Peripheral Edema Phenotyping Using Artificial Intelligence

Peripheral edema is the most common symptom of heart failure. Phenotyping of and continuous monitoring of edema provides critical clinical information and can be used for averting episodes of acute decompensation and hospitalizations. Students will do research on applying video-based AI techniques to measure, estimate and track edema grades. They will use videos of skin during the edema pitting-test, acquired from models and from real patients. They will design and experiment with deep learning models and video preprocessing techniques. The focus of the work during the spring of 2020 will be data augmentation based on videos acquired from patients.

Name of Lab: Kostic lab

Dates: 6/1/2020 - 8/21/2020

Direct Supervisor: Zoran Kostic

Paid: Yes

Credit: Yes

Hours per week: 20

Number of positions: 1

 

Skills required: signal processing, video processing, deep learning frameworks, software engineering, real-time embedded coding.

Eligibility: Junior, Senior, Master's

Zoran Kostic, zk2172@columbia.edu