Chittagong Medical College (CMC) is situated in the port city of Chittagong, Bangladesh and is the oldest medical college in the south-eastern part of the country. It is affiliated with a tertiary level hospital where clinical students are trained in medicine, surgery, obstetrics and gynaecology and other sub-specialties.
Background: Cardiovascular diseases remain one of the leading causes of mortality worldwide, particularly in low- and middle-income countries. Early and accurate prediction of heart disease is essential for timely intervention and improved patient outcomes. Machine learning techniques offer promising solutions; however, challenges such as class imbalance, lack of interpretability, and limited real-world validation persist. Methods: In this study, a machine learning-based heart disease prediction framework was developed using a real-world clinical dataset comprising 5000 patient records collected from healthcare facilities in Bangladesh. Data preprocessing included cleaning, feature encoding, train–test splitting, and class imbalance handling using the Synthetic Minority Oversampling Technique (SMOTE). Multiple machine learning models — Logistic Regression, Decision Tree, Support Vector Machine, and Random Forest — were evaluated using 10-fold stratified cross-validation. Model performance was assessed using accuracy, precision, recall, and F1-score. SHAP (SHapley Additive exPlanations) was employed to enhance model interpretability. The best-performing model was deployed as a web-based decision support system. Results: Among the evaluated models, the Random Forest classifier achieved the best performance, with an accuracy of 98%, recall of 96%, and F1-score of 96%. Ablation studies demonstrated the effectiveness of SMOTE, feature integration, and ensemble modeling. SHAP analysis identified clinically relevant features contributing to heart disease prediction, enhancing transparency and trust in model decisions. Conclusions: The proposed framework provides an accurate, interpretable, and practical solution for heart disease prediction using real-world clinical data. The integration of explainable machine learning and web-based deployment highlights its potential for clinical decision support. Future work will focus on multi-center prospective validation and adaptive model updating to further improve generalizability and real-world applicability.
BACKGROUND:Online education gained its popularity in the education system during the COVID-19 pandemic lockdown. The online platform, including social media, was institutionalized globally for the purpose of tutoring to keep the education process ongoing under feasible circumstances. However, the post-pandemic continuation of online education and prolonged usage of electronic devices imposed a greater risk of health issues related to sensory impairment. Our study aimed to determine the impact of online education on students' hearing status and its associated factors. METHODS:A cross-sectional study was conducted among 1030 students of 11th grade and above who were undergoing online education in Dhaka and Chattogram. Data were collected through the online administration of a structured questionnaire containing questions on sociodemographic status, family history of diseases, personal history of diseases, information related to screentime exposure, and SSQ-12 (Speech, Spatial, and Qualities of Hearing -12) scale. Descriptive statistics, Pearson's chi-square test, two independent sample t-tests, and multiple linear regression analysis were employed to obtain the results. RESULT:The mean SSQ score of the study participants was 7.74 ± 1.37. In bivariate analysis, gender, family income, family history of diseases (e.g., obesity, headache, hearing problem), personal history of diseases (e.g., obesity, insomnia), device type (mobile/tablet, computer), average daily screen time with sound, and break pattern during online learning were significantly (p < 0.05 for all) associated with hearing status. In multivariate analysis, being female (coefficient -0.293, p = 0.001), using mobile/tablet (coefficient -0.836, p = 0.001), and continuous screen use (coefficient -0.348, p = 0.003) were significantly associated with poor hearing status. CONCLUSION:This current study indicates the detrimental effect of online education on the hearing of young students in Bangladesh. Future studies should explore the long-term hearing effects of online education and guide the policy makers towards necessary preventive approaches.
BACKGROUND:Menstrual hygiene management is an important public health issue related to reproductive health, dignity, and well-being. Although medical students are expected to have basic knowledge of reproductive physiology, gaps may persist in practical menstrual hygiene knowledge, attitudes, and practices (KAP). This study aimed to assess KAP regarding menstrual hygiene among second-year female medical students. METHODS:This single-institution cross-sectional study was conducted from January 2025 to June 2025 among second-year female medical students at Shahabuddin Medical College Hospital, Dhaka, Bangladesh. Of 63 eligible female second-year MBBS students, 46 participated and were recruited using convenience sampling. Data were collected using a pre-tested, structured, self-administered questionnaire assessing sociodemographic characteristics and menstrual hygiene-related KAP. Composite KAP scores were calculated. Multiple linear regression analysis was performed to examine factors associated with KAP scores. RESULTS:The mean age of the participants was 20.98 ± 0.61 years. Twenty-one participants 21 (45.65%) resided in hostels, and 42 (91.30%) were from urban backgrounds. All participants reported awareness of menstruation, and 41 (89.13%) correctly identified the normal menstrual cycle length. Most participants strongly agreed that menstruation is a natural and normal process (42, 91.30%) and that proper menstrual hygiene has health benefits (44, 95.65%). Sanitary pads were the most commonly used menstrual product (43, 93.48%). However, only 26 (56.52%) correctly identified changing menstrual products every two to four hours as appropriate. Hostel residence was significantly associated with higher KAP scores (all p < 0.001). Urban background was also significantly associated with higher knowledge (p < 0.001), attitude (p = 0.005), and practice scores (p < 0.001). CONCLUSIONS:Second-year female medical students demonstrated good general awareness and favorable attitudes regarding menstruation and menstrual hygiene. However, practical gaps remained, particularly regarding appropriate menstrual product-changing frequency and product diversity. Hostel residence and urban background were associated with higher KAP scores, although findings should be interpreted cautiously because of the small sample size and single-institution design. Structured menstrual health education may help improve practical hygiene knowledge and future counseling capacity among medical students.
Background: The blood coagulation system is very delicately balanced; is maintained in a fluid state in the vasculature and rapidly clots to seal an injury. This study aims to determine the demographic and clinical pattern of bleeding disorders. Materials and Methods: This prospective type of cross-sectional study comprised 112 patients with bleeding or a history of bleeding in the Department of Haematology and Bone Marrow Transplantation Unit, Dhaka Medical College Hospital, Dhaka, from January to December 2016. Patients were examined for clinical presentation and underwent haematological investigations to arrive at a diagnosis. The data was obtained, tabulated and analysed by using SPSS 22. Results: In this study, 81.2% of patients suffered from congenital coagulation factors deficiency, and haemophilia was found to be the most common congenital haemostatic defect (61.6%). Bleeding disorder was found predominantly among males in 86.6% compared to females (13.4%). 1st-degree consanguinity was mainly observed in Glanzmann thrombasthenia and Von Willebrand Disease. History of prolonged bleeding was statistically associated with clinical diagnosis of the patient (p<0.01). Conclusions: Haemophilia A and B were found as the leading causes of bleeding disorder. Von Willebrand Disease (VWD) also played a role and Glanzmann thrombasthenia was the most common one that had been identified among other disorders.
Background: As dermatologic care is largely based on visual cues, teledermatology has emerged as a potential solution to improve access to care while maintaining clinical efficacy. The study aimed to understand the magnitude of skin diseases and related treatment-seeking behaviour in tribal villages of Rangamati Sadar, Rangamati of Chittagong Hill Tracts using a simple “Store and Forward” technique of teledermatology. Materials and methods: A cross-sectional study was conducted in five randomly selected villages of Rangamati Sadar, Rangamati district of Bangladesh, during June–August 2023. Images of skin conditions were captured on smartphones by trained surveyors. Deidentified images were transferred over WhatsApp to the principal investigator, who formulated diagnosis and treatment in real-time and sent them via WhatsApp to surveyors. Subsequently, the patient was examined face-to-face by a dermatologist at Rangamati General Hospital. The diagnoses formulated during teledermatology and face-to-face consultations were analyzed and compared. Results: A total of 200 participants consented. Among them, 107 (53.5%) were identified with suspected skin diseases and consented to photograph. The treatment-seeking behaviour and compliance with treatment were poor. Tinea infections were the most frequent skin disease present in 21 (19.6%) of the participants, followed by acne (15%), scabies (14%), eczema (14%), atopic dermatitis (9.3%), skin and soft tissue infection (9.3%), Seborrhoeic dermatitis (6.5%). There was a 93.5% concordance between the diagnoses of skin diseases by mobile teledermatology and face-to-face consultation (Cohen K coefficient 0.89). More than 75% of the photos were of excellent to good quality. Conclusions: Skin diseases were highly prevalent among tribal people, and the pattern was similar to that of the country’s other populations. Mobile teledermatology using smartphones is a reliable, useful, and cost-effective method of providing expertise for improving dermatologic care for the needy population in rural and remote areas of Bangladesh IAHS Medical Journal Vol 9(1), June 2026;73-78
