Related Experiment Video
Updated: Sep 19, 2025
Image Acquisition using Portable Sonography for Emergency Airway Management
Published on: September 28, 2022
Deep learning detects retropharyngeal edema on MRI in patients with acute neck infections
Oona Rainio1, Heidi Huhtanen2, Jari-Pekka Vierula2
1Turku PET Centre, University of Turku and Turku University Hospital, Turku, Finland. ormrai@utu.fi.
Background:
In acute neck infections, magnetic resonance imaging (MRI) shows retropharyngeal edema (RPE), which is a prognostic imaging biomarker for a severe course of illness. This study aimed to develop a deep learning-based algorithm for the automated detection of RPE.
Methods:
We developed a deep neural network consisting of two parts using axial T2-weighted water-only Dixon MRI images from 479 patients with acute neck infections annotated by radiologists at both slice and patient levels. First, a convolutional neural network (CNN) classified individual slices; second, an algorithm classified patients based on a stack of slices. Model performance was compared with the radiologists' assessment as a reference standard. Accuracy, sensitivity, specificity, and area under receiver operating characteristic curve (AUROC) were calculated. The proposed CNN was compared with InceptionV3, and the patient-level classification algorithm was compared with traditional machine learning models.
Results:
Of the 479 patients, 244 (51%) were positive and 235 (49%) negative for RPE. Our model achieved accuracy, sensitivity, specificity, and AUROC of 94.6%, 83.3%, 96.2%, and 94.1% at the slice level, and 87.4%, 86.5%, 88.2%, and 94.8% at the patient level, respectively. The proposed CNN was faster than InceptionV3 but equally accurate. Our patient classification algorithm outperformed traditional machine learning models.
Conclusion:
A deep learning model, based on weakly annotated data and computationally manageable training, achieved high accuracy for automatically detecting RPE on MRI in patients with acute neck infections.
Relevance Statement:
Our automated method for detecting relevant MRI findings was efficiently trained and might be easily deployed in practice to study clinical applicability. This approach might improve early detection of patients at high risk for a severe course of acute neck infections.
Key Points:
Deep learning automatically detected retropharyngeal edema on MRI in acute neck infections. Areas under the receiver operating characteristic curve were 94.1% at the slice level and 94.8% at the patient level. The proposed convolutional neural network was lightweight and required only weakly annotated data.
Insights
A deep learning model accurately detects retropharyngeal edema (RPE) on MRI scans for acute neck infections. This automated method aids in early identification of patients at high risk for severe illness.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- Retropharyngeal edema (RPE) on MRI indicates severe acute neck infections.
- Accurate RPE detection is crucial for prognostic assessment.
- Current detection relies on manual radiologist interpretation.
Purpose of the Study:
- To develop and validate a deep learning algorithm for automated RPE detection on MRI.
- To improve the efficiency and accuracy of identifying RPE in acute neck infections.
- To provide a tool for early risk stratification in patients with neck infections.
Main Methods:
- A two-part deep neural network was developed using T2-weighted MRI images from 479 patients.
- A convolutional neural network (CNN) performed slice-level classification.
- A patient-level algorithm aggregated slice classifications for final assessment.
- Model performance was benchmarked against radiologist assessments and traditional machine learning models.
Main Results:
- The deep learning model achieved high accuracy in detecting RPE at both slice (94.1% AUROC) and patient levels (94.8% AUROC).
- The proposed CNN was computationally efficient, outperforming InceptionV3 in speed.
- The patient-level algorithm surpassed traditional machine learning models in performance.
- The model demonstrated high sensitivity and specificity in identifying RPE.
Conclusions:
- A deep learning model can accurately and automatically detect RPE on MRI in acute neck infections.
- The algorithm, trained on weakly annotated data, is computationally manageable and potentially deployable.
- This automated approach can enhance early detection of severe infection courses.
- The findings suggest improved clinical applicability for RPE detection in managing acute neck infections.
More Related Videos
Related Concept Videos
01:24Magnetic Resonance Imaging
01:30Acute Pharyngitis
Acute pharyngitis is the inflammation of the back of the throat (pharynx), commonly resulting in a sore throat. It is a frequently encountered condition that prompts individuals to seek medical advice.
Classification
Acute pharyngitis can be categorized based on its underlying cause:
01:21Imaging Studies for Cardiovascular System IV: CMRI
01:30Tonsillitis I: Introduction
Etiology
Three primary contributing factors have been identified.