View Proposal

Proposer
Chengjia Wang
Title
XAI in the Prediction of COVID-19 Clinical Outcome
Goal
This project will develop deep learning methods that can directly benefit the efficiency and accuracy of clinical analysis for COVID-19
Description
SARS-CoV-2 pandemic has more than 1.6 million deaths worldwide by the end of 2020 and has overwhelmed health care resources in most countries. Medical imaging, especially chest CT and X-ray techniques have played critical roles in the diagnosis and treatment planning of COVID-19. In the past two years, a large number of deep learning methods have been proposed to: 1. assist the analysis and post-processing of chest imaging data; 2. predict the possible clinical outcomes and development of disease; 3. predict the spreading speed and pandemic status in human society; 4. etc.. This project will develop deep learning methods that can directly benefit the efficiency and accuracy of clinical analysis for COVID-19 using the available public challenge dataset (the STOIC2021 competition: https://stoic2021.grand-challenge.org/stoic2021/). Then focus on the analysis and assessment of explainability of currently mainstream models (ResNet variations: ConvNeXt, Transformer, gMLP, GNN, etc.) Purposes and milestones: Specifically, this project will develop DL models for to predict: 1. Predict COVID19 positivity. 2. Predict occurance of severe COVID-19 cases, defined as intubation or death within one month from the acquisition of the CT scan (metric: AUC). Milestones of this project: 1. A simple ConvNeXt model applied to STOIC2021 data and produce result (successfully submit to the competition) 2. review, collect, implement and compare the SOTA deep learning models on STOIC2021 models, you may use some extra data 3. review and implement different ways to assess the explainability of different models 4. Assess the explainability for different models
Resources
https://arxiv.org/abs/2201.03545 https://pubmed.ncbi.nlm.nih.gov/34184935/
Background
python, deep learning basics
Url
External Link
Difficulty Level
Easy
Ethical Approval
None
Number Of Students
2
Supervisor
Chengjia Wang
Keywords
machine learning, neural network, deep learning, ai, computer vision, explainability
Degrees
Bachelor of Science in Computer Science
Bachelor of Science in Computer Systems
Bachelor of Science in Information Systems
Master of Engineering in Software Engineering
Master of Science in Artificial Intelligence
Master of Science in Artificial Intelligence with SMI
Master of Science in Business Information Management
Master of Science in Computing (2 Years)
Master of Science in Data Science
Master of Science in Human Robot Interaction
Master of Science in Information Technology (Software Systems)
Master of Science in Robotics
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
Bachelor of Science in Computer Science (Cyber Security)
Back to List