Abstract Medical trainees depend on diverse visual reference material to develop diagnostic pattern-recognition skills, yet such material is often limited in diversity and accessibility. This paper presents a classifier-in-the-loop generative framework for synthesising realistic histopathology images: a Stable Diffusion v1.5 model is fine-tuned using Low-Rank Adaptation (LoRA) on the PathMNIST dataset, consisting of nine colorectal […]

Abstract The number of diabetes patients is increasing worldwide, but many cases stay undiagnosed in low-resource regions. The A1C blood test is invasive and costly, therefore it is often not available in these regions. In this research, DiaBreath is proposed as a low-cost noninvasive screening system based on exhaled breath analysis. Volatile organic compounds (VOCs) […]

Abstract Digital commerce platforms compete with local vendors under frictions of time, price, and visibility. We test whether an interpretable choice model with reinforcement and social influence can yield substantial changes in aggregate market share as control parameters vary. We simulate repeated customer choice among vendors using a utility function that combines vendor attributes (price, […]

Abstract Local businesses are important to neighborhood service access, but access to these services is limited in some areas. This study uses GIS and machine learning to identify possible limited service zones for restaurants and supermarkets across Dublin, Pleasanton, and Livermore in Alameda County, California. The model dataset included 1,134 Alameda County census block groups, […]