
BACKGROUND Early identification of medical students at risk of poor academic performance is essential for timely academic intervention and student support. This study aimed to develop and evaluate machine learning models to predict the performance of second-year MBBS students in summative assessments. METHODS This observational analytical study was conducted at All India Institute of Medical Sciences, Kalyani, using academic data from five consecutive MBBS batches (2019-2023). Variables included demographic information, attendance records, internal assessment scores, previous academic performance, and final summative examination results in microbiology, pathology, and pharmacology. Multiple machine learning algorithms including logistic regression, decision tree, random forest, support vector machine (SVM), k-nearest neighbors, Naïve Bayes, XGBoost, and LightGBM were implemented using R statistical software. The models were trained using data from four batches and validated using the fifth batch. Performance was evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve (AUC). RESULTS Machine learning models demonstrated strong predictive performance across all three subjects. Among the algorithms evaluated, the SVM model showed the highest sensitivity (89%) in identifying students at risk of failing in final examinations. Specificity across models reached 100%, indicating that students predicted to pass were accurately classified. Random forest, XGBoost, and LightGBM also demonstrated high predictive accuracy and AUC values. CONCLUSION Machine learning-based predictive models can effectively identify medical students at risk of poor academic performance. Integrating predictive analytics into medical education systems may enable early academic support, reduce student stress, and improve educational outcomes.
BACKGROUND This study aimed to evaluate ocular findings in transfusion-dependent beta-thalassemia patients and determine whether they exhibit a higher incidence of ocular abnormalities compared to healthy controls owing to effects of iron overload and chelation therapy. MATERIALS AND METHODS An analytical cross-sectional study was conducted in the Departments of Ophthalmology and Pediatrics at a tertiary care center in South India. Patients aged 10-18 years with transfusion-dependent thalassemia, receiving regular blood transfusions (every 2-4 weeks for at least 2 years), were included as cases. Age- and sex-matched healthy individuals without ocular complaints served as controls. All participants underwent a comprehensive ophthalmic evaluation, including visual acuity, color vision, slit-lamp examination, dilated fundus examination, dry eye testing (tear film breakup time [TBUT] and Schirmer's), and intraocular pressure measurement. Clinical history, annual transfusion requirement, and mean serum ferritin levels were recorded for cases. Data were analyzed using IBM SPSS Statistics for Windows, version 25.0 (IBM Corp., Armonk, NY, USA) with appropriate parametric or nonparametric tests, and P < 0.05 was considered statistically significant. RESULTS Of the thalassemia patients, 65.1% were male. Retinal vascular tortuosity was observed in 14% (P = 0.026) and was significantly associated with higher serum ferritin (P = 0.004). Retinal pigment epithelium changes (9.3%) and increased cup-to-disc ratio (11.6%) showed nonsignificant trends. Tear film instability (TBUT <10 s) was present in 20.9% (P = 0.649), while Schirmer's test indicated dry eye in 18.6% (P = 0.044), both associated with elevated ferritin. Refractive errors were significantly more frequent in the case group (18.6%, P = 0.005). Lenticular opacities and Bitot's spots were found in 4.7% and 2.3%, respectively. CONCLUSIONS Ocular abnormalities are common among beta-thalassemia major patients, with some correlating with serum ferritin levels. Regular ophthalmic screening should be integrated into multidisciplinary care to facilitate early detection and management. Larger longitudinal studies are warranted to assess the progression and visual impact of these changes.
INTRODUCTION Urticaria is a commonly reported cutaneous adverse event following vaccination, which may range from mild self-limiting reactions to more severe hypersensitivity manifestations. Monitoring vaccine-associated urticaria is important for ongoing vaccine safety surveillance and pharmacovigilance. However, large-scale population-based analyses evaluating demographic characteristics, vaccine associations, onset intervals, and severity patterns remain limited. AIM The aim of the study was to analyze reports of urticaria in the Vaccine Adverse Event Reporting System (VAERS) database between 1990 and 2025 and evaluate demographic characteristics, vaccine type distribution, onset intervals, temporal trends, and seriousness patterns. METHODOLOGY A retrospective observational study was conducted using data from the VAERS database. Reports of urticaria submitted between January 1990 and March 2025 were included. The variables analyzed included age, sex, vaccine type, onset interval, and seriousness classification. Descriptive statistics were generated using frequencies and percentages, and comparative analyses were performed between serious and nonserious reports. RESULTS A total of 61,996 urticaria reports were identified during the study period. Most reports occurred among adults aged 18-49 years and females. Most reports were classified as nonserious and occurred within the early postvaccination period. COVID-19 vaccines accounted for the highest proportion of reports, followed by influenza and measles, mumps, and rubella vaccines. Temporal analysis demonstrated a marked increase in reporting during 2021, followed by declining trends thereafter. Comparative analysis identified differences in demographic and clinical characteristics between serious and nonserious reports. CONCLUSION Urticaria is a frequently reported adverse event following vaccination, with identifiable demographic, temporal, and clinical patterns. Most reports were nonserious and occurred shortly after vaccination, although differences were observed between serious and nonserious reports. Continued pharmacovigilance and future studies incorporating detailed clinical data are needed to better understand factors associated with vaccine-related urticaria.
BACKGROUND Spontaneous intracerebral hemorrhage (ICH) represents a disproportionately high stroke burden (20%-30%) in low- and middle-income countries like India, necessitating accurate, early risk stratification using readily available clinical and radiographic features. MATERIALS AND METHODS This prospective observational study included 185 patients with spontaneous ICH at a tertiary care center. Baseline clinical severity was assessed using the Glasgow Coma Scale (GCS) and the ICH score. Radiographic markers included hematoma volume, intraventricular hemorrhage (IVH), and radiographic signs (swirl, blend, black hole, satellite, and spot signs). The primary outcome was unfavorable functional status (modified Rankin scale: 4-6, including mortality) at 90 days. Univariate analyses focused on identifying predictors associated with a poor outcome, followed by multivariable logistic regression. RESULTS The mean age was 56.0 ± 12.4 years, with 72.4% prevalence of hypertension. The unfavorable outcome rate was 37.3% (N = 69), with 22.2% mortality. Univariate analysis (P < 0.05) revealed strong associations between poor outcome and greater hematoma volume, lower GCS score, higher ICH score, IVH, swirl sign, blend sign, satellite sign, hydrocephalus, herniation, and ≥2 multiple noncontrast computed tomography (NCCT) signs. Multivariable analysis identified hematoma volume, GCS score, ICH score, herniation, satellite sign, and male gender as independent predictors. CONCLUSION Clinical severity indicators and multiple NCCT signs in univariate analysis serve as robust markers of poor functional outcome. The independent prognostic value of the satellite sign and herniation highlights the utility of readily available radiographic features for immediate risk stratification.