Presenter(s)
Dhaval Dilip Kadia
Files
Download Project (1.8 MB)
Description
Artificial Intelligence (AI) is growing exponentially with novel computational architectures and their cognitive capabilities. AI helps solve complex problems in medical imaging. Lung segmentation is essential since it processes the volumetric information of the lungs, removes the unnecessary areas of the scan, and segments the actual area of the lungs in CT scan. This research focuses on deep learning applications to segment lungs and further develop a novel algorithm to make them robust. Supervised learning requires data to train a deep neural network. The deep learning model, such as U-Net, outperforms other network architectures for biomedical image segmentation. We propose a deep neural network based on U-Net for the lung and lung lesion segmentation tasks. The proposed model integrates convolution into the sophisticated Multiscale Recurrent Residual Neural Network based on U-Net. Both deep neural network (DNN) and availability of diverse annotated data make the given deep learning based solution robust and generalized for practical use. Even if having sophisticated DNN, scarcity of annotated data challenges the expected outcomes. Robust segmentation of COVID-19 infected lungs requires rich labeled data. Accurate pixel-level annotation tasks to generate such data are time-consuming, and that delays data preparation. We propose a novel algorithm to generate lesion-like artificial patterns, and U-Net based deep neural network for robust lung segmentation further helps segment COVID-19 lung infection. The pattern generation algorithm generates 2D and 3D patterns to create an enormous amount of synthetic data. This algorithm and DNN give accurate lung segmentation results for highly infected lungs and provides infection segmentation. This research applies to the preprocessing stages of different applications of deep learning, medical imaging, and data annotation. The proposed work helps the deep neural network to generalize on a given domain to accomplish robust segmentation results in the absence of exact data.
Publication Date
4-22-2021
Project Designation
Graduate Research
Primary Advisor
Van Tam Nguyen, K. Asari Vijayan
Primary Advisor's Department
Computer Science
Keywords
Stander Symposium project, College of Arts and Sciences
United Nations Sustainable Development Goals
Good Health and Well-Being
Recommended Citation
"Lesion Synthesis Algorithm and Multi-Scale U-Net for Lung and Lesion Segmentation" (2021). Stander Symposium Projects. 2155.
https://ecommons.udayton.edu/stander_posters/2155
![Lesion Synthesis Algorithm and Multi-Scale U-Net for Lung and Lesion Segmentation](https://ecommons.udayton.edu/stander_posters/3156/thumbnail.jpg)