Zhihua Tian

M.S. Candidate in Operations Research and Cybernetics · Nankai University

Zhihua Tian portrait

AI4S · Computer Vision · Large Language Models

My research focuses on building clinically meaningful AI systems for movement disorder assessment, with particular interest in Parkinson’s disease analysis from smartphone videos and sequential human motion features.

I work on problems where robustness, interpretability, and reproducibility matter as much as raw accuracy, especially when the final system is meant to survive contact with real clinical workflows.

email github google scholar cv

selected publications

  1. Deep learning-enabled accurate assessment of gait impairments in Parkinson's disease using smartphone videos
    Jianda Han* , Zhihua Tian* , Jialing Wu* , Kai Zhang , Shaohua Li , Fahd Baig , Peipei Liu , Ravi Vaidyanathan , Francesca Morgante , Weiguang Huo
    * Equal contribution
    npj Digital Medicine, 2025 IF 15.1
    Source Code
  2. IMU2SKE: Hierarchical Contrastive Learning-Empowered Video-Based Detection of Freezing of Gait in Parkinson's Disease
    Zhihua Tian , Yanwei Zhao , Jianda Han , Weiguang Huo
    In IEEE CBS, 2025
    Source Code

projects

Voice agent thumbnail

Voice agent

A full-duplex, real-time voice agent for hospital appointment workflows, with interruptible interaction and an LLM backend.

View project
Website
romanti.cc thumbnail

romanti.cc

An education-email-based campus social platform centered on signal-based matching, direct chat, and trusted student identity.

View project
Website
FAGI-PD thumbnail

FAGI-PD

FAGI-PD is an intelligent gait analysis platform for Parkinson’s disease screening and management, built on our research published in npj Digital Medicine …

View project
Website