Visual Analytics for Explainable Artificial Intelligence on Parkinson's Disease Diagnosis

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Defense

Title: "Visual Analytics for Explainable Artificial Intelligence on 
        Parkinson's Disease Diagnosis"

by

PAN Linfei

Abstract:

With the wide prevalence of Parkinson's Disease and the hardship in 
concrete diagnosing this disease in the early stage, some assistant 
diagnosis methods can be of great value. Pioneering research has found the 
comparatively low level of brain connectivity in the patients with 
Parkinson's Disease which can be detected by various machine learning 
models. To better communicate such findings with domain experts, this 
project aims to serve as the communication bridge between the model and 
domain experts. This project consists of three parts: a machine learning 
model for classification and feature importance ranking, a visual system 
to demonstrate the findings in the machine learning model, and a user 
study. (subject to change)


Date            : 15 May 2020 (Friday)

Time            : 10:00 - 10:40

Zoom Meeting    : https://hkust.zoom.us/j/141521296

Advisor         : Prof. QU Huamin

2nd Reader      : Dr. MA Xiaojuan