MRI multi-sequence deep learning integration with clinical profiles for pediatric viral encephalitis diagnosis
Keyu Lu1, Ruying Liang2, Jinlian Che1
1Department of Radiology, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region, Xiangzhu Avenue, Nanning, 530021, China.
Scientific Reports
|November 18, 2025
Summary
A new AI model combining clinical data and MRI scans accurately diagnoses pediatric viral encephalitis. This fusion model offers improved accuracy and sensitivity, aiding early clinical decision-making for this challenging CNS infection.
Area of Science:
- Neurology
- Infectious Diseases
- Medical Imaging
- Artificial Intelligence
Background:
- Pediatric viral encephalitis (VE) presents diagnostic challenges due to varied symptoms and limitations of traditional methods.
- Accurate and early diagnosis is crucial for effective clinical management of VE.
- Advanced neuroimaging and machine learning offer potential for improved diagnostic tools.
Purpose of the Study:
- To develop and evaluate a novel clinical-imaging fusion model for diagnosing pediatric viral encephalitis (VE).
- To integrate clinical factors with advanced magnetic resonance imaging (MRI) deep features for enhanced diagnostic accuracy.
- To assess the clinical utility and application value of the developed fusion model.
Main Methods:
- Retrospective analysis of 525 pediatric patients (VE vs. non-VE groups) using clinical data and multi-sequence MRI (T1, T2, DWI).
- Logistic regression for identifying independent clinical factors associated with VE.
- Convolutional neural networks (CNNs) for extracting deep MRI features, followed by LASSO for dimensionality reduction and fusion model construction using machine learning.
Main Results:
- Independent clinical factors for VE included fever, white blood cell (WBC) count, and C-reactive protein (CRP) levels.
- The Logistic Regression (LR) classifier showed strong performance for MRI deep features (AUC up to 0.934).
- The clinical-imaging fusion model achieved high diagnostic performance (AUC up to 0.985 in training, 0.934 in testing), with superior accuracy, sensitivity, and specificity compared to MRI features alone.
Conclusions:
- The developed clinical-imaging fusion model demonstrates significant diagnostic efficacy for pediatric viral encephalitis.
- Integrating deep MRI features with clinical data provides a powerful, non-invasive tool for early VE diagnosis.
- This fusion model holds considerable promise for improving clinical decision-making in pediatric neurology.
Related Concept Videos
01:24Magnetic Resonance Imaging
9.0K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
9.0K
01:27Imaging Studies IV: Magnetic Resonance Imaging
216
Introduction:Magnetic Resonance Imaging, or MRI, can include a specialized imaging technique of the urinary system known as Magnetic Resonance Urography (MRU). This radiation-free technique uses strong magnetic fields and radio waves to produce detailed images with the help of a computer. MRU is particularly effective for visualizing fluid-filled structures like the kidneys, ureters, and bladder.Applications of MRI in the Genitourinary SystemKidneys and Ureters: MRI detects tumors, cysts,...
216