Background/Objectives: Alzheimer's disease (AD) is a progressive neurodegenerative disorder necessitating accurate and timely diagnosis for effective clinical intervention. While deep learning methods have shown promise in AD classification, many rely on computationally intensive architectures and high-dimensional feature representations. This study introduces a lightweight hybrid framework combining deep feature extraction, dimensionality reduction, and adaptive classification for MRI-based Alzheimer's disease stage classification. Methods: Utilizing MRI images from a publicly available Alzheimer's disease dataset encompassing four clinical stages (Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented), deep features were extracted using a pre-trained SqueezeNet model as a fixed feature extractor, generating 1000-dimensional feature vectors. Due to the computational complexity and for the improvement of the model efficiency, the dimensionality reduction technique, Principal Component Analysis (PCA) was then applied. This resulted in an optimum representation of 100 principal components, retaining about 96% of the variance. Then, the performances of various machine learning classifiers such as k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Decision Tree (DT), Neural Network (NN), Naïve Bayes (NB), Logistic Regression (LR) and Enhanced Fuzzy Min-Max Neural Network (EFMM) were tested. The accuracy, precision, recall, F1-score, area under the receiver operating characteristic curve (AUC), and confusion matrices were used to evaluate the performance. Stratified 5-fold cross validation was used to ensure the strength of our results. Results: The findings show that PCA has a significant improvement in classification accuracy for most of the models. In particular, the EFMM classifier outperformed the other classifiers, with an accuracy of 97.19% on the independent test set. After PCA, the AUC values for classes such as Mild Demented, Moderate Demented, Non-Demented and Very Mild Demented were obtained as 97.12%, 99.97%, 93.79% and 95.26% respectively. We further validated our proposed framework using stratified 5-fold cross validation which further corroborated the robustness of our proposed framework. The EFMM achieved a mean accuracy of 98.38% ± 0.36 and a mean macro-F1 score of 98.48% ± 0.43. Friedman statistical testing demonstrated that there were significant differences between the performance of the classifiers evaluated (p < 0.001), which further validated the performance of the EFMM. Conclusions: To sum up, the proposed SqueezeNet-PCA-EFMM is an effective and efficient method for Alzheimer's disease stage classification under MRI images. The combination of SqueezeNet, PCA, and EFMM-led not only to high classification performance, but also to good cross validation results. Furthermore, this property of incremental learning is the intrinsic one of the EFMM and renders this framework interesting for its incorporation in the next-generation intelligent clinical decision supports in particular, as medical care evolves.
Background: High-resolution ultrasound is increasingly used as a complementary tool in carpal tunnel syndrome (CTS), but the value of median nerve cross-sectional area (CSA) for severity stratification remains uncertain. This study evaluated whether median nerve CSA measured at the pisiform level can distinguish severe from moderate CTS classified according to the Bland grading scale. Methods: This prospective cross-sectional study included 72 participants: moderate CTS (n = 34), severe CTS (n = 24), and healthy controls (n = 14). Nerve conduction studies, abductor pollicis brevis needle electromyography, and ultrasound examinations were performed on the same day. CSA was measured three times by a blinded examiner, and the mean value was used for analysis. Correlation, age-adjusted linear regression, and receiver operating characteristic curve analyses were performed. Results: Median nerve CSA increased progressively from controls to moderate and severe CTS groups (8.0 ± 1.1, 16.5 ± 4.6, and 19.8 ± 3.4 mm2, respectively; p < 0.001), with significant pairwise differences. Intraobserver and interobserver reliability were excellent (ICC = 0.95 and 0.91, respectively). CSA correlated positively with distal motor and sensory latencies and inversely with compound muscle action potential and sensory nerve action potential amplitudes. CSA also showed a weak positive association with neurogenic motor unit action potentials. In age-adjusted regression analysis, distal motor latency remained independently associated with CSA (B = 0.873, p = 0.002). A CSA cut-off of 18.5 mm2 distinguished severe from moderate CTS with an area under the curve (AUC) of 0.728, sensitivity of 66.67%, and specificity of 79.41%. Conclusions: Median nerve CSA measured at the pisiform level may provide useful adjunctive structural information for CTS severity assessment and may show moderate discriminatory performance for differentiating severe from moderate CTS. CSA should be interpreted in conjunction with clinical and electrodiagnostic findings rather than as a stand-alone severity marker.
Background/Objectives: Accurate peripheral blood cell detection remains difficult in dense smear fields, where touching erythrocytes, small platelets, and staining variability reduce the reliability of box-based object detectors. This study aims to address these limitations by proposing BloodContourNet, a contour-informed and density-aware detection framework for red blood cell, white blood cell, and platelet localization. Method: BloodContourNet is built on YOLOv26-L and integrates three complementary modules. The Hematological Attention Module (HAM) refines high-level features using local convolution and shifted-window attention. The Density-Aware Reweighting Head (DARH) modulates the classification loss for locally sparse or difficult instances without image-level resampling. The Bézier Contour Refinement Head (BCRH) provides a weakly supervised geometric overlap cue for ambiguous dense-region post-processing. Results: Experiments on the TXL-PBC dataset, containing 1260 peripheral smear images and 18,143 annotated cells, benchmarked 25 YOLO variants and 10 Transformer-based or reference detectors under a unified split and evaluation protocol. Across five random seeds, BloodContourNet achieved 99.1% mean Average Precision (mAP) at 50% Intersection-over-Union (IoU) and 87.8% mAP@50:95, improving strict-IoU performance by 1.4 points over YOLOv26-L and 0.7 points over DINO-DETR. Both margins were statistically significant across the five seeds (paired two-sided t-tests, both p < 0.05), and the five-seed 95% confidence interval of the proposed model was [87.5, 88.0]. Conclusions: BloodContourNet primarily improved the strict-localization setting on TXL-PBC, where dense erythrocyte regions and small platelet instances remain challenging for box-based detectors. External patient- or slide-independent validation remains necessary before broader clinical claims can be made.
Background/Objectives: Artificial intelligence (AI) is increasingly being evaluated for ophthalmic diagnosis, screening, and triage, yet its role in paediatric eye care remains less established than in adult ophthalmology. This systematic review aimed to synthesise evidence on AI-enabled tools for paediatric ophthalmic diagnosis, screening, triage, surveillance, and referral, with an emphasis on diagnostic performance, safety, workflow integration, equity, and implementation readiness in primary, community, and primary care-relevant settings. Methods: A PRISMA-guided systematic review was conducted using MEDLINE, Embase, Web of Science, Scopus, and IEEE Xplore from inception to 30 March 2026. Eligible studies evaluated AI or machine-learning tools for children and adolescents aged 0-18 years in relation to paediatric eye conditions. Study selection and data extraction were undertaken independently by reviewers, with disagreements resolved by consensus or third-reviewer adjudication. Methodological and reporting quality was evaluated using an author-adapted six-domain rubric informed by APPRAISE-AI. Diagnostic-accuracy studies were assessed using an author-adapted QUADAS-2 framework incorporating QUADAS-AI-informed AI-specific considerations, the prediction-model study was assessed using PROBAST+AI, and the non-randomised treatment-effect study was assessed using ROBINS-I. The public dataset descriptor was evaluated separately using an author-developed dataset-quality, representativeness, and applicability framework. Because of clinical and methodological heterogeneity, findings were synthesised thematically. Results: Twelve empirical studies and one public dataset descriptor were included, covering retinopathy of prematurity, retinoblastoma, amblyopia risk, myopia, congenital cataract, and visual-acuity assessment. AI systems frequently demonstrated promising diagnostic or screening performance, including sensitivity-first detection of treatment-requiring retinopathy of prematurity, high discrimination for retinoblastoma activity, and strong myopia prediction using fundus images. Several studies supported feasibility in neonatal, school, and community workflows using smartphone-based imaging, task-shifted operators, tele-referral, and human-in-the-loop review. However, external and temporal validation, calibration, patient-level reporting, subgroup and fairness assessment, and economic evaluation were limited. Conclusions: AI-enabled tools show promise for supporting selected paediatric ophthalmic screening, triage, and surveillance pathways, particularly when combined with image-quality control, explicit escalation, and human oversight. However, confidence in the reported performance is limited by single-centre studies and enriched samples, small numbers of clinically important cases, heterogeneous analytical units, potentially optimistic aggregation procedures, limited external or temporal validation, incomplete calibration, and absent fairness analyses. Routine autonomous implementation remains premature.
Objectives: This study aims to compare the imaging performance of spectral-domain optical coherence tomography (SD-OCT) and swept-source OCT (SS-OCT) in the early postoperative assessment of patients undergoing vitreoretinal surgery with intraocular gas or air tamponade. Methods: Seventeen eyes of 17 patients who underwent pars plana vitrectomy with either sulfur hexafluoride (SF6, 20%) or air tamponade were prospectively enrolled. All patients underwent OCT imaging on postoperative day 1 using both the SD-OCT system and the SS-OCT platform, without pharmacological mydriasis. Images were independently evaluated by two experienced ophthalmologists based on the ability to delineate four prespecified anatomical layers: inner retinal layers, ellipsoid zone (EZ), retinal pigment epithelium (RPE) and choroid. Images were classified as adequate quality if two or more of these structures were identifiable. Results: SS-OCT provided adequate-quality images in all 17 eyes (100%), with complete visualization of the inner retinal layers, EZ, RPE, and choroid in 15 eyes (88.2%). In contrast, SD-OCT yielded adequate-quality images in only three of 17 eyes (17.6%), demonstrating marked signal attenuation, interface artifacts, and inability to resolve deeper retinal structures in most cases. No difference in imaging performance was observed between SF6 and air tamponade subgroups. Conclusions: SS-OCT demonstrates markedly superior imaging performance in gas-filled eyes on postoperative day 1 compared to SD-OCT, primarily attributable to its longer wavelength, reduced sensitivity roll-off, and superior penetration through optically challenging media. These findings suggest that SS-OCT may offer meaningful advantages for early postoperative monitoring following vitreoretinal surgery with tamponade; confirmation in larger prospective cohorts with clinical-outcome correlation is warranted before it can be recommended as the preferred modality.
Background: CAV (cardiac allograft vasculopathy) is considered the leading cause of late post-transplant mortality and affects approximately 50% of transplant patients at 10 years post-transplant. Its etiopathogenetic mechanism is considered to be immune-mediated, but the identification of other non-immunological risk factors could represent new therapeutic targets for this pathology. Lipofuscin is due to reactive oxygen species (ROS) and appears to have a determining role in CAV. The aim of this study is to investigate the association between the amount of lipofuscin identified on EMB (endomyocardial biopsy) from patients followed up after heart transplantation, and the presence of CAV detected by coronary angiography. Methods: This retrospective study includes 99 EMBs from 47 transplanted patients, who also had coronary angiography. The amount of lipofuscin, damage to intramyocardial small vessels, vasculitis, Quilty effect, and acute cellular and humoral rejection was evaluated microscopically. Results: 19 CAV cases (19.2%) were identified, of which 15 (68.18% of total CAV cases) were insignificant. Grade 3 lipofuscin affected equally the cases with insignificant and significant CAV, 3 cases (37.5%) from each group. Grade 2 lipofuscin was reported in 10 cases (58.8%) of insignificant CAV, respectively 1 case of significant CAV (5.9%), (p-value of 0.0001). Quantitative evaluation of lipofuscin on microscopic sections revealed 8 EMBs (8.1%) with grade 3 lipofuscin. Two cases (25.0%) with lipofuscin score 3 were associated with moderate ACR (acute cellular rejection), ISHLT 2R and 3 cases (37.5%) with lipofuscin score 3 were associated with mild ACR ISHLT 1R, the differences being statistically significant, (p = 0.0001). Lipofuscin grades 2 and 3 were associated with severe fibrosis in 6 cases (35.3%) and 2 cases (25.0%), respectively (p = 0.042). A statistically significant association between the degree of damage to the intramyocardial small vessels and the amount of intracytoplasmic lipofuscin was observed (p = 0.00014). Discussion: Our study revealed that lipofuscin was more frequently associated with CAV, fibrosis, and damaged small vessels. Oxidative stress influences lipofuscinogenesis and CAV, which leads to endothelial dysfunction and neointimal hyperplasia, which over time will produce progressive narrowing of the vascular lumen and dysfunction of the cardiac allograft. Of the total number of 19 cases with CAV, 17 (89.47%) presented a lipofuscin score of 2 or 3 concomitantly with CAV, (p = 0.0001). This could mean that lipofuscin is not a harmless degradation product. At the same time, the association of a large number of cases with lipofuscin score 2, 10 cases (58.8%) with insignificant CAV could lead to the idea of using lipofuscin as a potential biomarker in the early diagnosis of CAV. Conclusions: We evidenced a significant association between the amount of intracytoplasmic lipofuscin and CAV. Accordingly, lipofuscin might be involved in the pathogenesis of CAV. Further research is needed to clarify the exact mechanisms of this association.
Background/Objectives: To determine the prevalence and distribution of cervical enamel projections (CEPs) and enamel pearls (EPs) in Saudi adults using cone-beam computed tomography (CBCT). Methods: In this retrospective cross-sectional study, CBCT scans of 1376 consecutive patients attending postgraduate periodontics clinics between January 2024 and February 2025 were screened. After applying eligibility criteria, 153 patients (732 first and second molars) were included. Three calibrated examiners independently assessed each molar for CEPs and EPs. Inter-rater agreement was quantified using Fleiss's kappa. Prevalence was reported at the patient and tooth levels with 95% confidence intervals, and bivariate associations with age, gender, and medical history were tested using chi-square, Fisher's exact, and Mann-Whitney U tests. Results: Of 732 molars in 153 patients, the patient-level prevalence was 29.4% for CEPs and 3.3% for EPs; tooth-level prevalence was 7.7% and 0.7%, respectively. Grade I CEPs were the most common, and mandibular first molars were most affected. Inter-rater agreement was excellent (Fleiss's κ = 0.86 for CEPs; κ = 1.00 for EPs). Neither dental anomaly was associated with age, gender, or medical history. Conclusions: CEPs are common in Saudi adults, while EPs are uncommon. Neither dental anomaly is associated with patient demographics or systemic health, supporting their developmental rather than acquired origin.
Background/Objectives: Large language models (LLMs) show promise for clinical decision support, yet their accuracy in interpreting specialized medical guidelines remains uncertain. Retrieval-augmented generation (RAG) may enhance performance by grounding responses in authoritative knowledge bases. This study aimed to compare the accuracy, comprehensiveness, and safety of RAG-enhanced versus standard LLMs for answering clinical questions derived from the German S3 guideline for oral cavity carcinoma. Methods: We conducted a prospective, single-blind benchmark study evaluating six LLMs: one RAG-enhanced model (Custom GPT with guideline access), one consensus-based model (ConsensusGPT), and four standard models (DeepSeek-V3.2, Mistral Small 3.2, Qwen3-Next-80B, GPT-OSS-120B). Fifty clinical questions covering 17 guideline domains were presented to each model three times, yielding 900 evaluations. Three expert reviewers assessed responses using 5-point Likert scales for accuracy, comprehensiveness, and clarity, under a single-blind procedure, the effectiveness of which was tested by a pre-specified manipulation check. We then ran a paired within-model experiment in which each base model was queried with and without guideline access through a transparent, openly released retrieval pipeline, and scored every response with a condition-blind automated judge alongside deterministic retrieval metrics computed from the logs. Secondary outcomes included hallucination rates and guideline citation behavior. Inter-rater reliability was assessed using intraclass correlation coefficients (ICCs). Results: In a paired within-model design that held each base model fixed, adding transparent guideline retrieval improved accuracy-significantly in the three weaker open-weight models (Mistral, Qwen3, and GPT-OSS) and directionally in the already-strong DeepSeek and GPT-5 bases. Because a pre-specified blinding check found that experts could still identify retrieval-augmented answers with 98.5% accuracy, we anchored causal interpretation on measures that do not depend on the human raters, ranked by their independence: deterministic, log-derived retrieval metrics first, and then an automated, condition-blind LLM judge, whose agreement with the experts (Spearman ρ = 0.81, 95.7% within-one agreement) establishes shared calibration rather than independence from their bias. Deterministically from the retrieval logs, citation groundedness rose from 0% to 51-89% and retrieval recall@5 was 92%. On the judge, content-level hallucination fell from 42% to 4% and accuracy rose by a pooled +0.64 points (95% CI 0.47-0.80); the accuracy gain persisted after adjustment for response length (+0.48, 95% CI 0.22-0.73), which retrieval shortened rather than lengthened. The accuracy gain was large for weaker base models and small or non-significant for already-strong ones, whereas the hallucination and auditability gains were consistent across all models. The human ratings reproduced the judge's accuracy effect (+0.61, 95% CI 0.49-0.74), and GPT-5 run through the transparent pipeline showed no significant difference from the proprietary Custom GPT (judge accuracy 4.48 vs. 4.58). Conclusions: Guideline retrieval yields a reproducible, largely base-independent improvement in the safety and auditability of LLM answers to clinical guideline questions, with accuracy gains concentrated in weaker base models. Because retrieval-augmented answers are recognizable to experts, rigorous evaluation should rely on rater-independent measures, and residual hallucination continues to require human oversight.
Background/Objectives: To test whether routine T1-weighted spin-echo (T1-SE) and proton-density fast spin-echo (PD-FSE) MRI signal intensity (SI) can index bone regeneration after osteochondral scaffold implantation by correlating MRI metrics with micro-CT and histology. Methods: Twenty-eight sheep with bilateral trochlear defects were evaluated in separate cohorts at 30, 180, and 365 days (n = 7, 11, and 10, respectively). One knee received a tri-layered resorbable scaffold; the contralateral defect was left empty. MRI (T1-SE, PD-FSE) was read by two blinded musculoskeletal radiologists using 10-point Likert scales (T1: apparent mineralization; PD: SI normalization). Contrast-to-noise ratio (CNR) between repair-bone and reference bone was computed. MRI measures were correlated with micro-CT (new bone volume; trabecular bone volume) and histology (ICRS subchondral bone reconstruction; new bone; filling). Results: Likert ratings on both T1-SE and PD-FSE correlated with micro-CT new bone volume and trabecular bone volume and with histological measures of repair. PD-FSE CNR was inversely associated with micro-CT new bone volume and trabecular bone volume (both p < 0.001), indicating lower CNR with greater bone regeneration, whereas T1-SE CNR showed no significant associations. Inter-reader agreement was excellent (ICC 0.947 individual; 0.973 average), with good-to-excellent intra-reader reliability (0.920-0.963). Conclusions: Reader-based Likert assessment on both sequences and PD-FSE CNR provide complementary, non-invasive markers of subchondral bone regeneration, supporting MRI for radiation-free follow-up and endpoint selection in future translational studies. T1-SE CNR did not track incremental mineralization and should not be used as a stand-alone quantitative marker in early healing.
Background/Objectives: A novel CT dynamic angiographic imaging (CT-DAI) analytic algorithm was evaluated against the clinical gold standard for fractional flow reserve (FFR) measurement in patients with coronary artery disease (CAD) characterized by diffuse dense calcification and previous stent implantation. Methods: This retrospective feasibility study included 24 coronary arteries in 16 patients (age 69.9 ± 8.9 years, 11 males) with CAD who underwent dynamic CT myocardial perfusion scanning using a dual-source CT scanner after intravenous infusion of adenosine triphosphate. The included patients had analyzable proximal and distal coronary artery segments adjacent to the stenosis in the myocardial perfusion images and had corresponding invasive catheter-based FFR measurements for that stenosis. An in-house software based on the CT-DAI algorithm was used to compute FFR using the coronary time-enhancement curves sampled across the stenosis from stress myocardial CT perfusion images. The CT-DAI derived FFR values were then compared to the corresponding catheter-based FFR values. A coronary stenosis was considered functionally significant for FFR values below 0.8. Results: The mean axial length and calcium score of the coronary stenoses were 47.9 mm and 1911.63 Agatston Units, respectively. Eight coronary arteries received stents from previous treatments. The CT-DAI derived FFR values (0.822 ± 0.143) showed an excellent linear correlation (R = 0.974) with and were indifferent from the invasive FFR values (0.826 ± 0.147, p = 0.537), resulting in 100% per-vessel and per-patient sensitivity and specificity for the detection of functionally significant coronary stenosis. Bland-Altman analysis revealed a minimal mean difference in FFR measurements (0.004) between the two modalities with the lower and upper limits of agreement at -0.061 and 0.069, respectively. Conclusions: The findings suggest that CT-DAI can derive FFR for the coronary arteries with heavy calcification and stents from dynamic myocardial CT perfusion images.
Background/Objectives: Ultra-high-magnification endoscopy (Endocyto) visualizes microscopic mucosal structures and has been proposed as a tool for assessing mucosal inflammation. We previously developed an endocytoscopic classification system (EC-A to EC-D) for grading the severity of inflammation. In this study, we investigated the relationship between endocytoscopic classification and microRNA (miRNA) expression in 36 patients with ulcerative colitis (UC). Methods: This cross-sectional study was conducted in two phases at two institutions. In the first phase, microarray analysis was performed on biopsy samples from patients with UC and healthy controls enrolled at Nagasaki University Hospital. In the second phase, 36 patients with UC who underwent total colonoscopy via Endocyto at Tottori University Hospital were included, and selected miRNAs were analyzed via quantitative polymerase chain reaction. Results: Differentially expressed miRNAs were identified, and four (miR-141-5p, miR-192-5p, miR-194-5p, and miR-215-5p) were downregulated in inflammatory areas. Notably, miR-192-5p expression was markedly lower in the EC-B and EC-C+D groups than in the EC-A group. Furthermore, expression of CXC motif ligand 2 (CXCL2), a pro-inflammatory chemokine, was upregulated in inflammatory tissues and negatively correlated with miR-192-5p expression. Conclusions: These results indicate that reduced miR-192-5p expression is inversely associated with CXCL2 expression and that, among the miRNAs examined, miR-192-5p downregulation most consistently corresponded to the endocytoscopic classification. Our findings suggest that Endocyto may serve as a real-time, in vivo adjunct for a histology-like assessment of mucosal inflammation; its clinical utility requires prospective validation.
Objective: To characterize the anatomical spectrum of a single coronary artery (SCA) and an anomalous coronary artery originating from the opposite sinus of Valsalva (ACAOS) and to assess high-risk features using Coronary computed tomography angiography (CCTA)-based quantitative parameters. Methods: A retrospective review of 2513 consecutive CCTA examinations performed between June 2021 and December 2025 was conducted. Patients diagnosed with ACAOS or SCA were included. Coronary origin, course patterns, and morphological characteristics were analyzed. High-risk features—including interarterial course, intramural segment, acute take-off angle, slit-like ostium, intramural length, and interluminal space (ILS)—were quantitatively assessed using multiplanar and curved MPRs. Results: Twenty-two patients (0.88%) were identified (ACAOS: 0.64%, n = 16; SCA: 0.24%, n = 6). All SCA cases showed benign courses. Among ACAOS patients, 62.5% had an interarterial course. In this subgroup, the mean take-off angle was 14.4° ± 4.5°, with universal slit-like ostium and intramural course. Mean intramural length was 9.21 ± 2.00 mm and minimal ILS was 0.87 ± 0.12 mm. Conclusions: CCTA enables comprehensive anatomical characterization of morphological features associated with higher risk in patients with ACAOS and SCA. Quantitative assessment of parameters such as the take-off angle, intramural course, slit-like ostium, intramural length, and interluminal space may complement clinical evaluation, although further prospective studies incorporating functional ischemia assessment and clinical outcome data are needed to clarify their clinical significance.
Cardiovascular complications increasingly challenge survivors of urological cancers, given the cardiotoxicity of therapies such as androgen deprivation, vascular endothelial growth factor receptor inhibitors, tyrosine kinase inhibitors, immune checkpoint inhibitors, and chemotherapy. This narrative review addresses the complex crosstalk between urological cancer treatment and cardiovascular disease. It summarizes cardiovascular toxicities linked to major antineoplastic agents, explores underlying mechanisms including metabolic and immune-mediated effects, and proposes strategies for surveillance, diagnosis, and management. Highlighting the need for multidisciplinary collaboration, it outlines future directions for research to optimize cardiovascular outcomes in this high-risk population. The increasing complexity of cardiovascular care in patients with urological malignancies highlights the need for closer collaboration between cardiologists, urologists, and oncologists, with uro-cardio-oncology emerging as an important multidisciplinary field.
Pilonidal sinus is a chronic inflammatory skin condition predominantly found in the sacrococcygeal region, characterized by recurrent infections and sinus tract formation. Occurrences outside this area, particularly on the anterior chest wall, are exceedingly rare and prone to misdiagnosis. We report a rare case of a 24-year-old Chinese man presenting with a 3-year history of a recurrent, draining lesion on the anterior chest wall. Initial treatment via simple incision and drainage for a presumed sebaceous cyst resulted in delayed wound healing and persistent purulent discharge. Subsequent magnetic resonance imaging (MRI) and computed tomography (CT) revealed a localized superficial lesion. Definitive surgical debridement uncovered a sinus tract containing embedded hair fragments. Histopathological examination confirmed a pilonidal sinus with a robust foreign body giant cell reaction and chronic inflammation. After complete resection, during the nearly two-year follow-up period, the healing has been smooth, and no recurrence has been observed. This case highlights the necessity of detailed clinical history-taking and the inclusion of ectopic pilonidal sinus in the differential diagnosis of refractory chest wall abscesses to prevent repeated, ineffective interventions.
Endometriosis is a common gynecological disorder that can involve the intestinal tract and may present with bleeding, bowel obstruction, and, rarely, perforation or malignant transformation. Intestinal endometriosis can mimic malignant tumors or other conditions, which may lead to unnecessary aggressive surgical resections. It continues to be a challenging diagnosis to make preoperatively. We report a case of intestinal endometriosis with endoscopic features of lymphatic dilatation-like changes, which correlated with the pathological findings, aiming to provide a reference for the endoscopic recognition of intestinal endometriosis.
Background/Objectives: Male chronic pelvic pain syndrome (CPPS) is a heterogeneous condition involving overlapping urinary, psychosocial, organ-specific, infectious, neurological, myofascial, and sexual phenotypes. This complexity limits the effectiveness of routine symptom assessment and empirical treatment strategies. Artificial intelligence (AI) may support more reproducible interpretation, differential diagnosis, phenotyping, and personalized management. This up-to-date narrative mapping review aimed to identify AI-assisted approaches that are directly or indirectly relevant to the assessment of males with CPPS, classify them according to UPOINTS phenotypic domains and clinical functions, and critically discuss the extent to which current evidence is disease-specific or extrapolated from related conditions. Methods: A literature search was performed in PubMed/MEDLINE, the Cochrane Library, and Google Scholar from database inception to May 2026 using terms related to male CPPS, UPOINTS domains, diagnosis, treatment, prognosis, digital solutions, artificial intelligence, machine learning, deep learning, natural language processing, computer vision, and decision support. Studies were included if they described AI-based or AI-adjacent computational approaches relevant to male CPPS or to related conditions important for UPOINTS-based phenotyping, differential diagnosis, or phenotype-specific assessment. Results: Available evidence remains fragmented and is largely extrapolated from related urological, chronic pain, pelvic floor, infectious, neurological, and sexual medicine conditions. AI applications were most developed in urinary and organ-specific domains, including uroflowmetry analysis, bladder volume assessment, cystoscopy, prostate imaging, urinary biomarkers, and differential diagnosis of lower urinary tract disorders. AI tools also showed potential for infection detection, psychosocial screening, chronic pain monitoring, neuroimaging-based phenotyping, pelvic floor dysfunction assessment, and evaluation of sexual dysfunction. However, male-CPPS-specific validation remains limited. Conclusions: AI has promising potential to improve differential diagnosis, multidomain phenotyping, and individualized management in males with CPPS. Current evidence is mainly translational and hypothesis-generating. Future studies should focus on prospective male-CPPS-specific cohorts, external validation, explainable multimodal models, and integration of AI tools into clinically meaningful, patient-centered workflows.
Background: Pancreatic steatosis (PS) has emerged as a potential risk factor for pancreatic ductal adenocarcinoma (PDAC). While increasing evidence supports its role in pancreatic carcinogenesis, its prognostic significance after PDAC diagnosis remains unclear. We performed a systematic review to evaluate the association between PS and survival outcomes in patients with pancreatic cancer. Methods: A systematic search of PubMed and Scopus was conducted in May 2026 according to PRISMA guidelines. Observational studies assessing pancreatic steatosis by imaging or histology and reporting survival outcomes in PDAC patients were included. Risk of bias was evaluated using the Quality in Prognosis Studies (QUIPS) tool. Results: Five studies met inclusion criteria, comprising three imaging-based and two histology-based papers. Three studies reported findings suggestive of an adverse prognostic role of PS, whereas two found no significant association with survival. The strongest evidence originated from a study using quantitative histological assessment of pancreatic fat, which demonstrated that greater pancreatic fat accumulation was independently associated with poorer overall survival and recurrence-free survival following surgical resection. In contrast, imaging-based studies yielded inconsistent results. Histology-based studies consistently supported a negative prognostic impact, whereas CT-derived attenuation measures showed substantial variability. Risk of bias was moderate to high in most studies, largely due to retrospective designs, small sample sizes, heterogeneous definitions of pancreatic steatosis, and incomplete adjustment for metabolic confounders. Conclusions: Current evidence regarding the prognostic value of pancreatic steatosis in PDAC is limited and conflicting. While histology-based studies suggest an adverse impact on survival, imaging-based data remain inconsistent.
Background/Objectives: Dental caries is one of the most common oral diseases worldwide, and early diagnosis is essential for effective treatment. However, detecting carious lesions in panoramic radiographs is challenging because of low image contrast, anatomical complexity, and overlapping structures. This study aimed to develop an improved object detection framework for dental caries localization in panoramic radiographs. Methods: A modified YOLOv8 architecture was developed by integrating ConvNeXtV2 blocks into the Cross-Stage Partial (C2f) modules to enhance feature extraction and gradient propagation. To improve detection performance and reduce overfitting, key training hyperparameters were optimized using a genetic algorithm. The proposed model was evaluated on a dataset of 474 panoramic dental radiographs annotated by experts using bounding boxes. Results: The optimized model achieved a box precision of 78.4%, a recall of 53.6%, and a mean average precision at 50% intersection-over-union threshold (AP50) of 62.6%. Compared with the baseline YOLOv8 model, the proposed approach improved precision and AP50. Visual and quantitative analyses demonstrated that the ConvNeXtV2-enhanced architecture enabled more accurate localization of carious regions. Conclusions: The results indicate that combining ConvNeXtV2-based architectural enhancement with genetic algorithm-based hyperparameter optimization is an effective strategy for dental caries localization in panoramic radiographs. Although recall remains a limitation, the proposed framework shows potential as a computer-aided diagnostic tool for supporting clinical caries assessment.
Background/Objectives: Automated dermoscopic image analysis can assist research on early skin cancer detection, but deep learning models may learn acquisition-related shortcuts, including terminal hairs, illumination gradients, gel bubbles, shadows and measurement markings. This study evaluates a causal uncertainty-decomposed ensemble (CUDE) for confounding-aware skin-lesion classification within the ISIC 2019 benchmark setting. Methods: CUDE uses a structured variational autoencoder (SVAE) to impose separate lesion-relevant and acquisition-related nuisance latent heads. Three stochastic latent-space experts are trained on complementary streams and are integrated by a Dirichlet-based fusion module trained with nuisance sampling. The causal graph is used as a modeling prior rather than proof of causal identification. Results: Using a stratified internal split of the labeled ISIC 2019 training collection, CUDE achieved a balanced accuracy of 89.7% and an AUROC of 0.972. Under the predefined synthetic artifact protocol, CUDE showed a relative performance drop of 5.0%, compared with 9.6% for the deep ensemble. Deferring the 10% most uncertain cases increased non-deferred accuracy from 89.7% to 93.2% and reduced false-negative rates for melanoma, BCC and SCC in the non-deferred subset. Conclusions: Within the evaluated ISIC 2019 split and controlled synthetic corruption protocol, structured latent separation, stochastic expert diversity and Dirichlet fusion were associated with improved benchmark robustness and calibration. These results should not be interpreted as evidence of broad real-world robustness or clinical deployment readiness; external multi-center, device-diverse and skin-tone-diverse validation remains required.
Background/Objectives: Transfemoral cerebral angiography (TFCA) involves significant radiation exposure because of prolonged fluoroscopy and repeated imaging sequences. Because complex dose simulations are often infeasible in routine clinical practice, this study employed PC-based Monte Carlo methods to estimate organ-absorbed and effective doses. The primary objective was to establish Kerma area product (KAP)-based conversion factors to facilitate efficient clinical dose assessment. Methods: We analyzed 30 patients who underwent TFCA. Key parameters, including total KAP, tube voltage, and projection angles, were obtained from DICOM Radiation Dose Structured Reports. Monte Carlo simulations were conducted for anteroposterior, lateral, and cranial planes. The conversion factors were then derived by calculating the ratio of the effective dose to the total KAP. Results: The average KAP was 36.23 ± 6.81 Gy·cm2. Effective doses calculated via Monte Carlo simulation were 1.70 ± 0.35 mSv under International Commission on Radiological Protection (ICRP) 60 and 1.90 ± 0.40 mSv under ICRP 103 standards. The corresponding conversion factors were 0.0468 ± 0.0054 and 0.0525 ± 0.0065 mSv/Gy·cm2 respectively. Notably, the ICRP 103-based factor was 12.2 percent higher than the ICRP 60-based factor, reflecting updated tissue weighting factors in modern dosimetry. Conclusions: This study provides standardized conversion factors that allow rapid and reliable dose estimation without requiring complex simulations. The findings demonstrated substantial radiation absorption in the head and neck regions, providing critical data for routine dose monitoring. These results emphasize the importance of rigorous radiation protection and safety protocols during interventional procedures.
