AI for healthcare

Rina Bao

Instructor, Harvard Medical School | Scientist, Boston Children's Hospital

My research focuses on multimodal AI systems for healthcare that connect medical images, clinical variables, and domain expertise to advance trustworthy and clinically meaningful diagnosis, prognosis, and decision-support systems.

rina.bao@childrens.harvard.edu Landmark Center, 401 Park Drive, Boston, MA Google Scholar

Medical AI Multimodal Learning Neonatal Brain Injury Clinical Reasoning Foundation Models

Research

Recent Research Highlights

Multimodality Learning

Building multimodal models for healthcare.

Medical Image Analysis 2025: PARADISE

Agentic Clinical Reasoning

Developing agentic AI systems for expert-level medical reasoning.

Clinical Graph-of-Thought medical reasoning teaser

ICML 2025 spotlight: Clinical CGoT

Benchmarking and Evaluation

Creating benchmarks and human-expert evaluation frameworks for healthcare AI.

BONBID-HIE II challenge benchmark teaser

IEEE TMI, Scientific Data: BONBID-HIE

Grants & Fellowships

Charles A. King Trust Research Fellowship

PI: Rina Bao | $206,200 | 2025-2027

Development of neuroimaging biomarkers for 2-year neurocognitive outcomes after neonatal brain injury

Thrasher Research Fund Early Career Award

PI: Rina Bao | $26,750 | 2024-2026

Development of neuroimaging biomarkers for 2-year neurocognitive outcomes in mild neonatal brain injury

German Academic Exchange Service AINet Fellowship

DAAD AINet Fellowship | 2025

Toward Clinical AI for Neonatal Brain Injury: Datasets, Algorithms, and Clinical Reasoning

Appointments

Instructor, Harvard Medical School, USA; Scientist, Boston Children's Hospital, USA 2025-Present
Postdoctoral Fellow, Harvard Medical School & Boston Children's Hospital, USA Working with Prof. P. Ellen Grant and Prof. Yangming Ou.

Organized Challenges and Workshops

I built and organized public MRI benchmarks and challenges for neonatal hypoxic-ischemic encephalopathy (HIE), used by researchers worldwide, with a focus on lesion segmentation, outcome prediction, and clinically grounded evaluation.

300+ researchers 20+ countries 5,000+ dataset downloads

1st BONBID-HIE Lesion Segmentation Challenge

MICCAI 2023

Challenge Website

2nd BONBID-HIE Lesion Segmentation and Outcome Prediction Challenge

MICCAI 2024

Challenge Website

1st Pediatric Brain Data Analysis Workshop

ISBI 2026

Workshop Website

Selected Publications

BONBID-HIE 2023: Lesion Segmentation Challenge in BOston Neonatal Brain Injury Data for Hypoxic Ischemic Encephalopathy IEEE TMI

Bao, R., Foster, A. N., Song, Y., Vyas, R., Kesri, A., Toubal, I. E., Soltani Kazemi, E., Rahmon, G., Kucukpinar, T., Almansour, M., Ho, M.-L., Palaniappan, K., Ninalga, D., Koirala, C. P., Mohapatra, S., Schlaug, G., Wodzinski, M., Muller, H., Ellis, D. G., Aizenberg, M. R., Aydın, M. A., Abdinli, E., Unal, G., Tahmasebi, N., Punithakumar, K., Song, T., Peng, Y., Bates, S. V., Hirschtick, R., Grant, P. E., and Ou, Y.

IEEE Transactions on Medical Imaging, 45(4), 1711-1725, 2026

Paper Project Page
PointGrad: A Unified Spatially Varying Tensor Gradient Operator for Point Cloud Segmentation IEEE TGRS

Bao, R., Palaniappan, K., and Zhao, Y.

IEEE Transactions on Geoscience and Remote Sensing, 2026

Paper
PARADISE: Personalized and regional adaptation for HIE disease identification and segmentation MIA

Bao, R., Weiss, R. J., Bates, S. V., Song, Y., He, S., Li, J., Bjørnerud, A., Hirschtick, R. L., Grant, P. E., and Ou, Y.

Medical Image Analysis, 102, 103419, 2025

Paper
BOston Neonatal Brain Injury Data for Hypoxic Ischemic Encephalopathy (BONBID-HIE): I. MRI and Lesion Labeling Scientific Data

Bao, R., Song, Y., Bates, S. V., Weiss, R. J., Foster, A. N., Jaimes, C., Sotardi, S., Zhang, Y., Hirschtick, R. L., Grant, P. E., and Ou, Y.

Scientific Data 12, 53, 2025

Paper Project Page
Visual and Domain Knowledge for Professional-level Graph-of-Thought Medical Reasoning ICML

Bao, R., Dong, S., Chen, Z., He, S., Grant, E., and Ou, Y.

International Conference on Machine Learning, 2025. Spotlight (2.6% acceptance rate)

Paper Project Page
Spatial Prior-Guided Boundary and Region-Aware 2D Lesion Segmentation in Neonatal Hypoxic Ischemic Encephalopathy MICCAI

Rao, A., Shukla, A., Jia, B., Ou, Y., and Bao, R.^†. Corresponding author.

MICCAI, 2025

Paper Project Page
Obscuration to Clarity: Bone Suppression for Enhanced Localization in Pneumothorax Segmentation of Chest Radiographs ISBI

Rao, A., Shukla, A., Jia, B., Siddharth, S., and Bao, R.. Corresponding author.

ISBI, 2026. Oral

Paper
Lightweight SDE-Net Fusing Model-Based and Learned Features for Computational Histopathology IEEE JSTSP

Bao, R., Zhao, Y., Srivastava, A., Frazier, S., and Palaniappan, K.

IEEE Journal of Selected Topics in Signal Processing, 18(6), 1123-1137, 2024

Paper
AGE2HIE: Transfer Learning from Brain Age to Predicting Neurocognitive Outcome for Infant Brain Injury ISBI

Bao, R., He, S., Grant, E., and Ou, Y.

IEEE International Symposium on Biomedical Imaging, 2025

Paper
Data Harmonization Framework for Neonatal Hypoxic-Ischemic Encephalopathy Studies JAMIA Open

Hsiao, C.-H., Foster, A. N., McDonald, S. A., Vyas, R., Ashraf, A., Bao, R., Tran, L., Kesri, A., Darzidehkalani, E., Soldatelli, M. D., Auman, J. O., Soul, J. S., Chalak, L. F., Cotten, C. M., Shankaran, S., Laptook, A. R., Grant, P. E., and Ou, Y.

JAMIA Open, 8(5), ooaf086, 2025

Paper Project Page
The Cell Tracking Challenge: 10 years of objective benchmarking Nature Methods

Maška, M., Ulman, V., Delgado-Rodriguez, P., Gómez-de-Mariscal, E., Nečasová, T., Guerrero Peña, F. A., Ing Ren, T., Meyerowitz, E. M., Scherr, T., Löffler, K., Mikut, R., Guo, T., Wang, Y., Allebach, J. P., Bao, R., Al-Shakarji, N. M., Rahmon, G., Toubal, I. E., Palaniappan, K., Lux, F., Matula, P., Sugawara, K., Magnusson, K. E. G., Aho, L., Cohen, A. R., Arbelle, A., Ben-Haim, T., Riklin Raviv, T., Isensee, F., Jäger, P. F., Maier-Hein, K. H., Zhu, Y., Ederra, C., Urbiola, A., Meijering, E., Cunha, A., Muñoz-Barrutia, A., Kozubek, M., and Ortiz-de-Solórzano, C.

Nature Methods, 20, 1010-1020, 2023

Paper Project Page
GLSNet++: Global and Local-Stream Feature Fusion for LiDAR Point Cloud Semantic Segmentation Using GNN Demixing Block IEEE Sensors Journal

Bao, R., Palaniappan, K., Zhao, Y., and Seetharaman, G.

IEEE Sensors Journal, 24(7), 11610-11624, 2024

Paper
Predicting carbon nanotube forest attributes and mechanical properties using simulated images and deep learning npj Computational Materials

Hajilounezhad, T.*, Bao, R.*, Palaniappan, K., Bunyak, F., Calyam, P., and Maschmann, M. R.

npj Computational Materials 7, 134, 2021. *Equal contribution

Paper
Relation U-Net ISBI

He, S., Bao, R., Grant, P. E., and Ou, Y.

IEEE International Symposium on Biomedical Imaging, 2025

Paper
Using Off-Target-Disease Data to Improve Target-Disease Lesion Segmentation ISBI

Li, J., He, S., Bao, R., Bjørnerud, A., Grant, E., and Ou, Y.

IEEE International Symposium on Biomedical Imaging, 2025

Paper
DMNet: Dual-stream marker guided deep network for dense cell segmentation and lineage tracking ICCV Workshop

Bao, R., Al-Shakarji, N. M., Bunyak, F., and Palaniappan, K.

ICCV Workshop, 2021. Best Paper

Paper Project Page

Conference Abstracts

Bao, R., Hsiao, C.-H., Bates, S. V., Weiss, R., Gollub, R., Grant, E., Ou, Y. Anatomy-normalized ZADC maps improve detection of neonatal brain injury in HIE. Pediatric Academic Societies Meeting, 2026. Oral
Bao, R., Grant, E., Ou, Y. Agentic artificial intelligence for hypoxic-ischemic encephalopathy prognosis through clinically guided reasoning on neonatal MRI. Pediatric Academic Societies Meeting, 2026. Oral
Ou, Y., Kesri, A., Bao, R., et al. Combining 47 clinical and neuroimaging variables to predict adverse 18-22-month outcomes for hypoxic-ischemic encephalopathy in Neonatal Research Network trials. Pediatric Academic Societies Meeting, 2025
Bao, R., Kesri, A., et al. 2-year neurocognitive outcome prediction of hypoxic-ischemic encephalopathy in neonates with brain MRIs. Society for Pediatric Radiology, 2023. Abstract
Bao, R., et al. Maternal dietary nutrition interacts with demographic and socioeconomic information to impact offspring neurocognition by early school age. Nutrition, 2023
Kesri, A., Bao, R., et al. The role of expert MRI scores in predicting adverse 18-22-month outcomes for hypoxic-ischemic encephalopathy in Neonatal Research Network trials. Annual Meeting of American Society of Pediatric Neuroradiology, 2025. Abstract
Bao, R., Grant, E., Ou, Y. Clinical data for predicting neonatal and 2-year outcomes following hypoxic-ischemic encephalopathy. Pediatric Academic Societies Meeting, 2023

Professional Activities

Organizer of the Pediatric Brain Analysis Workshop at ISBI 2026.
Organizer of the 1st and 2nd BONBID-HIE challenges and workshops at MICCAI 2023 and MICCAI 2024.
Editor for AI for Brain Lesion Detection and Trauma Video Action Recognition, LNCS proceedings for MICCAI 2023 challenge workshops.
Book chapter: Bao, R., et al. "Linking Deep Learning and MRI to Brain Disorders" in Handbook of the Biology and Pathology of Mental Disorders.

Education

Ph.D. in Computer Science, University of Missouri-Columbia, USA Advisor: Prof. Kannappan Palaniappan.
B.E. in Communication Engineering, Tianjin University, China

Teaching & Internship

Microsoft Research Asia, Intelligent Multimedia Group, 2018.

Teaching Assistant, Department of Electrical Engineering and Computer Science, University of Missouri-Columbia

CMP_SC 3050: Advanced Algorithm Design, Fall 2016
CMP_SC 4850/7850: Computer Networks I, Spring 2016

Research

Recent Research Highlights

Multimodality Learning

Agentic Clinical Reasoning

Benchmarking and Evaluation

Grants & Fellowships

Charles A. King Trust Research Fellowship

Thrasher Research Fund Early Career Award

German Academic Exchange Service AINet Fellowship

Appointments

Organized Challenges and Workshops

1st BONBID-HIE Lesion Segmentation Challenge

2nd BONBID-HIE Lesion Segmentation and Outcome Prediction Challenge

1st Pediatric Brain Data Analysis Workshop

Selected Publications

BONBID-HIE 2023: Lesion Segmentation Challenge in BOston Neonatal Brain Injury Data for Hypoxic Ischemic Encephalopathy IEEE TMI

PointGrad: A Unified Spatially Varying Tensor Gradient Operator for Point Cloud Segmentation IEEE TGRS

PARADISE: Personalized and regional adaptation for HIE disease identification and segmentation MIA

BOston Neonatal Brain Injury Data for Hypoxic Ischemic Encephalopathy (BONBID-HIE): I. MRI and Lesion Labeling Scientific Data

Visual and Domain Knowledge for Professional-level Graph-of-Thought Medical Reasoning ICML

Spatial Prior-Guided Boundary and Region-Aware 2D Lesion Segmentation in Neonatal Hypoxic Ischemic Encephalopathy MICCAI

Obscuration to Clarity: Bone Suppression for Enhanced Localization in Pneumothorax Segmentation of Chest Radiographs ISBI

Lightweight SDE-Net Fusing Model-Based and Learned Features for Computational Histopathology IEEE JSTSP

AGE2HIE: Transfer Learning from Brain Age to Predicting Neurocognitive Outcome for Infant Brain Injury ISBI

Data Harmonization Framework for Neonatal Hypoxic-Ischemic Encephalopathy Studies JAMIA Open

The Cell Tracking Challenge: 10 years of objective benchmarking Nature Methods

GLSNet++: Global and Local-Stream Feature Fusion for LiDAR Point Cloud Semantic Segmentation Using GNN Demixing Block IEEE Sensors Journal

Predicting carbon nanotube forest attributes and mechanical properties using simulated images and deep learning npj Computational Materials

Relation U-Net ISBI

Using Off-Target-Disease Data to Improve Target-Disease Lesion Segmentation ISBI

DMNet: Dual-stream marker guided deep network for dense cell segmentation and lineage tracking ICCV Workshop

Conference Abstracts

Professional Activities

Education

Teaching & Internship