About Me

I am a Third-year Ph.D. student in Computer Science at West Virginia University, working with Dr. Prashnna K. Gyawali at the Machine Intelligence Lab. Before starting my Ph.D., I worked as a Machine Learning Engineer for four years, where I developed and deployed machine learning models across various applications. I completed my Bachelor's degree in Computer Engineering from Pulchowk Campus (Institute of Engineering), Tribhuvan University, Nepal.

I enjoy tackling challenging problems, learning new techniques, and applying them creatively to make an impact. Outside of research, I enjoy playing chess and can be found on chess.com under the username AlinaDevkota.

Please do not hesitate to reach out to me via LinkedIn or email at devkota.alina@gmail.com.

Research Interests

My research is focused on applying Artificial Intelligence (AI) to healthcare. I am currently focusing my research on multimodal out-of-distribution (OOD) detection. My earlier work has focused on foundation models and privacy-preserving techniques, such as federated learning for medical imaging, particularly in gastrointestinal endoscopy. I am also passionate about exploring broader areas of computer vision, large language models, and other applications of deep learning and machine learning.

Looking ahead, I plan to concentrate future research at the intersection of advanced computer vision and language models for medical imaging, with the goal of enhancing diagnostic accuracy and improving patient outcomes.

News

2026
Currently serving as the Web and Publicity Chair for the 4th DEMI Workshop at MICCAI 2026.
2026
Received 2nd place in the Rapid Fire Presentation at Statler College Research Week.
2026
Our paper, FedVG: Gradient-Guided Aggregation for Enhanced Federated Learning, has been accepted to CVPR Findings 2026.
2025
Served as the lecturer for the WVU SURE Generative AI Workshop.
2025
Received 2nd place in the poster presentation at the WVU AI Symposium.
2025
Our paper, AI Analysis for Ejection Fraction Estimation from 12-Lead ECG, was published in Scientific Reports.