AI-based analysis of CT images for rapid triage of COVID-19 patients

Abstract

The COVID-19 pandemic overwhelms the medical resources in the stressed intensive care unit (ICU) capacity and the shortage of mechanical ventilation (MV). We performed CT-based analysis combined with electronic health records and clinical laboratory results on Cohort 1 (n = 1662 from 17 hospitals) with prognostic estimation for the rapid stratification of PCR confirmed COVID-19 patients. These models, validated on Cohort 2 (n = 700) and Cohort 3 (n = 662) constructed from nine external hospitals, achieved satisfying performance for predicting ICU, MV, and death of COVID-19 patients (AUROC 0.916, 0.919, and 0.853), even on events happened two days later after admission (AUROC 0.919, 0.943, and 0.856). Both clinical and image features showed complementary roles in prediction and provided accurate estimates to the time of progression (p < 0.001). Our findings are valuable for optimizing the use of medical resources in the COVID-19 pandemic. The models are available here: https://github.com/terryli710/COVID_19_Rapid_Triage_Risk_Predictor .

Fig. 1. Illustration of workflow in this study.

a Our primary cohort (Cohort 1, n = 1662) for model development included patients from 17 hospitals, and our validation cohort (Cohort 2, n = 700) consisted of patients from 7 external and independent medical centers. In addition, we built a specific cohort (Cohort 3, n = 662) for patients from the 7 medical centers whose interval between admission and progression to critical outcomes (ICU/MV/death) were more than two days, aiming to evaluate the performance of our models on predicting events happening at least two days after admission. b Explanation of our data split and the corresponding usages. (1) Step one: feature visualization of Cohort 1 and Cohort 2 to get the preliminary intuitive sense; (2) Step two: 70% samples of Cohort 1 were picked as the training set using stratified sampling based on death cases, where fivefold cross-validation was used to tune the hyperparameters of the models; (3) Step three: model selection was performed on the remaining 30% samples of Cohort 1; (4) Step four: Cohort 2 and Cohort 3 were used to evaluate model performance in different aspects.

Fig. 2. Radiomics and clinical data heatmap.

Heatmap showing the prognostic performance of a radiomics data and b clinical data and R-score data on Cohort 2 with clustering of features. Hundred and fifty negative patients were randomly selected as well as all patients having outcomes of ICU admission, Mechanical Ventilation or Death to draw the heatmap. For patients with more than one adverse outcome, they will appear as samples in each corresponding category. The patients were grouped based on adverse outcomes (i.e., ICU admission, MV, and death) and whether the event occurred within 48 h after admission. The features were clustered within their categories to better visualize the data. The differences between negative outcome patients (yellow) and positive outcome patients can be seen from both (a) and (b), with some features showing different patterns for negative (patients discharged without any adverse outcomes) or positive patients (patients who required ICU, MV, or death while hospitalized). Almost all CT image features showed good discrimination between negative and severe outcome patients and had more obvious distinctions compared to clinical data. Among clinical data, lab results and demographics had good discriminating power. Part of radiologists’ score features had good discriminating power while clinical features have comparatively weak discriminating power. Regarding the distinctions between ICU admission, mechanical ventilation, and death, CT image features showed better discriminating power than clinical data. In CT image features, from ICU to MV to death, trends of value increasing or decreasing can be observed while in clinical data, this kind of trend is not visible.

Fig. 3. The model performances in the prediction of three outcomes (Cohort 2) and the ten most important features in the three outcome prediction tasks.

The first and second row presented ROC curves and PR curves for predicting three events of models based on different data types. a and d, b and e, c and f indicated that RadioClinLab based models for predicting ICU/MV/death achieved the highest AUROC (0.944/0.942/0.860) and AUPRC (0.665/0.551/0.346), respectively. g–i The ten most important features and their relative importance based on thirty bootstrapping experiments for the three prediction tasks based on the feature importance of the LightGBM classifiers.

Fig. 4. Kaplan–Meier curves for 3 tasks in Cohort 2.

Risk groups were divided according to model predicted scores. a ICU admission, b mechanical ventilation, and c death (high-risk: risk = 1, low-risk: risk = 0).

Fig. 5. Examples of lesion segmentation by the AI system.

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