I am Hugo Aerts, a professor, scientist, and entrepreneur using artificial intelligence to uncover new insights into cancer, aging, and human health.

Hugo Aerts
About
Hugo Aerts
Director, AIM Program · Professor, Harvard University

Hugo Aerts PhD is a Professor at Harvard University and the Director of the Artificial Intelligence in Medicine (AIM) Program. AIM's mission is to accelerate the application of AI algorithms in medical sciences and clinical practice. This academic program centralizes AI expertise stimulating cross-pollination among clinical and technical expertise areas, and provides a common platform to address a wide range of clinical challenges.

Dr. Aerts is a leader in medical AI and Principal Investigator on major NIH-supported efforts, including the Quantitative Imaging Network (U01) and Informatics Technology for Cancer Research (U24) initiatives of the NCI. In 2020 he was awarded a prestigious ERC Consolidator grant from the European Union's Horizon program. In 2022 he was recognized by Web of Science as among the top 1% highest cited scientists worldwide.

Dr. Aerts is a Professor at Harvard University and an Adjunct Professor at Maastricht University. He earned his Master in Engineering from Eindhoven Institute of Technology, his PhD from Maastricht University, and his postdoctoral fellowship from Harvard School of Public Health.

Research
01
Radiomics & Deep Radiomics
Quantifying thousands of imaging features from CT, PET, and MRI to build non-invasive biomarkers predicting tumour biology, treatment response, and patient survival. Our 2014 Nature Communications paper has 4,600+ citations — the most-cited in the field.
Imaging · Biomarkers · NCI U01
02
Foundation Models for Cancer Imaging
Large-scale self-supervised models pre-trained on 11 million medical images to discover novel cancer imaging biomarkers across organ systems — reducing the need for manually labelled training data. Nature Communications, 2024.
Deep Learning · NCI · Nature Comm 2024
03
FaceAge — Biological Ageing AI
Deep learning system estimating biological age from facial photographs, achieving ~80% accuracy predicting cancer patient survival. Lancet Digital Health, 2025. Covered by NYT, Washington Post, CNN, BBC, Fox News.
Ageing · Lancet Digital Health 2025
04
AI in Cardiology — DeepCAC
Predicting cardiovascular risk from routine chest X-rays and CT, linking cancer therapy toxicity to heart disease via AI-automated body composition. Validated on 40,000+ patients. Nature Medicine, 2022 — CNN Morning Show.
Cardiology · Nature Medicine 2022
05
LLMs in Clinical Care
Evaluating and developing large language models for patient portal messaging, cancer treatment recommendations, and social determinants of health from clinical notes — including TRIPOD+AI guideline development.
LLMs · Lancet · NEJM AI
06
Radiogenomics & Multimodal AI
Linking imaging phenotypes to molecular and genomic profiles to understand tumour biology non-invasively. Building multimodal models that fuse CT, genomics, pathology, and clinical data into unified predictors of disease outcome.
Genomics · Precision Medicine
Publications
Nature — Thymic Health
Publication · 2026 · Nature
Thymic Health consequences in adults
AI thymus biomarker linked to mortality and disease incidence
Thymic Health and Immunotherapy
Publication · 2026 · Nature
Thymic Health and Immunotherapy Outcomes
Thymus Health linked to immunotherapy outcomes in patients with cancer
nature
Publication · 2025
Guidelines for studies using large language models (TRIPOD+AI)
LLMs
Publication · 2024
Responding to patient messages using large language models in oncology care
Foundation model
Publication · 2024
A foundation model for cancer imaging biomarkers trained on 11 million medical images
SDOH
Publication · 2024
LLMs for extraction of social determinants of health from clinical notes
ChatGPT
Publication · 2023
Reliability of large language models for cancer treatment recommendations
DeepCAC
Publication · 2022
Deep learning to predict cardiovascular risk from chest radiographs
nature
Publication · 2017
Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach
naturemedicine
Publication · 2019
Deep learning for prediction of obstructive disease from fast myocardial perfusion SPECT
Full list on Google Scholar — 428 papers, 80k+ citations ↗
Media
Featured in CNN The New York Times The Washington Post BBC FOX NEWS FORBES WSJ Nature SCIENCE NPR BLOOMBERG
The New York Times
News · 2025
New York Times features AIM Study on Biological Age using Face Photos
FOX NEWS
News · 2025
AIM Study on Biological Age using Face Photos picked up by Fox News
BBC
News · 2025
Novel AI for estimation of biological age from face photographs
NIHNational Cancer Institute
News · 2025
AIM Investigator awarded 5-year grant to improve cancer survivorship
The New York Times
News · 2024
AIM Study on LLMs for patient portal messaging featured in the New York Times
Google
Award · 2024
AIM Student Awarded 2024 Google PhD Fellowship in Natural Language Processing
BLOOMBERG
News · 2023
AIM ChatGPT reliability study headlines across Bloomberg, WBUR, Yahoo and major outlets
CNN
News · 2022
Harvard AI predicts cardiovascular risk directly from a routine chest X-ray
Nature
Press · 2024
AIM foundation model study highlighted by MGB News, ScienceMag and ecancer
Speaking
2026
Upcoming
The next frontier of AI in oncology imaging — from biomarkers to clinical deployment
MICCAI 2026 Keynote · Marrakech, Morocco
Apr 2026
FaceAge and biological age estimation as a clinical tool in cancer treatment planning
AACR Annual Meeting · Chicago, USA
2025
Foundation models for medical imaging: progress, pitfalls, and the path to clinical use
RSNA Annual Meeting · Chicago, USA
2024
Large language models for clinical oncology — reliability, bias, and real-world deployment
ASCO Annual Meeting · Chicago, USA
2023
Artificial intelligence for medical imaging — from radiomics to deep learning at scale
CANSSI Ontario STAGE · University of Toronto
2022
Radiomics: there is more than meets the eye in medical imaging
SPIE Medical Imaging Plenary · San Diego, USA
2022
Deep learning applied to X-rays as a new biomarker source: cardiovascular risk prediction
NCI CBIIT Speaker Series · Bethesda, USA
Contact
Hugo Aerts
Director, AIM Program · Professor, Harvard University
Brigham & Women's Hospital
Mass General Hospital
Harvard University
Emailhaerts@bwh.harvard.edu
Labaim.mgh.harvard.edu
HarvardAI in Medicine (AIM) Program
Mass General Brigham
75 Francis St, Boston MA 02115
MaastrichtDept. of Precision Medicine
Maastricht University
P.O. Box 616, 6200 MD, NL
ScholarGoogle Scholar — 80k+ citations
ORCID0000-0002-2122-2003
Twitter@hugoaerts
LinkedInlinkedin.com/in/hugoaerts