NA-MIC Project Weeks

Back to Projects List

Clinical trajectory visualization based on edema quantification from chest x-ray

Key Investigators

• Ruizhi Liao (MIT)
• Steve Pieper (ISOMICS)
• Polina Golland (MIT)

Project Description

We have developed machine learning algorithms to automatically quantify the severity of pulmonary edema from chest x-rays on a continuous scale. The resulting assessment can be used for visualization of heart failure patient recovery trajectories in prior episodes of heart failure to support physicians with a data-driven approach to treating patients. We would like to investigate different ways of visualizing the clinical tracjectories that can inform physicians how different patients responded to different medications/treatment plans.

ML algorithms for quantifying pulmonary edema in chest x-ray: https://www.csail.mit.edu/research/chest-x-ray-analysis

Example x-ray images:

Approach

1. Look into Crossfilter: http://square.github.io/crossfilter/
2. Look into Vega-lite: https://vega.github.io/vega-lite/
3. Look into i2b2: https://www.i2b2.org/
4. …

Progress

1. Collected visualization ideas and option from several investigators
2. Developed a prototyping plan

Next steps

1. Implement some visualization ideas using a few patient datasets
2. Iterate with clinical colleagues on best implementation options

Background and References

ML algorithms for quantifying pulmonary edema in chest x-ray: https://www.csail.mit.edu/research/chest-x-ray-analysis