•  
  •  
 

SMU Data Science Review

Abstract

Rapid and accurate detection of COVID-19 remains the best weapon to control and prevent the spread of this pandemic, at least before a vaccine or treatment is available. In this study, we trained our custom computer vision models to predict COVID-19 from patients’ CT scans. We trained one model using Google’s AutoML Vision platform and achieved comparable accuracy with previously reported models. We also trained several custom models using transfer learning by taking advantage of several well-unknown pre-trained computer vision models, including Resnet and Inception models. The models are fine-tuned with a relatively large dataset and their high accuracy should make them more generalizable for potential clinical application.

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Share

COinS