five deep learning models. Results: The principal outcome of this study was the enhancement of diagnostic accuracy for gastrointestinal disease classification, assessed through classification accuracy, precision, recall, and F1 score. The findings illustrate the efficacy of the SAVE method in improving diagnostic performance without requiring specialised equipment. With the best accuracy of 98% attained using EfficientNetB7, compared to 97% with WLI, experimental data show that SAVE greatly increases classification metrics across all models. With relative improvement from 85% (WLI) to 92% (SAVE), VGG16 showed the highest accuracy. Conclusions: These results confirm that the SAVE algorithm significantly improves the early identification and classification of GID, therefore providing a potential development towards more accurate, non-invasive GID diagnostics with WCE. Full article

Background/Objectives: Pediatric inflammatory bowel disease (pIBD), including Crohn’s disease (CD), ulcerative colitis (UC), and IBD-unclassified (IBD-U), exhibits unique clinical features compared to adult-onset disease. This study aimed to describe phenotypic characteristics of pIBD in the north-west region of Romania over a 21-year period and to compare our findings with those of other studies worldwide. Methods: We conducted a retrospective study of children under 18 years of age, from the north-west region of Romania, diagnosed with pIBD between 2000 and 2020 at the Emergency Clinical Hospital for Children, Cluj-Napoca. Disease phenotype at diagnosis was established according to the Paris classification. Data were collected from the hospital records and analyzed using descriptive statistics and univariate analysis of categorical variables. A p-value < 0.05 was considered statistically significant. Results: Ninety-four patients were included (CD: 51.0%; UC: 43.6%; IBD-U: 5.4%), with a median age at diagnosis of 14 years (11–15.7). Very early-onset IBD accounted for 5.3% of cases. The likelihood of being diagnosed with CD after 10 years of age was significantly higher compared to UC (OR = 4.75, 95% CI: 1.10–29.07, p = 0.03). UC most frequently presented as pancolitis (51.2%), while CD most often involved the ileocolonic region (56.3%). Inflammatory behavior was the most common CD phenotype (69%). Upper gastrointestinal involvement was documented in 18.7% of CD cases, with detection rates increasing after 2014. Perianal disease and growth impairment were significantly associated with complicated CD behavior (p = 0.03, and p = 0.007 respectively). Our findings are broadly consistent with other published reports. Conclusions: This study provides the first detailed phenotypic characterization of pIBD in this region. Our findings reflect trends observed in other populations and underscore the importance of standardized diagnostic evaluation. Full article

Inflammatory bowel disease (IBD) is a chronic inflammatory condition of the gastrointestinal tract characterized by the deregulation of immuno-oncology markers. IBD includes ulcerative colitis and Crohn’s disease. Chronic active inflammation is a risk factor for the development of colorectal cancer (CRC). This technical note describes a dataset of histological images of ulcerative colitis, CRC (adenocarcinoma), and colon control. The samples were stained with hematoxylin and eosin (H&E), and immunohistochemically analyzed for LAIR1 and TOX2 markers. The methods used for collecting, processing, and analyzing scientific data, including this dataset, using convolutional neural networks (CNNs) and information about the dataset’s use are also described. This article is a companion to the manuscript “Ulcerative Colitis, LAIR1 and TOX2 Expression, and Colorectal Cancer Deep Learning Image Classification Using Convolutional Neural Networks”. Full article

Gastrointestinal graft-versus-host disease (GvHD) is a frequent and severe complication after allogeneic stem cell transplantation (aSCTx). Although biopsy and histopathology remain the gold standard for diagnosis of GvHD, this approach can be limited by thrombocytopenia accompanying aSCTx and the diagnostic delay associated with routine histopathology. Here, we report on two patients in which dye-based contact microscopy using a latest generation endocytoscope with 520-fold magnification enabled in vivo diagnosis of GvHD. The first patient was a 23-year-old man with acute lymphoblastic leukemia presenting with non-bloody diarrhea 3 months after aSCTx. After topical staining with crystal violet and methylene blue, endocytoscopy in the rectum showed several apoptotic epithelial cells. Histopathology confirmed GvHD grade III according to the Lerner classification. The second patient was a 59-year-old female with diarrhea 3 months after aSCTx. Apart from pathognomic apoptotic bodies, EC additionally revealed crypt lumina enlargement and mononuclear cell infiltrates in the lamina propria with subsequent crypt distension. The duration of the procedure was less than 5 min in each patient. These findings illustrate that in vivo microscopy using endocytoscopy can enable instantaneous diagnosis of GvHD with the benefit of accelerating therapeutic decisions in patients with suspected severe GvHD. Full article

Background: Chronic rhinosinusitis with or without nasal polyps (CRSwNPs/CRSsNPs) is an inflammatory disease that is becoming increasingly associated with laryngopharyngeal reflux disease (LPRD). Although symptom-based questionnaires, such as the Reflux Symptom Index (RSI) and Reflux Symptom Score (RSS), are widely used, there is a lack of objective endoscopic tools for assessing the nasopharyngeal and nasal manifestations of reflux. The Nasopharyngeal Reflux Endoscopic Score (NRES) is a novel endoscopic scoring system that was developed to address this issue. Objective: The objective of this study was to evaluate the diagnostic accuracy of the NRES in identifying LPRD in patients with CRS, compared with a clinical reference standard. Methods: A prospective diagnostic accuracy cohort study was conducted at two tertiary care centers in Astana, Kazakhstan, from September 2023 to February 2025. A total of 216 adults were enrolled and divided into three groups: CRS with suspected LPRD (n = 116), CRS without LPRD (n = 69), and healthy controls (n = 31). CRS was diagnosed according to the EPOS 2020 criteria. LPRD was defined using a composite reference standard comprising clinical assessment, RSS > 13, RSI, and selective 24 h pH monitoring and gastrointestinal endoscopy. All participants underwent nasopharyngeal and laryngeal endoscopy, with NRES, L-K, RFS, RSI, and RSS assessments at baseline and at 6 and 12 months. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance, and Wilcoxon tests were used to analyze the changes in scores. Correlation and regression analyses were used to explore associations between scales and predictive factors. Results: At baseline, NRES scores were significantly higher in the CRS with LPRD group (mean: 11.59) than in the CRS without LPRD group (mean: 3.10) and the healthy control group (mean: 2.16) (p < 0.001). ROC analysis demonstrated excellent diagnostic accuracy, with an area under the curve (AUC) of 0.998 (95% confidence interval (CI): 0.994–1.000), a sensitivity of 98% (95% CI: 94–100%) and a specificity of 96% (95% CI: 91–99%) at an optimal cut-off point of 8.5. NRES scores showed strong correlations with RSI, RSS, and RFS scores (r > 0.76, p < 0.001). A longitudinal assessment revealed significant reductions in all scores after treatment with proton pump inhibitors and lifestyle modifications, with sustained improvement at 12 months. Regression analysis found no significant effect of age, gender, or GERD severity (LA classification) on NRES scores. Conclusions: The NRES is a highly sensitive and specific endoscopic tool for identifying nasopharyngeal changes associated with LPRD in CRS patients. It demonstrates strong correlations with established symptom-based and laryngoscopic reflux assessments and responds to anti-reflux therapy over time. The NRES may, therefore, be a valuable objective adjunct in the comprehensive evaluation and longitudinal monitoring of LPRD-associated CRS. Full article

This study presents an enhanced version of the StarGAN model, with a focus on medical applications, particularly endoscopic image augmentation. Our model incorporates novel Perspective Transformation and Viewpoint Attention Modules for StarGAN that improve image classification accuracy in a multiclass classification task. The Perspective Transformation Module enables the generation of more diverse viewing angles, while the Viewpoint Attention Module helps focus on diagnostically significant regions. We evaluate the performance of our enhanced architecture using the Kvasir v2 dataset, which contains 8000 images across eight gastrointestinal disease classes, comparing it against baseline models including VGG-16, ResNet-50, DenseNet-121, InceptionNet-V3, and EfficientNet-B7. Experimental results demonstrate that our approach achieves better performance in all models for this eight-class classification problem, increasing accuracy on average by 0.7% on VGG-16 and 0.63% on EfficientNet-B7 models. The addition of perspective transformation capabilities enables more diverse examples to augment the database and provide more samples of specific illnesses. Our approach offers a promising solution for medical image generation, enabling effective training with fewer data samples, which is particularly valuable in medical model development where data are often scarce due to challenges in acquisition. These improvements demonstrate significant potential for advancing machine learning disease classification systems in gastroenterology and medical image augmentation as a whole. Full article

Spondyloarthropathies (SpAs) represent a diverse group of seronegative immune-mediated inflammatory diseases characterized by a genetic predisposition and an association with human leukocyte antigen-B27. This narrative review aims to explore juvenile spondyloarthropathies (JSpAs), their classification, clinical manifestations, diagnostic challenges, and contemporary treatment strategies. According to the International League of Associations for Rheumatology criteria, JSpAs include several specific forms: enthesitis-related arthritis, psoriatic arthritis, and undifferentiated arthritis. Despite established classifications, the terms and definitions surrounding these conditions can often lead to confusion among healthcare professionals. This ambiguity underscores the need for a standardized approach to nosological classification. The clinical presentation of JSpAs can be multifaceted, encompassing both articular and extra-articular manifestations. Articular symptoms may include enthesitis and varying forms of arthritis, while extra-articular involvement can range from uveitis to gastrointestinal, cardiovascular, pulmonary, neurological, and renal complications. These diverse manifestations highlight the systemic nature of the disease and the importance of a holistic approach to diagnosis and treatment. While laboratory tests for SpAs are often non-specific, imaging modalities such as musculoskeletal ultrasound and magnetic resonance imaging play a crucial role in the early detection of inflammatory lesions. These imaging techniques can provide valuable insights into disease progression and aid in the formulation of appropriate treatment plans. Current treatment guidelines advocate for a “stepwise” approach to therapy, beginning with nonsteroidal anti-inflammatory drugs and progressing to glucocorticoids, disease-modifying antirheumatic drugs, and biological agents, particularly anti-tumor necrosis factor alpha agents. The primary objective of treatment is to achieve clinical remission or, at a minimum, to attain low disease activity. Regular monitoring of disease activity is imperative; however, the lack of validated assessment tools for the pediatric population remains a significant challenge. JSpAs pose unique challenges in terms of diagnosis and management due to their diverse manifestations and the complexities of their classification. Ongoing research and clinical efforts are essential to refine our understanding of these conditions, improve treatment outcomes, and enhance quality of life for affected children and their families. Effective management hinges on early detection, individualized treatment plans, and continuous monitoring, ensuring that patients receive the most appropriate care tailored to their specific needs. Full article

Plant-derived small RNAs (sRNAs) have garnered significant attention in nucleic acid therapeutics, driven by their distinctive cross-kingdom regulatory capabilities and extensive therapeutic promise. These sRNAs exhibit a wide range of pharmacological effects, including pulmonary protection, antiviral, anti-inflammatory, and antitumor activities, underscoring their substantial potential for clinical translation. A key advantage lies in their delivery, facilitated by plant-specific nanovesicular carriers—such as plant exosomes, herbal decoctosomes, and bencaosomes—which protect sRNAs from gastrointestinal degradation and enable precise, tissue-specific targeting. This review provides a comprehensive analysis of plant-derived sRNAs, detailing their classification, gene-silencing mechanisms, and nanovesicle-mediated cross-kingdom delivery strategies. It further explores their therapeutic potential and underlying molecular mechanisms in major human diseases. Additionally, we critically evaluate current technical challenges and propose future directions to advance the development of plant-derived sRNAs for precision therapeutics. This work aims to offer a robust theoretical framework and practical guidance for the clinical advancement of plant-derived sRNA-based therapies. Full article

Inflammatory bowel diseases (IBDs), comprising Crohn’s disease (CD) and ulcerative colitis (UC), are chronic immune-mediated inflammatory diseases of the gastrointestinal (GI) tract with still-elusive etiopathogeneses and an increasing prevalence worldwide. Despite the growing availability of more advanced therapies in the last two decades, there are still a number of unmet needs. For example, the achievement of mucosal healing has been widely demonstrated as a prognostic marker for better outcomes and a reduced risk of dysplasia and cancer; however, the accuracy of endoscopy is crucial for both this aim and the precise and reproducible evaluation of endoscopic activity and the detection of dysplasia. Artificial intelligence (AI) has drastically altered the field of GI studies and is being extensively applied to medical imaging. The utilization of deep learning and pattern recognition can help the operator optimize image classification and lesion segmentation, detect early mucosal abnormalities, and eventually reveal and uncover novel biomarkers with biologic and prognostic value. The role of AI in endoscopy—and potentially also in histology and imaging in the context of IBD—is still at its initial stages but shows promising characteristics that could lead to a better understanding of the complexity and heterogeneity of IBDs, with potential improvements in patient care and outcomes. The initial experience with AI in IBDs has shown its potential value in the differentiation of UC and CD when there is no ileal involvement, reducing the significant amount of time it takes to review videos of capsule endoscopy and improving the inter- and intra-observer variability in endoscopy reports and scoring. In addition, these initial experiences revealed the ability to predict the histologic score index and the presence of dysplasia. Thus, the purpose of this review was to summarize recent advances regarding the application of AI in IBD endoscopy as there is, indeed, increasing evidence suggesting that the integration of AI-based clinical tools will play a crucial role in paving the road to precision medicine in IBDs. Full article

Inflammatory bowel disease (IBD) in the pediatric population presents distinct characteristics compared to adult cases. Pathology plays a critical role in its diagnosis, and this review underscores key considerations in the pathologic evaluation of pediatric IBD. Recognizing inflammatory patterns in the upper gastrointestinal tract can improve disease classification and aid in diagnosing IBD in certain scenarios, such as isolated upper gastrointestinal or small bowel involvement. Additionally, familiarity with distinctive subtypes, including IBD associated with primary sclerosing cholangitis and monogenic forms of IBD, supports early comorbidity detection, enhances patient management, and improves prognostication. Full article

Objective: The objective was to evaluate the impact of perioperative immunonutrition (IN) on postoperative complications in patients undergoing radical cystectomy (RC) for bladder cancer (BC). Methods: A prospective case series of 19 patients treated with perioperative IN between October 2022 and July 2023 was conducted. Patients received preoperative IN based on nutritional risk and postoperative IN with gradual recovery of normal feeding. The inclusion criteria encompassed clinically node-negative patients without metastatic disease. The outcomes were assessed using Clavien–Dindo classification and included infectious complications, wound healing disorders, ileus, anemia, genitourinary issues, recovery time, and compliance with the nutritional regimen. Results: Sixteen patients (84.2%) experienced complications. Most were low-grade (CD 1–2), with no CD > 3a. Wound disorders affected 10.5% and anemia requiring transfusion occurred in 47.4% of patients, infectious complications were reported in 26.3%, and ileus in 36.8%. The median time to first flatus was 2 days (IQR 2–3), while resumption of oral feeding occurred after 4 days (IQR 2–5), like mobilization (IQR 2–5). The median hospital stay was 14 days (IQR 11–18). Compliance with IN was 78.9%, with gastrointestinal intolerance being the primary cause of discontinuation. Conclusions: Patients with RC undergoing perioperative IN showed low rates of high-grade complications and promising results in bowel function recovery and infection rates. Further randomized controlled trials are required to validate these results. Full article