DeSPPNet: A Deep Learning Model for Cardiac MRI Segmentation
Author Information
Author(s): Elizar Elizar, Muharar Rusdha, Zulkifley Mohd Asyraf
Primary Institution: Universiti Kebangsaan Malaysia
Hypothesis
The study aims to develop a multiscale deep learning model to segment cardiac organs based on MRI imaging data.
Conclusion
The proposed DeSPPNet model achieves high segmentation performance for cardiac MRI images by effectively utilizing multiscale features.
Supporting Evidence
- The DeSPPNet model achieved dice, IoU, and accuracy scores of 0.859, 0.800, and 0.993, respectively.
- The study utilized the ACDC cardiac MRI dataset, which includes diverse cardiac conditions.
- Multiscale deep learning techniques significantly improved segmentation accuracy.
Takeaway
This study created a smart computer program that helps doctors see and understand heart images better, making it easier to find problems.
Methodology
The study utilized a multiscale deep learning network called DeSPPNet, which incorporates a Spatial Pyramid Pooling layer for improved cardiac segmentation.
Limitations
The model's complexity may require high-performance hardware and careful tuning of hyperparameters.
Participant Demographics
The dataset included cardiac MRI scans from 100 patients across various cardiac conditions.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website