Name
Title | ||
|---|---|---|
Coronary Artery Fibrous Plaque Detection Based on Multi-Scale Convolutional Neural Networks |
Abstract | ||
|---|---|---|
One of the major causes of the coronary heart disease is vascular stenosis and thrombosis that is generally caused by development of fibrous plaques. Therefore, detection of a fibrous plaque in coronary arteries for the diagnosis and treatment of coronary heart disease is of clinical significance. Technical challenges are in reading the optical coherence tomography (OCT) images which is tedious and inaccurate. In response, we propose an automated coronary artery fibrous plaque detection method based on deep learning with Convolutional Neural Networks (CNN). We present our novel techniques of identifying a contracting path to capture the context and a symmetric expanding path that enables the precise localization. The algorithm utilizes the features of the contracting path and the expanding path, so that the merged features can present the context and accurate localization, and uses the multi-scale feature maps for detection. Experimental results show that the proposed method achieved a coincidence of 91.04%, accuracy of 94.12%, and recall of 94.12%. Compared with the previously published work the proposed method is advantageous in both accuracy and robustness. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1007/s11265-019-01501-5 | Journal of Signal Processing Systems |
Keywords | Field | DocType |
Coronary heart disease, Fibrous plaque, Optical coherence tomography, Multi-scale convolution neural networks | Vascular Stenosis,Artery,Optical coherence tomography,Coronary arteries,Pattern recognition,Convolutional neural network,Computer science,Robustness (computer science),Artificial intelligence,Deep learning,Machine learning,Heart disease | Journal |
Volume | Issue | ISSN |
92 | 3 | 1939-8018 |
Citations | PageRank | References |
1 | 0.36 | 0 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xiu-Ling Liu | 1 | 9 | 2.16 |
| Jiaxing Du | 2 | 1 | 0.36 |
| Jianli Yang | 3 | 1 | 0.36 |
| Peng Xiong | 4 | 28 | 3.80 |
| Jing Liu | 5 | 250 | 27.30 |
| Feng Lin | 6 | 1 | 0.36 |