Background and Aim: Large language models (LLMs) demonstrate significant potential in assisting with medical image interpretation. However, the diagnostic accuracy of general-purpose LLMs on image-based internal medicine cases and the added value of brief clinical history remain unclear. This study evaluated three general-purpose LLMs (ChatGPT, Gemini, and DeepSeek) on expert-curated cases to quantify diagnostic accuracy with image-only input versus image plus brief clinical context. Methods: We conducted a comparative evaluation using 138 expert-curated cases from Harrison’s Visual Case Challenge. Each case was presented to the models in two distinct phases: Phase 1 (image only) and Phase 2 (image plus a brief clinical history). The primary endpoint was top-1 diagnostic accuracy for the textbook diagnosis, comparing performance with versus without a brief clinical history. Secondary/Exploratory analyses compared models and assessed agreement between model-generated differential lists and the textbook differential. Statistical analysis included Wilson 95% confidence intervals, McNemar’s tests, Cochran’s Q with Benjamini–Hochberg correction, and Wilcoxon signed-rank tests. Results: The inclusion of clinical history substantially improved diagnostic accuracy for all models. ChatGPT’s accuracy increased from 50.7% in Phase 1 to 80.4% in Phase 2. Gemini’s accuracy improved from 39.9% to 72.5%, and DeepSeek’s accuracy rose from 30.4% to 75.4%. In Phase 2, diagnostic accuracy reached at least 65% across most disease nature and organ system categories. However, agreement with the reference differential diagnoses remained modest, with average overlap rates of 6.99% for ChatGPT, 36.39% for Gemini, and 32.74% for DeepSeek. Conclusions: The provision of brief clinical history significantly enhances the diagnostic accuracy of large language models on visual internal medicine cases. In this benchmark, performance differences between models were smaller in Phase 2 than in Phase 1. While diagnostic precision improves markedly, the models’ ability to generate comprehensive differential diagnoses that align with expert consensus is still limited. These findings underscore the utility of context-aware, multimodal LLMs for educational support and structured diagnostic practice in supervised settings while also highlighting the need for more sophisticated, semantics-sensitive benchmarks for evaluating diagnostic reasoning.

Background: The SeDeM-ODT expert system is designed to assess the suitability of the pharmaceutical ingredients for their conversion into an orodispersible formulation by direct compression. The tool can be utilized to select the most appropriate excipients that improve the compressibility and buccodispersibility of the formulation. Objective: This study aimed to utilize the SeDeM-ODT expert system to evaluate the performance of superdisintegrants and select an appropriate superdisntegrant for Doxylamine Succinate orodispersible formulation. Method: The SeDeM-ODT expert system scrutinized the excipients to develop an orodispersible Doxylamine Succinate formulation. Among the 15 parameters of the tool, some of them were determined through experimental work, while the remaining were calculated through the experimental values of other parameters. The central composite design approach was used for formulation development. The prepared powder blends were compressed using the direct compression method and evaluated for different parameters (hardness, thickness, diameter, friability, weight variation, water absorption ratio, wetting time, and disintegration time). Results: The results of the SeDeM-ODT expert system were correlated with the values obtained by the post-compression tests. The Crospovidone formulation (F7) was found to be an optimized formulation as it disintegrated quickly compared with the other formulations containing other superdisintegtrants. The results perfectly endorsed the SeDeM-ODT expert system evaluation, as Crospovidone showed the highest IGCB value of 6.396. Conclusion: The study observed the effectiveness of the expert system in accurately examining the performance of disintegrating agents. The study observed the effectiveness of the expert system in accurately examining the performance of disintegrating agents. The assessment proved Crospovidone to produce quicker disintegration in Doxylamine Succinate orodispersible formulation.

Background and objective: Adverse drug reactions (ADRs) cause substantial harm, and a considerable proportion may be preventable, but under-reporting persists and weakens pharmacovigilance. Spontaneous reporting depends on clinicians, yet under-reporting persists and weakens pharmacovigilance. To map barriers to adverse drug reaction reporting among physicians, a comparison of nurses and pharmacists in single-country studies was carried out between high-income countries (HICs) and low- and middle-income countries (LMICs). Methods: A scoping review was conducted following PRISMA-ScR guidance. PubMed and Web of Science were searched for studies published from 2016 onward. Eligible studies were single-country primary empirical studies including physicians, nurses, or pharmacists and examining ADR reporting. Only barriers that were measured or explicitly explored and reported as extractable results were included. Barriers were coded into 12 domains and summarised by income group and profession. Results: A total of 44 studies were included, with 18 from HICs and 26 from LMICs. Survey designs were most common. Pharmacists were the most frequently studied cadre. Knowledge and training barriers were reported in all studies in both income groups. Fear of legal or punitive concerns was reported in 13/18 (72.2%) HIC studies and 17/26 (65.4%) LMIC studies. Time and workload barriers were reported in 10/18 (55.6%) HIC studies and 11/26 (42.3%) LMIC studies. Access barriers to tools, forms, and information technology showed the clearest income group difference: these were reported in 5/18 (27.8%) HIC studies versus 16/26 (61.5%) LMIC studies. Lack of feedback or acknowledgement was reported in 8/18 (44.4%) HIC studies and 10/26 (38.5%) LMIC studies. Conclusions: Barriers extend beyond individual knowledge in all settings. The main income group difference was the greater prominence of reporting system access barriers in LMICs compared with workflow and time pressure barriers in HICs. Addressing fear and building a supportive non-punitive reporting culture remains a cross-cutting priority because these were common issues in both income groups and can limit reporting even when infrastructure and training exist.

Introduction: The COVID-19 pandemic that originated in Wuhan, China in December 2019 results in respiratory and gastrointestinal infections. Elderly patients are at high risk. Preventive measures like avoiding contact with COVID-19 patients and wearing N95 masks can contribute to reducing the risk of infection, but vaccination remains crucial. Methodology: A cross-sectional survey-based study was conducted among the elderly population (≥ 50 years) in Pakistan. A 16-items questionnaire explored the socio-demographic profile of the participants, including questions about their age, gender, etc. The other sections included questions regarding vaccine registration, vaccine acceptance, and hesitation and fears towards vaccines. The items were developed to collect the data on the basis of the Likert scale. Results: There was a total of 3059 respondents. The majority of the participants (n = 2726, 89.1%) were aged below 70 years, while 10.9% (n = 333) were aged above 70 years. Many participants (47.1%) expressed their concerns about the efficacy of COVID-19 vaccine booster dose. Despite their willingness to vaccinate with the cOVID-19 vaccine, 24.7% female participants expressed concerns related to the safety of the vaccine. Among all the respondents, 23.1% participants had no concerns about the cOVID-19 vaccine booster dose. The study identified a reluctance and lack of confidence in the efficiency of the COVID-19 booster dose. Conclusions: The alarming situation is rigidity towards vaccination among the elderly population. Necessary measures must be taken by the health department of Pakistan to manage this reluctant behavior and increase confidence on the efficiency of the COVID-19 booster dose.

1. Hospitalized adults often start treatment with IV antibiotics and many later move to oral therapy once infection control and clinical stability are achieved. The timing of this transition matter...

Background: Large language models (LLMs) are increasingly used in medical education, but their performance and reliability on mechanistically demanding topics like acid–base interpretation are unclear. Methods: We conducted a cross-sectional head-to-head evaluation of three LLMs (ChatGPT, DeepSeek, Perplexity) using 75 textbook acid–base cases related to management and drug treatment. One obsolete case was excluded, leaving 74 vignettes (29 metabolic, 10 respiratory, 35 mixed) that generated 510 multiple-choice questions (MCQs). Each MCQ was mapped to case category, one of seven cognitive domains, and one of four distractor-based difficulty bands. All questions were posed via public web interfaces in two independent phases (Phase I and II). Primary outcome was accuracy (proportion correct); consistency outcomes were identical answers across phases and reproducibly correct answers (correct in both phases). Results: Overall accuracy was 59.8% (305/510) for ChatGPT, 58.8% (300/510) for DeepSeek, and 53.7% (274/510) for Perplexity (Cochran’s Q p = 0.16); ChatGPT and DeepSeek each outperformed Perplexity (p = 0.0086 and p = 0.0374). In metabolic cases (164 items), Perplexity scored 50.6% versus 60.4% and 62.8% for ChatGPT and DeepSeek. Accuracy was highest for compensation/expected response (116 items; 70.7% for ChatGPT and DeepSeek, 62.1% for Perplexity) and lowest for metabolic acidosis subtypes (59 items; 52.5%, 42.4%, 32.2%). Accuracy declined with difficulty, with Perplexity significantly lower than the other models in moderate–hard and hard bands. Identical answers across phases occurred in 81.7% of items for ChatGPT, 81.1% for DeepSeek, and 76.1% for Perplexity, but reproducibly correct answers were 52.1%, 48.8%, and 43.2%. Conclusions: ChatGPT and DeepSeek showed moderate performance and generally higher accuracy than Perplexity. Stable repeated errors despite only moderate reproducibly correct responses support cautious interpretation of LLM outputs and argue against their use as stand-alone tools for acid–base learning or clinical decision making of management and drug treatment .