Jiacheng Xie

I'm a Ph.D student and research assistant at Digital Biology in University of Missouri.

Email  /  CV  /  Scholar  /  Github

Research

I'm interested in medical imaging analysis, clinical decision support systems, and the application of artificial intelligence in medicine and healthcare. My recent work also explores epidemiological insights derived from social media data. Much of my research focuses on building AI systems that can assist clinical diagnosis, treatment planning, and public health surveillance. Selected projects and publications are highlighted below.

In this work, we present TOM, an open-source tongue image segmentation method based on multi-teacher knowledge distillation and task-specific diffusion-based data augmentation. The method improves segmentation robustness while reducing model size, with the lightweight student model retaining high segmentation accuracy after substantial parameter compression.

In this paper, we introduce TCM-Ladder, the first unified multimodal question-answering benchmark for Traditional Chinese Medicine, consisting of over 52,000 questions across text, image, and video formats.

In this work, we introduce Ladder-base, a TCM-focused language model trained with Group Relative Policy Optimization on the textual subset of TCM-Ladder. The model is designed to improve reasoning and factual consistency for Traditional Chinese Medicine question answering.

In this study, we develop BenCao, a ChatGPT-based multimodal assistant for Traditional Chinese Medicine that integrates structured knowledge, diagnostic data, expert feedback, and external tools for tongue-image analysis.

This work presents ChatGPT-P3DB, a multimodal question-answering widget that connects ChatGPT with the Plant Protein Phosphorylation Database and uses multimodal LLMs to extract regulatory pathways from scientific figures.

We present a real-time oil spill assessment system that integrates fluorescence imaging, deep learning, a mobile app, and a data management platform for rapid field assessment.

This data release provides 1,530 fluorescence images of two crude oil types across concentrations from 0 to 500 mg/L, along with metadata describing oil type and concentration.

In this work, we developed a real-time COVID-19 surveillance system based on self-reported cases from Twitter, using large language models to automatically detect infections, symptoms, recoveries, and reinfections.

In this paper, we propose a universal tongue optimization model for tongue segmentation and developed an online tongue segmentation tool based on TOM.

This work presents Covlab, an online platform for monitoring COVID-19 trends and geographic spread using self-reported tweets.

We developed a portable device and deep learning model that estimate oil concentration in water using fluorescence images captured by an iPhone.

In this work, we reviewed recent advances in computerized tongue diagnosis for health assessment, covering tongue image acquisition, segmentation, feature extraction, color correction, and intelligent diagnosis systems.

@2025 Jiacheng Xie