1
|
Chu M, De Maria GL, Dai R, Benenati S, Yu W, Zhong J, Kotronias R, Walsh J, Andreaggi S, Zuccarelli V, Chai J, Channon K, Banning A, Tu S. DCCAT: Dual-Coordinate Cross-Attention Transformer for thrombus segmentation on coronary OCT. Med Image Anal 2024; 97:103265. [PMID: 39029158 DOI: 10.1016/j.media.2024.103265] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2024] [Revised: 06/02/2024] [Accepted: 07/01/2024] [Indexed: 07/21/2024]
Abstract
Acute coronary syndromes (ACS) are one of the leading causes of mortality worldwide, with atherosclerotic plaque rupture and subsequent thrombus formation as the main underlying substrate. Thrombus burden evaluation is important for tailoring treatment therapy and predicting prognosis. Coronary optical coherence tomography (OCT) enables in-vivo visualization of thrombus that cannot otherwise be achieved by other image modalities. However, automatic quantification of thrombus on OCT has not been implemented. The main challenges are due to the variation in location, size and irregularities of thrombus in addition to the small data set. In this paper, we propose a novel dual-coordinate cross-attention transformer network, termed DCCAT, to overcome the above challenges and achieve the first automatic segmentation of thrombus on OCT. Imaging features from both Cartesian and polar coordinates are encoded and fused based on long-range correspondence via multi-head cross-attention mechanism. The dual-coordinate cross-attention block is hierarchically stacked amid convolutional layers at multiple levels, allowing comprehensive feature enhancement. The model was developed based on 5,649 OCT frames from 339 patients and tested using independent external OCT data from 548 frames of 52 patients. DCCAT achieved Dice similarity score (DSC) of 0.706 in segmenting thrombus, which is significantly higher than the CNN-based (0.656) and Transformer-based (0.584) models. We prove that the additional input of polar image not only leverages discriminative features from another coordinate but also improves model robustness for geometrical transformation.Experiment results show that DCCAT achieves competitive performance with only 10% of the total data, highlighting its data efficiency. The proposed dual-coordinate cross-attention design can be easily integrated into other developed Transformer models to boost performance.
Collapse
Affiliation(s)
- Miao Chu
- Biomedical Instrument Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China; Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK
| | - Giovanni Luigi De Maria
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; National Institute for Health Research, Oxford Biomedical Research Centre, UK.
| | - Ruobing Dai
- Biomedical Instrument Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
| | - Stefano Benenati
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; University of Genoa, Genoa, Italy
| | - Wei Yu
- Biomedical Instrument Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
| | - Jiaxin Zhong
- Biomedical Instrument Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China; Department of Cardiology, Fujian Medical University Union Hospital, Fujian, China
| | - Rafail Kotronias
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; National Institute for Health Research, Oxford Biomedical Research Centre, UK
| | - Jason Walsh
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; National Institute for Health Research, Oxford Biomedical Research Centre, UK
| | - Stefano Andreaggi
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiology, Department of Medicine, University of Verona, Italy
| | | | - Jason Chai
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK
| | - Keith Channon
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; National Institute for Health Research, Oxford Biomedical Research Centre, UK
| | - Adrian Banning
- Oxford Heart Centre, Oxford University Hospitals NHS Trust, UK; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK; National Institute for Health Research, Oxford Biomedical Research Centre, UK
| | - Shengxian Tu
- Biomedical Instrument Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China; Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, UK.
| |
Collapse
|
2
|
Ishii M, Nakamura T, Tsujita K. Intravascular Imaging and Coronary Microvascular Dysfunction After Percutaneous Coronary Intervention in ST-Segment Elevation Myocardial Infarction. Circ J 2023; 87:1633-1634. [PMID: 37460314 DOI: 10.1253/circj.cj-23-0437] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 10/27/2023]
Affiliation(s)
- Masanobu Ishii
- Department of Medical Information Science, Graduate School of Medical Sciences, Kumamoto University
- Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kumamoto University
| | - Taishi Nakamura
- Department of Medical Information Science, Graduate School of Medical Sciences, Kumamoto University
| | - Kenichi Tsujita
- Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kumamoto University
| |
Collapse
|
3
|
Computational Analysis of the Related Factors of Deep Vein Thrombosis (DVT) Formation in Patients Undergoing Hip Fracture Surgery. EVIDENCE-BASED COMPLEMENTARY AND ALTERNATIVE MEDICINE 2022; 2022:1127095. [PMID: 35668774 PMCID: PMC9166936 DOI: 10.1155/2022/1127095] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 05/07/2022] [Accepted: 05/14/2022] [Indexed: 12/04/2022]
Abstract
A retrospective study was conducted on 51 patients undergoing hip fracture surgery to investigate the factors associated with the formation of deep venous thrombosis (DVT). The independent sample t-test and correlation analysis were used to sort out and analyze the data. The findings are as follows. (1) Different gender samples showed significant differences in the Caprini score and thrombus location. Most DVTs in females are located in the posterior tibial vein and intermuscular veins. The Caprini score of females was significantly higher than that of males. (2) Age displays a positive correlation with DVT, coronary heart disease, hypertension, and different surgical types, respectively. (3) There is a correlation between age and operation duration. (4) Hyperlipidemia and cerebrovascular disease show a positive correlation with DVT. (5) There was a significant negative correlation between the Caprini score and the quantification of D-dimer. This indicates that in this sample, the higher the patients' Caprini score is, the lower the quantitation of D-dimer will be. (6) Hyperlipidemia and cardiac insufficiency show a positive correlation with cerebrovascular disease. Patients with hyperlipidemia and cardiac insufficiency may also suffer from cerebrovascular diseases.
Collapse
|