Medical Imaging AI researcher applying opportunistic screening and deep learning to uncover hidden cardiovascular clues in routine radiology scans.
About Me
I am Theo Dapamede, a Postdoctoral Fellow at Emory University. My research focuses on applying AI in medical imaging — particularly using opportunistic screening to uncover hidden disease clues in routine scans.
I develop AI models for cardiovascular risk prediction from mammograms, lung function decline prediction using social determinants of health, and AI-enhanced prognostication of kidney failure and lung cancer risk from chest radiographs.
I also explore LLMs for automated post-deployment monitoring of commercial AI, granular pathology detection in radiology reports, and synthetic data to improve generalizability in medical imaging AI.
Research Interests
Background
An international academic journey across four countries — Indonesia, Australia, New Zealand, and the USA.
Recognition
Society of Imaging Informatics in Medicine
Best
Research Paper Award
Helen & Paul Chang Foundation
New
Investigator Travel Award
University of Otago Doctoral Scholarship
Postgraduate High Honour Roll
University of
Sydney
Australia Awards Scholarship
Professional
Member · 2015–present
Indonesian Medical Association
Ikatan Dokter Indonesia
Member In-Training
Society for Imaging Informatics in Medicine
SIIM
Member In-Training
Radiological Society of North America
RSNA
Associate Member
Sigma Xi, The Scientific Research Honor Society
Fellow In-Training
North American Society for Cardiovascular Imaging
NASCI
Fellow In-Training
American College of Cardiology
ACC
Open Source
A Python library for simplified, reproducible LLM prompting — tailored for biomedical and radiology applications. Co-authored with Dr. Bardia Khosravi.
Presentations
A hands-on walkthrough covering key DICOM preprocessing steps for medical imaging AI pipelines — from raw pixel data to model-ready inputs.
Press
American College of Cardiology
Physician's Weekly
NBC Right Now
Cardiovascular Business
Emory Health Digest
Scholarship
Sorted by year, most recent first.