Mars (Shih-Cheng) Huang 🚀

Mars (Shih-Cheng) Huang

Ph.D. Candidate @

Stanford University

Hi there - I am a PhD candidate in Biomedical Informatics at Stanford University, studying artificial intelligence and clinical informatics. I am currently advised by Serena Yeung, Curtis Langlotz, Nigam Shah and previously by Matthew P. Lungren. I am affiliated with Stanford’s MARVL lab and AIMI Center, and have experience working at Google Research , Microsoft Research , Salesforce AI research and The ChanZuckerberg Initiative.

My research focuses on the intersection of multimodal and self-supervised learning, and the application of these methods to improve healthcare.

All Publications

Quickly discover relevant content by filtering publications.
(2023). DrML: Diagnosing and Rectifying Vision Models using Language. ICLR.

PDF Cite Code

(2023). Label-efficient self-supervised federated learning for tackling data heterogeneity in medical imaging. IEEE Transactions on Medical Imaging.

PDF Cite Code DOI

(2023). Self-supervised learning for medical image classification: a systematic review and implementation guidelines. Nature Digital Medicine (Under Review).

PDF Cite

(2022). Adapting pre-trained vision transformers from 2D to 3D through weight inflation improves medical image segmentation. ML4H.

PDF Cite Code

(2022). Developing medical imaging AI for emerging infectious diseases. Nature Communications.

PDF Cite DOI

(2022). AI recognition of patient race in medical imaging: a modelling study. The Lancet Digital Health.

PDF Cite Code DOI

(2022). Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials. Nature Digital Medicine.

PDF Cite DOI

(2022). Development and validation of a prognostic AI biomarker using multi-modal deep learning with digital histopathology in localized prostate cancer on NRG Oncology phase III clinical trials.. Journal of Clinical Oncology.

PDF Cite DOI

(2021). Automatic lung nodule segmentation and intra-nodular heterogeneity image generation. IEEE Journal of Biomedical and Health Informatics.

PDF Cite DOI

(2021). RadFusion: Benchmarking Performance and Fairness for Multimodal Pulmonary Embolism Detection from CT and EHR. arXiv.

PDF Cite DOI

(2021). GLoRIA: A Multimodal Global-Local Representation Learning Framework for Label-Efficient Medical Image Recognition. ICCV.

PDF Cite Code

(2020). Multimodal fusion with deep neural networks for leveraging CT imaging and electronic health record: a case-study in pulmonary embolism detection. Nature Scientific Reports.

PDF Cite DOI

(2020). Biomedical Graph Visualizer for Identifying Drug Candidates. biorXiv.

PDF Cite DOI

(2020). Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines. Nature Digital Medicine.

PDF Cite DOI

(2020). OncoNet: Weakly Supervised Siamese Network to automate cancer treatment response assessment between longitudinal FDG PET/CT examinations. arXiv.

PDF Cite DOI

(2020). PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging. Nature Digital Medicine.

PDF Cite DOI

(2019). Ten simple rules for writing and sharing computational analyses in Jupyter Notebooks. PLOS Computational Biology.

PDF Cite DOI

Experience

 
 
 
 
 
Google Research
Research Scientist Intern
September 2023 – Present Mountain View

Designed object embeddings to improve dense semantic understanding in Vision Language Models (VLMs) and curated a multi-image Visual Question Answering dataset to benchmark VLM’s dense semantic understanding

 
 
 
 
 
Microsoft Research
Research Scientist Intern
June 2023 – September 2023 Seattle

Developed a VLM for generating radiology reports from Chest X-rays that achieves state-of-the-art performance

 
 
 
 
 
Salesforce
AI Research Summer Intern
June 2021 – September 2021 San Francisco

Designed and implemented a multimodal self-supervised framework for prostate cancer long-term outcome prediction.

screen reader text
The multimodal architecture is composed of two parts: a stack to parse a variable number of digital histopathology slides and another stack to merge the resultant features and predict binary long-term outcomes.

Project Link

 
 
 
 
 
Chan Zuckerberg Initiative (CZI)
Computational Biology Summer Intern
June 2019 – September 2019 Redwood City

Created Segmentify, an interactive and general-purpose cell segmentation plugin for the image viewer Napari

screen reader text
In the example above, the user is using segmentify to segment out the nucleus, cytoplasm and the background for all the cells in the image. The trained classifier is used to predict the label for all the remaining unlabeled pixels, and the segmentation output is displayed at the segmentation labels layer.

Project Link

 
 
 
 
 
San Diego Supercomputer Center
Research Programmer
March 2015 – December 2016 San Diego

Developed mmtf-pyspark, a python package that parallelizes analysis and mining of protein data using Apache-Spark

screen reader text
Visualization of Protein-DNA complex using MMTF-PySpark.

Project Link

Amateur Photography