Deep learning project
Budget: $250 – $750 USD
1. Problem Statement and Overview
Using clinical information and x-ray images, it is intended to predict the severity of COVID-19
infection by probabilistically predicting infected patients to be admitted to the ICU. Achieving this
objective will require implementing various techniques for handling missing data, appropriate
use and combination of the different types of data available, and the adequate selection of the
deep learning architecture.
2. Project Objectives
Implement a method capable of predicting if a patient infected by covid-19 will be admitted to
the ICU using their clinical information and X-ray images with high sensitivity (0.95) and good
specificity (0.9).
3. Suggested Methodology
A good start for approaching this problem would be implementing a siamese network
architecture utilizing the triplet loss function to learn how to differentiate between positive and
negative ICU labels given the x-ray image and clinical data.
I will give the dataset after we meet
Using clinical information and x-ray images, it is intended to predict the severity of COVID-19
infection by probabilistically predicting infected patients to be admitted to the ICU. Achieving this
objective will require implementing various techniques for handling missing data, appropriate
use and combination of the different types of data available, and the adequate selection of the
deep learning architecture.
2. Project Objectives
Implement a method capable of predicting if a patient infected by covid-19 will be admitted to
the ICU using their clinical information and X-ray images with high sensitivity (0.95) and good
specificity (0.9).
3. Suggested Methodology
A good start for approaching this problem would be implementing a siamese network
architecture utilizing the triplet loss function to learn how to differentiate between positive and
negative ICU labels given the x-ray image and clinical data.
I will give the dataset after we meet