New paper published in npj Digital Medicine

Published:

Our paper on interpretable retrieval-augmented diagnosis for breast ultrasound, a collaboration with Seoul National University Hospital, has been published in npj Digital Medicine. Lab member Subin An is co-first author.

Title: Interpretable Multimodal Retrieval Augmented Diagnosis for Breast Ultrasound with Multinational Clinical Validation and Reader Study Authors: Han Jang, Subin An, Sunghyun Jung, Hye Ji, Miran Kim, Soochahn Lee, Kyu Sung Choi, Su Min Ha Paper: npj Digital Medicine article Project page: B-RAD project page Code: GitHub repository

B-RAD mirrors the radiologist workflow of retrieving reference cases, localizing the lesion, and assigning a BI-RADS category. It was validated on 8,311 images from eleven cohorts in seven countries, and in a four-reader study it improved reader accuracy and inter-reader agreement.

Congratulations to all authors!

You May Also Enjoy

SERVAL launched as a web application

Published:

We are excited to launch SERVAL, our software platform for retinal vessel analysis in fundus images, as a web application.

New paper published in IEEE Access

Published:

Our paper on weakly-supervised hand pose estimation for reading guitar fingerings has been published in IEEE Access.

Selected as an NRF Basic Research Laboratory

Published:

Our lab has been selected as a Basic Research Laboratory (BRL) by the National Research Foundation of Korea (NRF).

1 paper accepted to Interspeech 2026

Published:

Our collaborative research with the University of Wisconsin-Madison on active speaker detection has been accepted to Interspeech 2026. This is a substantially revised version of our earlier work, UniTalk, retitled to better reflect its focus on in-the-wild generalization and robustness.