Related Experiment Video
Updated: Nov 1, 2025
Evaluation of the Cognitive Performance of Hypertensive Patients with Silent Cerebrovascular Lesions
Published on: April 23, 2021
Novel Approaches to Detection of Cerebral Microbleeds: Single Deep Learning Model to Achieve a Balanced Performance
Min Jae Myung1, Kyung Mi Lee1, Hyug-Gi Kim1
1Department of Radiology, Kyung Hee University College of Medicine, Kyung Hee University Hospital, #23 Kyungheedae-ro, Dongdaemun-gu, Seoul 02447, Republic of Korea.
Purpose:
Cerebral microbleeds (CMBs) are considered essential indicators for the diagnosis of cerebrovascular disease and cognitive disorders. Traditionally, CMBs are manually interpreted based on criteria including the shape, diameter, and signal characteristics after an MR examination, such as susceptibility-weighted imaging or gradient echo imaging (GRE). In this paper, an efficient method for CMB detection in GRE scans is presented.
Materials And Methods:
The proposed framework consists of the following phases: (1) pre-processing (skull extraction), (2) the first training with the ground truth labeled using CMB, (3) the second training with the ground truth labeled with CMB mimicking the same subjects, and (4) post-processing (cerebrospinal fluid (CSF) filtering). The proposed technique was validated on a dataset of 1133 CBMs that consisted of 5284 images for training and 1737 images for testing. We applied a two-stage approach using a region-based CNN method based on You Only Look Once (YOLO) to investigate a novel CMB detection technique.
Results:
The sensitivity, precision, F1-score and false positive per person (FPavg) were evaluated as 80.96, 60.98, 69.57 and 6.57, 59.69, 62.70, 61.16 and 4.5, 66.90, 79.75, 72.76 and 2.15 for YOLO with a single label, YOLO with double labels, and YOLO + CSF filtering, respectively, and YOLO + CSF filtering showed the highest precision performance, F1-score and lowest FPavg.
Conclusions:
Using proposed framework, we developed an optimized CMB learning model with low false positives and a balanced performance in clinical practice.
Insights
This study presents an efficient YOLO-based framework for detecting cerebral microbleeds (CMBs) in gradient echo scans. The YOLO + CSF filtering method achieved the best precision and lowest false positives, optimizing CMB detection.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Machine Learning
Background:
- Cerebral microbleeds (CMBs) are crucial indicators for cerebrovascular diseases and cognitive disorders.
- Traditional CMB detection relies on manual interpretation of MR images like GRE scans.
- An efficient and automated method for CMB detection is needed.
Purpose of the Study:
- To develop an efficient method for detecting cerebral microbleeds (CMBs) in gradient echo (GRE) scans.
- To optimize CMB detection using a region-based Convolutional Neural Network (CNN) approach.
- To improve the accuracy and reduce false positives in CMB identification.
Main Methods:
- A four-phase framework was proposed: skull extraction, two-stage training using a region-based CNN (YOLO), and cerebrospinal fluid (CSF) filtering.
- The model was trained and tested on a large dataset of 5284 and 1737 images, respectively.
- Performance was evaluated using sensitivity, precision, F1-score, and average false positives per person (FPavg).
Main Results:
- The YOLO + CSF filtering approach demonstrated superior performance with the highest precision (79.75) and F1-score (72.76).
- This method also achieved the lowest average false positives per person (FPavg) of 2.15.
- Compared to single and double label YOLO, the YOLO + CSF filtering significantly improved detection accuracy.
Conclusions:
- The developed framework provides an optimized CMB learning model for clinical practice.
- The YOLO + CSF filtering method offers a balanced performance with low false positives.
- This automated approach enhances the efficiency and reliability of CMB detection in GRE scans.
