SSU computer science major Jorge Bautista-Martinez and Dr. Gurman Gill have presented their work at the International Conference on Computational Science and Computational Intelligence in Las Vegas, NV. Their paper—"Comparison of Pre-trained vs Domain-specific Convolutional Neural Networks for Classification of Interstitial Lung Disease"—explores how a generalized pre-trained neural network compares to a domain-specific neural network in the context of interstitial lung disease. Jorge, a computer science and physics dual-major, graduated at the end of Fall 2019. He began working with Dr. Gill on this project in Fall 2018, by investigating the efficacy of pre-trained Convolutional Neural Networks (CNNs) for classifying 5 different types of Interstitial Lung Disease (ILD). That early work, "Applications of Convolutional Neural Network Model for classifying Interstitial Lung Disease images from Computed Tomography scans," done with CS major Skyler Penna was previous selected to represent Sonoma State University at the 33rd Annual CSU Student Research Competition held at CSU Fullerton in April 2019.
- J. Bautista-Martinez, and G. Gill, Comparison of Pre-trained vs Domain-specific Convolutional Neural Networks for Classification of Interstitial Lung Disease, Proceedings of the 2019 International Conference on Computational Science and Computational Intelligence, Dec. 2019, Las Vegas.
- Proceedings via IEEE
- PDF via Gill's website