VS-Edge Project

In Collaboration with University of Chicago

VS-EDGE (Visual-Semantic Explanations for Diagnostic Guidance) aims to bridge the gap between research Computer-Aided Diagnosis (CAD) systems and clinical use by developing explainable CAD systems for lung cancer and connecting image features extracted computationally with semantic understanding.

This research is supported by the National Institute of Health (NIH) under grant number 1R15CA297521-01.

For more information on the project, contact the Principal Investigator (PI) Dr. Daniela Stan Raicu.

People

Faculty

Students

Rashi Jain, MS Artificial Intelligence
Charmi Patel, PhD Computer Vision
Rohan Patil, MS Computer Science
Miranda Price, MS Data Science

Publications

Lucas M., Raicu D., Furst JD., Lerma M., "Human‑centered contrastive explanations for medical imaging using VAE‑AC‑WGAN". SPIE Medical Imaging Symposium: Computer-Aided Diagnosis , Vancouver, British Columbia, February 15–19, 2026
Patel C., Wang Y., Patil R., Tchoua R., Furst JD., Raicu DS., "Evaluating ImageNet and Domain-Specific Pretraining for Variational Autoencoder Reconstruction of Lung Nodules". SPIE Medical Imaging Symposium: Computer-Aided Diagnosis , Vancouver, British Columbia, February 15–19, 2026
Wang Y., Patel C., Tchoua R., Furst JD., Raicu DS., "Harnessing Generative AI for Lung Nodule Spiculation Characterization". Journal of Imaging Informatics in Medicine , 2025
Patel C., Wang Y., Tchoua R., Orhean A., Furst JD., Raicu DS., "Enhancing lung nodule classification with variational autoencoder-based image augmentation". SPIE Medical Imaging Symposium: Computer-Aided Diagnosis, San Diego, California, February 16-22, 2025
Wang Y., Patel C., Ramaraj T., Tchoua R., Furst JD., Raicu DS., “Enhancing sensitivity in lung nodule malignancy classification: incorporating cost values into deep learning-based CAD systems”. SPIE Medical Imaging Symposium, San Diego, California, February 18-22, 2024