Related Experiment Video
Updated: Jul 29, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
Severe/critical COVID-19 early warning system based on machine learning algorithms using novel imaging scores.
Qiu-Yu Li1, Zhuo-Yu An2, Zi-Han Pan1
1Department of Respiratory and Critical Care Medicine, Peking University Third Hospital, Beijing 100191, China.
World Journal of Clinical Cases
|May 22, 2023
Summary
Early identification of severe coronavirus disease 2019 (COVID-19) is vital. This study developed a prediction model using chest CT scans and clinical data to effectively identify severe COVID-19 cases early.
Area of Science:
- Medical imaging and diagnostics
- Computational biology and bioinformatics
- Infectious disease epidemiology
Background:
- Early identification of severe/critical COVID-19 is crucial for timely intervention.
- Chest CT scores are significant in pneumonia diagnosis but effective early warning systems for severe COVID-19 based on dynamic CT evolution are lacking.
Purpose of the Study:
- To develop a severe/critical COVID-19 prediction model.
- The model integrates imaging scores, clinical features, and biomarker levels for enhanced early detection.
Main Methods:
- Utilized an improved scoring system for chest CT characteristics in COVID-19 patients.
- Incorporated clinical indicators like dyspnea, oxygen saturation, ALT, and AST.
- Employed lasso regression to identify significant disease characteristics and machine learning algorithms (XGBClassifier, Logistic Regression, MLPClassifier, RandomForestClassifier, AdaBoost Classifier) to build the prediction model.
Main Results:
- Key predictors for severe/critical COVID-19 included blood oxygen saturation, ALT, IL-6/IL-10, combined CT score, ground glass opacity score, age, crazy paving mode score, qSOFA, AST, and overall lung involvement score.
- An early warning system for severe/critical COVID-19 was established using multiple machine learning algorithms.
- The prediction model demonstrated feasibility for early detection of severe/critical COVID-19 evolution.
Conclusions:
- A prediction model integrating improved CT scores and machine learning algorithms effectively detects early warning signals of severe/critical COVID-19.
- This approach offers a feasible method for timely intervention in severe COVID-19 cases.
