Related Experiment Video
Updated: Nov 9, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Detecting cerebral microbleeds via deep learning with features enhancement by reusing ground truth
Tianfu Li1, Yan Zou2, Pengfei Bai3
1Guangdong Provincial Key Laboratory of Optical Information Materials and Technology & Institute of Electronic Paper Displays, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006, China.
Background And Objectives:
Cerebral microbleeds (CMBs) are cerebral small vascular diseases and are often used to diagnose symptoms such as stroke and dementia. Manual detection of cerebral microbleeds is a time-consuming and error-prone task, so the application of microbleed detection algorithms based on deep learning is of great significance. This study presents the feature enhancement technology applying to improve the performances of detecting CMBs. The primary purpose of the feature enhancement is emphasizing the meaningful features, leading deep learning network easier and correctly to optimize.
Method:
In this study, we applied feature enhancement in detecting CMBs from brain MRI images. Feature enhancement enhanced specific intervals and suppressed the useless intervals of the feature map. This method was applied in SSD-512 and SSD-300 algorithm, using VGG architecture pre-trained in the ImageNet dataset.
Results:
The proposed method was applied in SSD-512. Moreover, the model was trained and tested on the sequence of SWAN images of brain MRI images. The results of the experiment demonstrate that our method effectively improves the detection performance of the SSD network in detecting CMBs. We train SSD-512 120000 iterations and test results on the test datasets, by applying the feature enhancement layer, improving the precision with 3.3% and the mAP of 2.3%. In the same way, we trained SSD-300, improving the mAP of 2.0%. 2.8% and 7.4% precision are improved by applying feature enhancement layer In ResNet-34 and MobileNet.
Conclusions:
The proposed method achieved more effective performance, demonstrated that feature enhancement can be a helpful algorithm to enhance the deep learning model.
Insights
Feature enhancement technology improves deep learning models for detecting cerebral microbleeds (CMBs) in brain MRI scans. This method boosts precision and mean average precision (mAP), aiding in diagnosing conditions like stroke and dementia.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neurology
Background:
- Cerebral microbleeds (CMBs) are indicative of small vascular diseases, crucial for diagnosing stroke and dementia.
- Manual detection of CMBs is labor-intensive and prone to errors.
- Deep learning algorithms offer significant potential for automated CMB detection.
Purpose of the Study:
- To introduce and evaluate a feature enhancement technology for improving CMB detection using deep learning.
- To emphasize meaningful features and suppress irrelevant ones in feature maps for better model optimization.
Main Methods:
- Feature enhancement was applied to the SSD-512 and SSD-300 algorithms, utilizing a VGG architecture pre-trained on ImageNet.
- The technique selectively enhances informative feature map intervals while suppressing noise.
- Models were trained and tested on SWAN brain MRI image sequences.
Main Results:
- The feature enhancement method improved precision by 3.3% and mAP by 2.3% for SSD-512.
- SSD-300 demonstrated an improved mAP of 2.0% with the enhancement.
- Precision gains of 2.8% and 7.4% were observed for ResNet-34 and MobileNet, respectively.
Conclusions:
- Feature enhancement is an effective technique for improving deep learning-based CMB detection.
- The proposed method offers a valuable tool for enhancing the performance of CMB detection models.
More Related Videos
14:08Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
12:50Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
Published on: April 14, 2014