Search Results (18,292)

Background: Irritable Bowel Syndrome (IBS) is a heterogeneous gastrointestinal disorder characterized by complex brain–gut interactions. Patient Similarity Networks (PSNs) offer a novel approach for exploring this heterogeneity and identifying clinically relevant patient subgroups. Methods: We analyzed data from 78 participants (49 IBS patients and 29 healthy controls) with 36 brain morphometric measures (FreeSurfer v7.4.1) and 6 measures of cognitive functions (5 RBANS domain indices plus a Total Scale score). PSNs were constructed using multiple similarity measures (Euclidean, cosine, correlation-based) with Gaussian kernel transformation. We performed community detection (Louvain algorithm), centrality analyses, feature importance analysis, and correlations with symptom severity. Statistical validation included bootstrap confidence intervals and permutation testing. Results: The PSN comprised 78 nodes connected by 469 edges, with four communities detected. These communities did not significantly correspond to diagnostic groups (Adjusted Rand Index = 0.011, permutation $p=0.212$), indicating IBS patients and healthy controls were intermixed. However, each community exhibited distinct neurobiological profiles: Community 1 (oldest, preserved cognition) showed elevated intracranial volume but reduced subcortical gray matter; Community 2 (youngest, most severe IBS symptoms) had elevated cortical volumes but reduced white matter; Community 3 (most balanced IBS/HC ratio, mildest IBS symptoms) showed the largest subcortical volumes; Community 4 (lowest cognitive performance across multiple domains) displayed the lowest RBANS scores alongside high IBS prevalence. Top network features included subcortical structures, corpus callosum, and cognitive indices (Language, Attention). Conclusions: PSN identifies brain–cognition communities that cut across diagnostic categories, with distinct feature profiles suggesting different hypothesis-generating neurobiological patterns within IBS that may inform personalized treatment strategies. Full article

Collateral status is an important therapeutic indicator for acute ischemic stroke (AIS), yet visual collateral grading remains subjective and suffers from inter-observer variability. To address this limitation, this study automatically extracted binarized vascular morphological features from CTA images and developed a convolutional neural network (CNN) for automated collateral classification. Performance trends were systematically analyzed across diverse hyperparameter combinations to meet different clinical decision needs. A total of 157 AIS patients (median age 65 [⁵⁷,⁵⁸,⁵⁹,⁶⁰,⁶¹,⁶²,⁶³,⁶⁴,⁶⁵,⁶⁶,⁶⁷,⁶⁸,⁶⁹,⁷⁰,⁷¹,⁷²,⁷³,⁷⁴] years; 61.8% were male) were retrospectively enrolled and stratified by Menon score into good (3–5, n = 117) and poor (0–2, n = 40) collateral groups. A total of 192 architectures were established, and three representative model tendencies emerged: a sensitivity-oriented model (AUC = 0.773; sensitivity = 87.18%; specificity = 65.00%), a balanced model (AUC = 0.768; sensitivity = 72.65%; specificity = 77.50%), and a specificity-oriented model (AUC = 0.753; sensitivity = 63.25%; specificity = 85.00%). These results demonstrate that kernel size, the number of filters in the first layer, and the number of convolutional layers are key determinants of performance directionality, allowing tailored model selection depending on clinical requirements. This work highlights the feasibility of CTA-based automated collateral classification and provides a systematic framework for developing models optimized for sensitivity, specificity, or balanced decision-making. The findings may serve as a reference for clinical model deployment and have potential for integration into multi-objective AI systems for endovascular thrombectomy patient triage. Full article

Background: Dietary transitions toward Westernized patterns (WDPs) (high in processed foods, sugars, and fats) pose a global public health challenge. The Westernized Diet Index (WDI) measures adherence to these patterns. However, its validity with respect to metabolic biomarkers warrants thorough evaluation for use in epidemiological and clinical research. Objectives: This study validates the WDI using metabolic biomarkers (including anthropometrics, blood pressure, fasting blood glucose (FBG), triglycerides, HDL-c, LDL-c, and total cholesterol), examines its association with metabolic syndrome (MetS), and compares scoring methods to identify the most effective measure of WDPs adherence. Methods: Data from 10,146 participants in the Fasa Adult Cohort Study (FACS) were used. We calculated the WDI using global (WDI-G) and population (WDI-P) Z scores and food group (WDI-FG)-based algorithms. Validation employed logistic and linear regression, ROC (receiver operating characteristic) curves, Youden’s index, and k-means clustering. Results: All WDI scoring methods (across all methods, higher scores indicated lower adherence to WDPs) demonstrated a strong, significant association with all three MetS definitions (WHO, NCEP: ATPIII, and IDF) and nearly all investigated metabolic biomarkers. In fully adjusted logistic models, WDI Global (WDI-G) (OR: 0.23) and WDI Food Groups (WDI-FG) (OR: 0.26) were significantly associated with MetS (based on the WHO definition). Also, in fully adjusted linear regression models, a 10% increase (reflecting lower adherence to WDPs) in the WDI-G score (range: −2.03 to 1.11) was significantly associated with a 3.96 mg/dL reduction in FBG and a 2.61 cm reduction in waist circumference. Additionally, ROC curves (AUC: 0.57–0.61) demonstrated that WDI predicts MetS with moderate accuracy. The strongest associations were observed with population-based scoring. In addition, based on comparative performance, WDI-G, WDI-P, and WDI-FG appear most suitable for cross-population, within-cohort, and mechanistic or intervention-focused research, respectively. Conclusions: The WDI shows promise as a nutritional tool for assessing adherence to WDPs and exploring associations with metabolic health outcomes, including MetS. These findings suggest that the WDI may be useful in future dietary, public health, and clinical research, although further validation in diverse populations is warranted. Full article

Background: Probable sarcopenia, indicated by low handgrip strength, is a prevalent condition among hospitalized older adults and may reflect broader functional and nutritional decline. Methods: We examined differences in nutritional, functional, and cognitive status between Alzheimer’s clinical syndrome (ACS) patients with probable sarcopenia and those without sarcopenia. A cross-sectional analysis was conducted on 194 hospitalized older adults with ACS. Probable sarcopenia was defined using European Working Group on Sarcopenia in Older People (EWGSOP2) handgrip strength thresholds. Results: Patients with probable sarcopenia (n = 137) had significantly lower Mini-Mental State Examination (MMSE) scores, Geriatric Nutritional Risk Index (GNRI), albumin, hemoglobin, and gait speed compared to those without. After age and sex adjustment, MMSE (p = 0.023), GNRI (p = 0.002), hemoglobin (p = 0.022), albumin (p = 0.003), and gait speed (p < 0.001) remained significantly different. In the sex- and age-adjusted multivariable model (adjusted R² = 0.442), higher nutritional risk (β = 0.26, p = < 0.001), lower MMSE scores (β = 0.17, p = 0.029), polypharmacy (β = –4.20, p = 0.002), and slower gait speed (β = 4.12, p = 0.010) were associated with reduced handgrip strength. In the multivariable binary logistic regression model (adjusted for age and sex), moderate or high nutritional risk and slow gait speed emerged as independent predictors of probable sarcopenia, with OR 5.14 (95% CI 1.34–19.75; p = 0.017) and OR 3.13 (95% CI 1.30–7.52; p = 0.011), respectively. Conclusions: Probable sarcopenia in hospitalized older adults with ACS is highly prevalent and is associated with higher nutritional risk, poorer cognitive and physical function, and polypharmacy; its early recognition may help to guide more targeted nutritional and functional interventions. Full article

Postoperative delirium (POD) is a frequent and serious complication among elderly surgical patients. Despite its clinical relevance, reliable biomarkers for early identification and pathophysiological insight remain limited. Recent evidence implicates systemic immune activation and complements dysregulation as contributors to cognitive decline after surgery. This study investigated the association between perioperative levels of selected complement pathway proteins and both the incidence and severity of POD. Methods: We performed a secondary analysis of 22 patients aged ≥ 60 years from the prospective CONFESS cohort undergoing elective spine surgery. Complement proteins (C1q, C2, C4), mannose-binding lectin (MBL), Factor D [FD], Factor B [FB], Factor I [FI] were quantified from blood samples collected at baseline, preoperatively, and on postoperative days 1 and 2. POD was assessed using the Nursing Delirium Screening Scale (Nu-DESC) and Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria. Delirium severity was rated with the Confusion Assessment Method–Severity (CAM-S) scale. Associations were tested using univariate and multivariate regression analyses. Preoperative levels of FD and C2 were significantly elevated in patients who developed POD (FD: p = 0.023; C2: p = 0.044), while C4 levels trended lower. FD remained an independent predictor of POD in multivariate regression (p = 0.049), although cognitive performance was the only significant predictor when adjusted for surgery duration. Delirium severity was associated with perioperative reductions in C1q, FI, and FB and with increased MBL levels, explaining up to 43% of CAM-S score variance. These findings highlight the role of complement activation—particularly FD, C2, MBL—in the development and clinical expression of POD. Complement profiling may offer a novel approach for risk stratification and therapeutic targeting in perioperative neurocognitive disorders. Full article

Background/Objectives: Primary Progressive Aphasia (PPA) is a neurodegenerative disorder characterized by gradual and progressive deterioration of speech and language abilities. Speech and language therapy is considered an important intervention to slow decline and support the recovery of linguistic functions in individuals with PPA. This study aims to examine the effectiveness of an elaborated Semantic Feature Analysis (SFA) approach in enhancing naming abilities and semantic networks in individuals with the logopenic and semantic variants of PPA. Methods: Fourteen participants were recruited, including seven individuals with logopenic PPA and seven with semantic PPA. All participants received an elaborated SFA intervention twice weekly for four weeks. The Aphasia Language Assessment Test (ADD), the Turkish Picture Naming Test (T-RAT), and the SAQOL-39 were conducted at the following three time points: prior to treatment (pre-test), immediately after treatment (post-test), and one month post-treatment (follow-up). Results: Significant improvements were observed in ADD, T-RAT, and SAQOL-39 scores in both logopenic and semantic PPA groups following treatment (p < 0.05). Although follow-up scores declined compared to posttest performance (p < 0.05), several follow-up scores remained higher than pretest levels. Between-group comparisons indicated no significant difference in ADD scores; however, logopenic PPA participants demonstrated higher T-RAT scores (p < 0.05), while semantic PPA participants showed higher SAQOL-39 scores, except at follow-up (p < 0.05). Conclusions: Preliminary results suggest that the elaborated SFA intervention is effective in improving naming skills, language functioning, and quality of life in both logopenic and semantic variants of PPA. Although treatment gains partially decreased after one month, many improvements were maintained above baseline, supporting the clinical value of SFA in managing language decline in PPA. Full article

Background: Colorectal cancer (CRC) is the second leading cause of cancer-related mortality worldwide, accounting for approximately 10% of all cancer cases. Despite the proven effectiveness of conventional screening modalities such as colonoscopy and fecal immunochemical testing (FIT), their invasive nature, high cost, and limited patient compliance hinder widespread adoption. Recent advancements in artificial intelligence (AI) and bowel sound-based signal processing have enabled non-invasive approaches for gastrointestinal diagnostics. Among these, bowel sound analysis—historically considered subjective—has reemerged as a promising biomarker using digital auscultation and machine learning. Objective: This review explores the potential of AI-powered bowel sound analytics for early detection, screening, and characterization of colorectal cancer. It aims to assess current methodologies, summarize reported performance metrics, and highlight translational opportunities and challenges in clinical implementation. Methods: A narrative review was conducted across PubMed, Scopus, Embase, and Cochrane databases using the terms colorectal cancer, bowel sounds, phonoenterography, artificial intelligence, and non-invasive diagnosis. Eligible studies involving human bowel sound-based recordings, AI-based sound analysis, or machine learning applications in gastrointestinal pathology were reviewed for study design, signal acquisition methods, AI model architecture, and diagnostic accuracy. Results: Across studies using convolutional neural networks (CNNs), gradient boosting, and transformer-based models, reported diagnostic accuracies ranged from 88% to 96%. Area under the curve (AUC) values were ≥0.83, with F1 scores between 0.71 and 0.85 for bowel sound classification. In CRC-specific frameworks such as BowelRCNN, AI models successfully differentiate abnormal bowel sound intervals and spectral patterns associated with tumor-related motility disturbances and partial obstruction. Distinct bowel sound-based signatures—such as prolonged sound-to-sound intervals and high-pitched “tinkling” proximal to lesions—demonstrate the physiological basis for CRC detection through bowel sound-based biomarkers. Conclusions: AI-driven bowel sound analysis represents an emerging, exploratory research direction rather than a validated colorectal cancer screening modality. While early studies demonstrate physiological plausibility and technical feasibility, no large-scale, CRC-specific validation studies currently establish sensitivity, specificity, PPV, or NPV for cancer detection. Accordingly, bowel sound analytics should be viewed as hypothesis-generating and potentially complementary to established screening tools, rather than a near-term alternative to validated modalities such as FIT, multitarget stool DNA testing, or colonoscopy. Full article

Background/Objectives: Large osteochondral lesions of the talus (OLT) pose a major challenge because their size and depth often exceed the indications for bone marrow stimulation, and durable biological repair remains difficult to achieve. However, evidence for autologous osteochondral transplantation (AOT) in extensive talar defects is still limited. Methods: In this retrospective cohort, 40 consecutive patients ≥ 14 years with ICRS grade III–IV lesions of the talar dome were treated with AOT at a tertiary referral center. One to three overlapping cylindrical osteochondral grafts (mean diameter 0.9 cm) were harvested from non-weight-bearing regions of the ipsilateral patellofemoral groove using a water-cooled diamond trephine system and implanted press-fit into the talar dome. Donor sites were refilled with autologous iliac crest bone cylinders and hydroxyapatite substitute. Pain (Numeric Rating Scale, NRS) and function (AOFAS Ankle–Hindfoot Score) were recorded preoperatively and at 3, 6, 9, and 12 months, and changes over time were analyzed using generalized estimating equations. Results: Mean defect size was 137.4 ± 31.9 mm², and 82.5% of lesions were ICRS grade III. NRS pain improved from 5.69 ± 2.52 preoperatively to 0.53 ± 0.98 at 12 months (p < 0.001). AOFAS score increased from 63.79 ± 2.55 to 97.36 ± 2.49 (p < 0.001). Age and graft location significantly influenced postoperative pain, whereas graft size and sex did not. No infections, graft failures, conversions to arthrodesis or arthroplasty, or clinically relevant donor-site symptoms occurred. Conclusions: Multi-plug AOT using a diamond trephine system provides substantial and durable pain relief and functional improvement in patients with large OLT, with low complication and donor-site morbidity rates. These findings support AOT as a joint-preserving option for extensive talar defects and justify further prospective, comparative studies with long-term follow-up. Full article

Background and Objectives: Early recognition of life-threatening organ dysfunction is central to modern sepsis frameworks. We systematically reviewed the prognostic performance and clinical utility of the Neonatal Sequential Organ Failure Assessment (nSOFA) for mortality and major morbidity in NICU populations. The search identified 939 records across databases; after screening and full-text assessment, 16 studies met the inclusion criteria. Methods: Following PRISMA guidance, we searched major databases (2019–2025) for observational or interventional studies reporting discrimination or risk stratification using nSOFA in neonates. Populations included suspected/proven infection and condition-specific cohorts. Heterogeneity in timing, thresholds, and outcomes precluded meta-analysis. Results: A cumulative sample exceeding 25,000 neonates was identified across late- and early-onset infection, all-NICU admissions, necrotizing enterocolitis, respiratory distress, and very preterm screening cohorts. Across settings and timepoints, nSOFA demonstrated consistent, good-to-excellent mortality discrimination, with reported AUROCs ≥ 0.80 and upper ranges near 0.90–0.92; serial scoring within the first 6–12 h generally improved risk classification. Disease-specific applications (NEC, early-onset infection) showed similar discrimination for death or composite adverse outcomes. Conclusions: Evidence from diverse NICU contexts indicates that nSOFA is a pragmatic, EHR-ready organ dysfunction score with robust discrimination for mortality and serious morbidity, supporting routine, serial use for risk stratification and standardized endpoints in neonatal sepsis pathways, aligned with contemporary organ dysfunction–based pediatric criteria. Full article

Background: Role clarity is a persistent challenge among Patient Care Technicians (PCTs), contributing to inconsistent task performance and safety risks. In Saudi Arabia, little is known about PCTs’ understanding of their responsibilities. This study evaluated the impact of a targeted educational program designed to improve PCTs’ role clarity, safety practices, and communication. Methods: A quasi-experimental pre-post study was conducted in September 2025 with 35 PCTs from the Hail Health Cluster. The one-day intervention included lectures, discussions, role-play, and case scenarios. Outcomes were measured using a validated instrument across four domains: role clarity; core clinical tasks and safety; communication and ethics; and objective knowledge. Pre-post changes were analyzed using paired t-tests (Cohen’s d), and subgroup differences in change scores were examined using one-way ANOVA (η²) in SPSS v29. Results: Baseline scores were lowest in objective knowledge (41.4%) and role clarity (62.8%). Post-training, total composite scores improved significantly (+10.88%, p < 0.001, d = 1.63), with the most significant gain in objective knowledge (+19.8%, p < 0.001, d = 0.99). Role clarity showed only a modest, non-significant increase (+3.98%, p = 0.088, d = 0.30). No demographic differences were found. Conclusions: Targeted training was effective in reducing knowledge gaps; however, improving role clarity may require organizational reinforcement beyond brief training. Full article

Objective: The objective was to evaluate the optimal timing of cervical cerclage insertion for perinatal outcomes, such as birthweight, gestational week, and pregnancy prolongation in women with diagnosed cervical insufficiency (CI). Methods: This retrospective study was conducted at the 1st Department of Obstetrics and Gynaecology of the Medical University of Warsaw, over a 10-year period. Maternal and perinatal outcomes were compared between 75 women divided into three groups based on the gestational week (GW) at cerclage insertion: (1) before 18 GW (n = 31), (2) 18–22 GW (n = 31), (3) after 22 GW (n = 13). Only single pregnancies were included in the final analysis in order to maintain the homogeneity of the population. The primary outcomes included the week of delivery and pregnancy prolongation following cervical cerclage insertion. Numerous secondary outcomes were also evaluated, including neonatal mortality, need for NICU hospitalization, Apgar score, birthweight, maternal white blood cell (WBC) count and C-reactive protein (CRP) levels. Results: Birth week was significantly associated with GW at insertion—35.8 ± 3.8 vs. 34.8 ± 5.2 vs. 32 ± 5.7, respectively, p = 0.016. Moreover, statistical difference was also found regarding birthweight of the analysed groups—2723.8 ± 951.6 g vs. 2518.5 ± 1167.9 g vs. 1886.7 ± 1011.2 g, respectively, p < 0.001, and pregnancy prolongation following cerclage insertion 20.4 ± 4.2 vs. 14.7 ± 5.5 vs. 7.3 ± 5.7 weeks, respectively, p < 0.001. Conclusions: Earlier cerclage placement (<18 weeks) is associated with significantly improved perinatal outcomes. However, this association largely reflects the benefit of prophylactic intervention over emergency ‘rescue’ procedures (common in the >22-week group). The sharp decline in outcomes after 22 weeks highlights the risks of advanced cervical dilation, suggesting that clinical management should prioritize risk assessment within the prophylactic window. Full article

Background/Objectives: Ulcerative colitis (UC) is an inflammatory bowel disease and a major cause of ulcers and chronic inflammation in the colon and rectum. Recurring symptoms include abdominal pain, rectal bleeding, and diarrhoea, and patients with UC are at a higher risk of developing comorbidities such as colorectal cancer and poor mental health. In UC, the decreased diversity and changed metabolic profile of gut microbiota, along with a diminished mucus layer, leads to disruption of the underlying epithelial barrier, with an ensuing excessive and detrimental inflammatory response. Treatment options currently rely on drugs that reduce the inflammation, but less emphasis has been placed on improving the resilience of the epithelial barrier. Macrolide antibiotics exhibit epithelial barrier-enhancing capacities unrelated to their antibacterial properties. Methods: We investigated two novel barriolides, macrolides with reduced antibacterial effects in common bacterial strains. Gut epithelial cell barrier resistance in the Caco-2 cell line, with and without co-culture with mucus-producing HT-29 cells, was increased when treated with barriolides. Using ^AXGut-on-Chip technology with inflammatory cytokine-stimulated Caco-2/HT-29 co-cultures, the effectiveness of the barriolides was confirmed. Lastly, we reveal the barrier-enhancing and inflammation-reducing effects of the barriolides in a dextran-sulphate sodium (DSS)-induced colitis mouse model. Results: We show the predictive power of the novel ^AXGut-on-Chip system and the effectiveness of the novel barriolides. Indications include reduced inflammatory response, increased epithelial barrier and decreased overall clinical score. Conclusions: The results of this study indicate the notion that barriolides could be used as a treatment option for UC. Full article

Background/Objectives: The increasing prevalence of overweight and obesity in adolescents represents a major global health concern. Adolescent weight gain frequently shows additional metabolic risk factors, including insulin resistance, hypertension, and dyslipidemia, whose co-occurrence defines the metabolic syndrome (MetS). Adherence to a healthy dietary pattern, such as the Mediterranean Diet (MD), has been shown to reduce the metabolic risk among adolescents. Skin carotenoid score has emerged as an objective and non-invasive indicator of MD adherence; however, its relationship with a cluster of metabolic parameters which characterize the MetS, including the triglyceride levels, diastolic blood pressure, and waist circumference, remains poorly explored. Here, we investigated the role of skin carotenoid score as an early biomarker of metabolic syndrome risk in adolescents. Methods: A sample of 634 healthy adolescents underwent anthropometric and clinical measurements, blood sample collection, and evaluation of the MD adherence by the Mediterranean Diet Quality Index for Children and Adolescents (KIDMED) questionnaire and the skin carotenoid levels by the Veggie Meter^®. Student’s t-test, chi-square test, Pearson correlation, and the multivariable linear regression model were used for analyses. Results: Participants had a mean BMI Z-score of 0.02 ± 1.01; the metabolic serum profile and the cardiovascular parameters were within the normal range. Mean KIDMED and skin carotenoid scores were 5.21 ± 2.56 and 357 ± 96.58, respectively. Skin carotenoids were positively associated with height (p = 0.02), while they were inversely associated with weight (p = 0.008), BMI Z-score (p < 0.0001), diastolic blood pressure (p = 0.013), and triglycerides (p = 0.003). Moreover, the carotenoid score was positively associated with male gender and KIDMED score and negatively associated with waist circumference and triglyceride levels in multivariable regression analyses. Conclusions: Our results suggested the potential application of skin carotenoid score as a complementary biomarker for the early identification of adolescents at increased metabolic risk. Full article

Background: Diagnosing CNS tumors through MRI is limited by significant variability in scanner hardware, acquisition protocols, and intensity characteristics at clinical centers, resulting in substantial domain shifts that lead to diminished reliability for automated models. Methods: We present a Domain-Adaptive MRI Learning Model (DA-MLM) consisting of an adversarially aligned hybrid 3D CNN–transformer encoder with contrastive regularization and covariance-based feature harmonization. Varying sequence MRI inputs ($T1$, $T1ce$, $T2$, and $FLAIR$) were inputted to multi-scale convolutional layers followed by global self-attention to effectively capture localized tumor structure and long-range spatial context, with domain adaptation that harmonizes feature distribution across datasets. Results: On the BraTS 2020 dataset, we found DA-MLM achieved 94.8% accuracy, 93.6% macro-F1, and 96.2% AUC, improving upon previously established benchmarks by 2–4%. DA-MLM also attained Dice score segmentation of 93.1% (WT), 91.4% (TC), and 89.5% (ET), improving upon 2–3.5% for CNN and transformer methods. On the REMBRANDT dataset, DA-MLM achieved 92.3% accuracy with segmentation improvements of 3–7% over existing U-Net and expert annotations. Robustness testing indicated 40–60% less degradation under noise, contrast shift, and motion artifacts, and synthetic shifts in scanner location showed negligible performance impairment (<0.06). Cross-domain evaluation also demonstrated 5–11% less degradation than existing methods. Conclusions: In summary, DA-MLM demonstrates improved accuracy, segmentation fidelity, and robustness to perturbations, as well as strong cross-domain generalization indicating the suitability for deployment in multicenter MRI applications where variation in imaging performance is unavoidable. Full article

Manual pill identification is often inefficient and error-prone due to the large variety of medications and frequent visual similarity among pills, leading to misuse or dispensing errors. These challenges are exacerbated when pill imprints are engraved, curved, or irregularly arranged, conditions under which conventional optical character recognition (OCR)-based methods degrade significantly. To address this problem, we propose GO-PILL, a geometry-aware OCR pipeline for robust pill imprint recognition. The framework extracts text centerlines and imprint regions using the TextSnake algorithm. During imprint refinement, background noise is suppressed and contrast is enhanced to improve the visibility of embossed and debossed imprints. The imprint localization and alignment stage then rectifies curved or obliquely oriented text into a linear representation, producing geometrically normalized inputs suitable for OCR decoding. The refined imprints are processed by a multimodal OCR module that integrates a non-autoregressive language–vision fusion architecture for accurate character-level recognition. Experiments on a pill image dataset from the U.S. National Library of Medicine show that GO-PILL achieves an F1-score of 81.83% under set-based evaluation and a Top-10 pill identification accuracy of 76.52% in a simulated clinical scenario. GO-PILL consistently outperforms existing methods under challenging imprint conditions, demonstrating strong robustness and practical feasibility. Full article