Crowdsourcing with Enhanced Data Quality Assurance: An Efficient Approach to Mitigate Resource Scarcity Challenges in Training Large Language Models for Healthcare.

Large Language Models (LLMs) have demonstrated immense potential in artificial intelligence across various domains, including healthcare. However, their efficacy is hindered by the need for high-quality labeled data, which is often expensive and time-consuming to create, particularly in low-resource domains like healthcare. To address these challenges, we propose a crowdsourcing (CS) framework enriched with quality control measures at the pre-, real-time-, and post-data gathering stages. Our study evaluated the effectiveness of enhancing data quality through its impact on LLMs (Bio-BERT) for predicting autism-related symptoms. The results show that real-time quality control improves data quality by 19% compared to pre-quality control. Fine-tuning Bio-BERT using crowdsourced data generally increased recall compared to the Bio-BERT baseline but lowered precision. Our findings highlighted the potential of crowdsourcing and quality control in resource-constrained environments and offered insights into optimizing healthcare LLMs for informed decision-making and improved patient care.

A Culinary Medicine Elective for Clinically Experienced Medical Students: A Pilot Study.

Purpose: Culinary medicine, a combination of nutrition science and the culinary arts, is an emerging approach for teaching nutrition to medical students and improving their competence in counseling patients with diet-associated diseases. Data are, however, lacking on the impact of culinary medicine courses directed at clinically experienced students. This study reports initial outcomes of a pilot nutrition and culinary medicine course targeting 4th-year medical students. Methods: An elective course on culinary medicine was offered to 4th-year medical students at the Perelman School of Medicine, comprising seven disease-focused sessions, with a final capstone session. Students read primary literature for each session. Individual sessions consisted of culinary literacy, cooking, and a case discussion led by physicians and registered dietitians. In addition, students participated in a nutrition education initiative in a local high school. Students completed pre-and postcourse surveys that evaluated perceived nutrition knowledge and counseling skills and personal dietary choices, and included free-text options for qualitative comments. Results: Thirty-one 4th-year medical students participated in three offerings of the course in 2018 and 2019. There was strong student enthusiasm for the course as (1) all the available slots for each course offering were filled within 2 h of an e-mail announcement to the students; (2) student attendance was consistently very high; and (3) student feedback about the course was uniformly positive. Students reported significant increases in their confidence regarding (1) knowledge of pertinent nutrition information; (2) discussing nutrition with patients; and (3) ability to impact patient behavior through counseling (p < 0.001). Qualitative comments suggested that students were contemplating or implementing changes in their dietary habits and food choices. Conclusion: A culinary medicine course for clinically experienced medical students may improve perceived nutrition knowledge and increases confidence in counseling patients with diet-associated diseases.

Exclusion of an Azygos Vein Varix With a Covered Stent.

An azygos vein varix was incidentally discovered in a 26-year-old man. Owing to the potential risk of pulmonary emboli, we implanted a covered stent in the superior vena cava, effectively excluding the varix. Eighth months later, the varix was thrombosed and involuted. (Level of Difficulty: Advanced.).