Bangwei Guo

I am a Ph.D. student in Computer Science at Rutgers University, advised by Distinguished Professor Dimitris N. Metaxas. Previously, I received my M.S. in Applied Statistics and B.S. in Mathematics from the University of Science and Technology of China (USTC).

My research focuses on medical foundation models, with an emphasis on universal, interactive, and in-context medical image segmentation. I work extensively to develop scalable 2D/3D segmentation frameworks that support semantic, prompt-based, and human-in-the-loop interaction. My broader research interests lie in integrating large multimodal models and agent-based reasoning into vision systems, aiming to build generalizable and adaptive AI models that can transfer across tasks and data distributions.

I am currently a research intern at NEC Laboratories America, where I work on large multimodal models and multi-agent systems for structured reasoning and complex information extraction. I am actively seeking 2026 summer internship opportunities in computer vision, medical imaging, and large language / multimodal models.

news

Jan 26, 2026	🎉 Our paper K-Prism has been accepted to ICLR2026. Congratulations to all collaborators!
Aug 01, 2025	💼 I started my part-time research internship at NEC Laboratories America in Princeton, New Jersey!
May 25, 2025	I will attend IPMI2025 in Kos, Greece.
Feb 13, 2025	🎉 Our paper VerSe has been accepted to IPMI2025. Congratulations to all collaborators!

selected publications

Journal

A robust and lightweight deep attention multiple instance learning algorithm for predicting genetic alterations

Bangwei Guo, Xingyu Li, Miaomiao Yang, and 2 more authors

Computerized Medical Imaging and Graphics, 2023

Bib PDF

@article{guo2023robust,
  title = {A robust and lightweight deep attention multiple instance learning algorithm for predicting genetic alterations},
  author = {Guo, Bangwei and Li, Xingyu and Yang, Miaomiao and Zhang, Hong and Xu, Xu Steven},
  journal = {Computerized Medical Imaging and Graphics},
  volume = {105},
  pages = {102189},
  year = {2023},
  publisher = {Elsevier},
}

Journal

Predicting microsatellite instability and key biomarkers in colorectal cancer from H&E-stained images: achieving state-of-the-art predictive performance with fewer data using swin transformer

Bangwei Guo, Xingyu Li, Miaomiao Yang, and 3 more authors

The Journal of Pathology: Clinical Research, 2023

Bib PDF Code

@article{guo2023predicting,
  title = {Predicting microsatellite instability and key biomarkers in colorectal cancer from H\&E-stained images: achieving state-of-the-art predictive performance with fewer data using swin transformer},
  author = {Guo, Bangwei and Li, Xingyu and Yang, Miaomiao and Jonnagaddala, Jitendra and Zhang, Hong and Xu, Xu Steven},
  journal = {The Journal of Pathology: Clinical Research},
  volume = {9},
  number = {3},
  pages = {223--235},
  year = {2023},
  publisher = {Wiley Online Library},
}

ICLR

K-Prism: A Knowledge-Guided and Prompt Integrated Universal Medical Image Segmentation Model

Bangwei Guo, Yunhe Gao, Meng Ye, and 4 more authors

arXiv preprint arXiv:2509.25594, 2025

Bib PDF Code

@article{guo2025k,
  title = {K-Prism: A Knowledge-Guided and Prompt Integrated Universal Medical Image Segmentation Model},
  author = {Guo, Bangwei and Gao, Yunhe and Ye, Meng and Gu, Difei and Zhou, Yang and Axel, Leon and Metaxas, Dimitris},
  journal = {arXiv preprint arXiv:2509.25594},
  year = {2025},
}

IPMI

Verse: Integrating multiple queries as prompts for versatile cardiac mri segmentation

Bangwei Guo, Meng Ye, Yunhe Gao, and 3 more authors

In International Conference on Information Processing in Medical Imaging, 2025

Bib PDF Code

@inproceedings{guo2025verse,
  title = {Verse: Integrating multiple queries as prompts for versatile cardiac mri segmentation},
  author = {Guo, Bangwei and Ye, Meng and Gao, Yunhe and Xin, Bingyu and Axel, Leon and Metaxas, Dimitris},
  booktitle = {International Conference on Information Processing in Medical Imaging},
  pages = {373--387},
  year = {2025},
  organization = {Springer},
}