Related Experiment Video
Updated: Jun 19, 2025
10:46
MRI and PET in Mouse Models of Myocardial Infarction
Published on: December 19, 2013
11.7K
Multi-modality deep learning-based [68Ga]Ga-DOTA-FAPI-04 PET polar map generation: potential value in detecting
Xiaoya Qiao1,2,3, Hanzhong Wang1,2,3, Hongping Meng1
1Department of Nuclear Medicine, Ruijin Hospital, Shanghai Jiao Tong University, Shanghai, China.
Summary
A new deep learning method accurately generates polar maps from [68Ga]Ga-DOTA-FAPI-04 PET images, improving detection of myocardial fibrosis after infarction and its impact on cardiac function.
Area of Science:
- Nuclear Medicine
- Cardiovascular Imaging
- Artificial Intelligence in Medicine
Background:
- Generating accurate polar maps (PM) from [68Ga]Ga-DOTA-FAPI-04 PET images is challenging due to limitations in existing automatic methods.
- Myocardial fibrosis assessment after myocardial infarction is crucial for understanding cardiac function and prognosis.
Purpose of the Study:
- To develop and validate a deep-learning-based method for accurate PM generation from [68Ga]Ga-DOTA-FAPI-04 PET images.
- To evaluate the potential of these PMs in detecting reactive fibrosis post-myocardial infarction and their correlation with cardiac function.
Main Methods:
- A deep learning approach fusing multi-modality [68Ga]Ga-DOTA-FAPI-04 PET/MR images was developed to compensate for lost cardiac structural information.
- 133 patient datasets were used for training and evaluation, with 26 undergoing longitudinal analysis.
Main Results:
- The proposed deep learning method demonstrated accuracy comparable to manual generation and superior to commercial software (PMOD) for [68Ga]Ga-DOTA-FAPI-04 PET PMs.
- Significant correlations were found between PM-derived fibrosis indices and changes in cardiac function parameters (LVESV%, LVEDV%, LVEF%), with p < 0.001.
Conclusions:
- The deep learning-generated [68Ga]Ga-DOTA-FAPI-04 PET PMs are clinically viable and accurate.
- These PMs show promise in detecting reactive myocardial fibrosis and assessing its relationship with cardiac function, potentially enhancing clinical diagnostics for myocardial infarction.
