Related Experiment Video
Updated: Dec 3, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
A Deep Learning Prognosis Model Help Alert for COVID-19 Patients at High-Risk of Death: A Multi-Center Study
Insights
A new AI tool, De-COVID19-Net, uses CT scans to predict high-risk coronavirus disease (COVID-19) patients. This prognostic tool identifies patients likely to die within 14 days, enabling early intervention.
Area of Science:
- Artificial Intelligence in Medicine
- Medical Imaging Analysis
- Prognostic Modeling
Background:
- Coronavirus disease (COVID-19) poses a significant global health challenge.
- Identifying high-risk patients is crucial for effective treatment planning.
- Existing prognostic tools require improvement for severe and critical COVID-19 cases.
Purpose of the Study:
- To develop and validate a prognostic tool for predicting short-term mortality in severe or critical COVID-19 patients.
- To leverage computed tomography (CT) imaging and clinical data for risk stratification.
- To create an AI-driven model for early identification of high-risk COVID-19 individuals.
Main Methods:
- Retrospective collection of data from 366 severe/critical COVID-19 patients across four centers.
- Development of a 3D densely connected convolutional neural network (De-COVID19-Net).
- Integration of CT scan and clinical information for model training and testing.
- Evaluation using Area Under the Curve (AUC) and Kaplan-Meier analysis.
Main Results:
- De-COVID19-Net achieved high predictive performance with an AUC of 0.952 (training set) and 0.943 (test set).
- Model accuracy was consistent across different demographics (age, sex) and comorbidities.
- Kaplan-Meier analysis confirmed the model's ability to significantly differentiate high-risk from low-risk patients (p < 0.001).
Conclusions:
- De-COVID19-Net demonstrates impressive non-invasive prognostic capability using initial CT scans.
- The AI model can accurately predict short-term mortality in severe/critical COVID-19 patients.
- This tool offers potential for early risk alerts and timely clinical intervention.
Abstract:
Since its outbreak in December 2019, the persistent coronavirus disease (COVID-19) became a global health emergency. It is imperative to develop a prognostic tool to identify high-risk patients and assist in the formulation of treatment plans. We retrospectively collected 366 severe or critical COVID-19 patients from four centers, including 70 patients who died within 14 days (labeled as high-risk patients) since their initial CT scan and 296 who survived more than 14 days or were cured (labeled as low-risk patients). We developed a 3D densely connected convolutional neural network (termed De-COVID19-Net) to predict the probability of COVID-19 patients belonging to the high-risk or low-risk group, combining CT and clinical information. The area under the curve (AUC) and other evaluation techniques were used to assess our model. The De-COVID19-Net yielded an AUC of 0.952 (95% confidence interval, 0.928-0.977) on the training set and 0.943 (0.904-0.981) on the test set. The stratified analyses indicated that our model's performance is independent of age, sex, and with/without chronic diseases. The Kaplan-Meier analysis revealed that our model could significantly categorize patients into high-risk and low-risk groups (p < 0.001). In conclusion, De-COVID19-Net can non-invasively predict whether a patient will die shortly based on the patient's initial CT scan with an impressive performance, which indicated that it could be used as a potential prognosis tool to alert high-risk patients and intervene in advance.
