A robust deep learning framework for cerebral microbleeds recognition in GRE and SWI MRI
Tahereh Hassanzadeh1, Sonal Sachdev2, Wei Wen3
1School of Computer Science and Engineering, University of New South Wales, Sydney, NSW 2052, Australia.
Abstract:
Cerebral microbleeds (CMB) are small hypointense lesions visible on gradient echo (GRE) or susceptibility-weighted (SWI) MRI, serving as critical biomarkers for various cerebrovascular and neurological conditions. Accurate quantification of CMB is essential, as their number correlates with the severity of conditions such as small vessel disease, stroke risk and cognitive decline. Current detection methods depend on manual inspection, which is time-consuming and prone to variability. Automated detection using deep learning presents a transformative solution but faces challenges due to the heterogeneous appearance of CMB, high false-positive rates, and similarity to other artefacts. This study investigates the application of deep learning techniques to public (ADNI and AIBL) and private datasets (OATS and MAS), leveraging GRE and SWI MRI modalities to enhance CMB detection accuracy, reduce false positives, and ensure robustness in both clinical and normal cases (i.e., scans without cerebral microbleeds). A 3D convolutional neural network (CNN) was developed for automated detection, complemented by a You Only Look Once (YOLO)-based approach to address false positive cases in more complex scenarios. The pipeline incorporates extensive preprocessing and validation, demonstrating robust performance across a diverse range of datasets. The proposed method achieves remarkable performance across four datasets, ADNI: Balanced accuracy: 0.953, AUC: 0.955, Precision: 0.954, Sensitivity: 0.920, F1-score: 0.930, AIBL: Balanced accuracy: 0.968, AUC: 0.956, Precision: 0.956, Sensitivity: 0.938, F1-score: 0.946, MAS: Balanced accuracy: 0.889, AUC: 0.889, Precision: 0.948, Sensitivity: 0.779, F1-score: 0.851, and OATS dataset: Balanced accuracy: 0.93, AUC: 0.930, Precision: 0.949, Sensitivity: 0.862, F1-score: 0.900. These results highlight the potential of deep learning models to improve early diagnosis and support treatment planning for conditions associated with CMB.
Insights
Deep learning accurately detects cerebral microbleeds (CMB) on MRI, improving early diagnosis of neurological conditions. This automated method enhances accuracy and reduces false positives, aiding treatment planning.
Area of Science:
- Neuroimaging and Artificial Intelligence
- Medical Diagnostics
- Cerebrovascular Diseases
Background:
- Cerebral microbleeds (CMB) are crucial MRI biomarkers for neurological conditions, but manual detection is inefficient and variable.
- Accurate CMB quantification is vital for assessing disease severity, stroke risk, and cognitive decline.
- Automated deep learning methods face challenges with CMB heterogeneity and false positives.
Purpose of the Study:
- To develop and validate a deep learning model for accurate and robust automated detection of cerebral microbleeds (CMB).
- To enhance CMB detection accuracy and reduce false positives using both 3D CNN and YOLO-based approaches.
- To ensure the model's performance across diverse public and private MRI datasets.
Main Methods:
- Utilized gradient echo (GRE) and susceptibility-weighted (SWI) MRI modalities from ADNI, AIBL, OATS, and MAS datasets.
- Developed a 3D convolutional neural network (CNN) for initial CMB detection.
- Integrated a You Only Look Once (YOLO)-based approach to refine detection and minimize false positives.
Main Results:
- The deep learning pipeline demonstrated high performance across all four datasets, with balanced accuracies ranging from 0.889 to 0.968.
- Achieved excellent AUC scores, indicating strong discriminative ability in CMB detection.
- Key metrics like precision, sensitivity, and F1-score were consistently high, underscoring the model's effectiveness.
Conclusions:
- Deep learning models show significant potential for improving the early diagnosis of conditions associated with cerebral microbleeds.
- The proposed automated detection method offers a robust and accurate solution, overcoming limitations of manual inspection.
- This advancement can support clinical decision-making and treatment planning for cerebrovascular and neurological disorders.
Related Concept Videos
01:24Magnetic Resonance Imaging
01:21Imaging Studies for Cardiovascular System IV: CMRI
01:27Imaging Studies IV: Magnetic Resonance Imaging
01:19Imaging Studies VII: Vascular Imaging
01:17Hemorrhagic Stroke l: Introduction