Multimodality artificial intelligence for cardiac diagnosis and prevention
ABOUT THE PROJECT
Project Description
The Tison lab leverages multi-modal medical data streams--such as data from cardiac ultrasound (echocardiograms), ECGs, electronic health record, photoplethysmography and other remote sensors--to perform cardiac phenotyping, diagnosis and disease prevention. This project aims to build a platform for machine learning-based interpretation of multiple data sources including at least echocardiogram, ECG and electronic health record-derived data. This will be performed first for discrete demonstration diseases including heart failure, pulmonary hypertension and hypertrophic cardiomyopathy, while adhering to the larger vision that the platform be broadly applicable across various cardiac phenotypes.
Funding
Might or might not
LOOKING FOR
Required Skills
- Python
- R
- SQL
- Git/Github
- HPC/GPU computing
- Apache Spark
- TensorFlow
- PyTorch
- MLlib / machine learning tools
- Natural Language Processing tools (NLTK, CTAKES, PyTorch-NLP, etc)
Required Course Work or Level of Knowledge
- Statistics, intermediate
- Machine Learning, intermediate - advanced
Acceptable Level of Education (eg. Undergrad, Grad Students, Post Docs, MD, PhD)
Undergraduate students, Graduate students, Postgraduates, Full-time workers who volunteer time
CONTACT
PI/Research group
Geoff Tison, MD, MPH
Contact
Geoff Tison MD MPH, [email protected]