AI based analysis of imaging, clinical & lab data to triage COVID19 patients.

Heatmap of radiomics features of COVID19 patients showing the reatlinships between severity (ICU admission, Machine Ventilator use (MV) and death). 

We took advantage of a large cohort of Chinese PCR-confirmed COVID19 patients to embark on a biomedical data fusion project to triage COVID19 patients. We develop a machine learning approach for risk stratification based on CT-based radiomics features and clinical data for COVID-19 patients in terms of stable or severe disease (requiring ICU) on admission, then developed specific outcome prediction (MV/death) models for critically ill patients. We also provided insights into estimating time to the progression (ICU/MV/death) for COVID-19 patients. Furthermore, we identified key clinical prognostic indicators and CT-based predictive features for outcome prediction (ICU/MV/death). This work is valuable for delivering timely treatment and optimizing the use of limited medical resources during the COVID-19 pandemic.



 

CT images of COVID19 patients: Left (a, c, e): original images; Right (b, d, f): pulmonary lobes (colored lines) and opacities segmentation (blue area).

The models are available here: https://github.com/terryli710/COVID_19_Rapid_Triage_Risk_Predictor.

You can find more thoughts on this work and biomedical data fusion in the context of quantitative imaging here. 

Reference: 

Xu Q, Zhan X, Zhou Z, Li Y, Xie P, Zhang S, Li X, Yu Y, Zhou C, Zhang L, Gevaert O, Lu G. AI-based analysis of CT images for rapid triage of COVID-19 patients. NPJ Digit Med. 2021 Apr 22;4(1):75. doi: 10.1038/s41746-021-00446-z. PMID: 33888856; PMCID: PMC8062628.