Learn more →
Back to Expert Scholars
research / researchPathAI, deep learning pathology, biomarker discovery, TIL assessment, companion diagnosticsBispecific T-cell Engager Pioneer

Andrew H. Beck

安德鲁·贝克

MD, PhD

🏢PathAI; Harvard Medical School (formerly)(PathAI公司;哈佛医学院(前任))🌐USA

Co-Founder and Chief Science Officer, PathAI; Formerly Associate Professor of Pathology, Harvard Medical School and Beth Israel Deaconess Medical CenterPathAI联合创始人兼首席科学官;前哈佛医学院及贝斯以色列女执事医疗中心病理学副教授

45
h-index
4
Key Papers
5
Awards
4
Key Contributions

👥Biography 个人简介

Andrew H. Beck, MD, PhD is Co-Founder and Chief Science Officer of PathAI, the leading AI-powered pathology company, and formerly Associate Professor of Pathology at Harvard Medical School and Beth Israel Deaconess Medical Center. Dr. Beck is widely regarded as one of the founders of modern computational pathology, having published seminal work on deep learning for pathology image analysis while at Harvard before founding PathAI in 2016 to translate academic advances in AI pathology into clinical-grade diagnostic and biomarker tools. Under his scientific leadership, PathAI has developed AI-powered pathology platforms deployed across major pharmaceutical companies for clinical trial biomarker assessment, drug development, and companion diagnostic development. His laboratory at Harvard generated landmark studies demonstrating that deep learning models applied to H&E-stained breast cancer slides can predict survival outcomes and identify immune infiltration patterns associated with treatment response, establishing AI-quantified tumor-infiltrating lymphocytes (TILs) as a reproducible, prognostic biomarker. Dr. Beck's PathAI platform has been used to develop AI-based biomarkers in clinical trials for liver fibrosis, NASH, breast cancer, and immunotherapy response, and several PathAI-derived assays are in regulatory pathways as companion diagnostics. He is a physician-scientist and practicing pathologist whose integration of clinical, academic, and industry perspectives has made him a defining figure in the transition of computational pathology from research to clinical reality.

Share:

🧪Research Fields 研究领域

PathAI — Founding and Scientific Direction of AI-Powered Pathology PlatformPathAI——AI驱动病理学平台的创建与科学指导
Deep Learning for Pathology Biomarker Discovery — TILs, Fibrosis, and Tumor Architecture病理生物标志物发现的深度学习——TIL、纤维化与肿瘤结构
Companion Diagnostics — AI-Powered Biomarker Assay Development for Clinical Trials伴随诊断——用于临床试验的AI驱动生物标志物检测开发
Tumor-Infiltrating Lymphocyte Assessment — AI Quantification for Immunotherapy Prediction肿瘤浸润淋巴细胞评估——用于免疫治疗预测的AI定量
Spatial Tumor Microenvironment Analysis — AI Mapping of Immune Cell Infiltration Patterns空间肿瘤微环境分析——免疫细胞浸润模式的AI映射

🎓Key Contributions 主要贡献

Systematic Analysis of Breast Cancer Morphology with AI — Foundational Computational Pathology

Published the foundational study "Systematic Analysis of Breast Cancer Morphology Uncovers Stromal Features Associated with Survival" in Science Translational Medicine (2011), one of the first demonstrations that computational analysis of digitized H&E-stained pathology slides using machine learning could extract morphological features predictive of breast cancer-specific survival. This work established deep quantitative histomorphological analysis as a viable approach for cancer prognosis independent of established clinical and pathological variables, directly inspiring a generation of computational pathology research.

PathAI — Clinical-Grade AI Pathology Platform

Co-founded PathAI in 2016 and led its scientific development into the world's leading AI-powered pathology company, providing pharmaceutical, biotech, and clinical partners with AI-based pathology analysis platforms for oncology and hepatology. Under his leadership, PathAI developed AISight and other enterprise pathology AI tools deployed across global pharma partnerships including Bristol Myers Squibb, Gilead, and Sanofi for clinical trial biomarker evaluation, with AI-derived biomarkers qualifying as clinical trial endpoints in multiple phase II/III oncology studies.

AI Assessment of Tumor-Infiltrating Lymphocytes for Immunotherapy Biomarkers

Led development of AI-based quantification of tumor-infiltrating lymphocytes (TILs) in breast cancer and other solid tumors, providing reproducible, scalable alternatives to manual pathologist TIL scoring. PathAI's AI-TIL assays have been applied in clinical trials to predict response to immunotherapy, demonstrating in multiple datasets that AI-quantified TIL density and spatial distribution patterns outperform or match manual expert scoring while enabling high-throughput biomarker assessment across large clinical cohorts.

Liver Fibrosis and NASH — AI Pathology for Non-Oncology Biomarker Development

Extended AI pathology biomarker methodology from oncology to hepatology, developing and validating PathAI's liver fibrosis and NASH activity scoring algorithms against expert pathologist consensus scoring (NASH CRN system). These algorithms achieved substantial clinical validation as digital pathology endpoints in pharmaceutical clinical trials, with PathAI's liver AI assays used by major pharma companies as primary efficacy endpoints in NASH clinical trials, demonstrating the generalizability of AI pathology biomarker methodology beyond oncology.

Representative Works 代表性著作

[1]

Systematic Analysis of Breast Cancer Morphology Uncovers Stromal Features Associated with Survival

Science Translational Medicine (2011)

Foundational computational pathology study demonstrating AI extraction of morphological prognostic features from breast cancer H&E slides, establishing stromal and immune features as survival predictors.

[2]

Assessment of Tumor-Infiltrating Lymphocytes Using Image Analysis and Multiplex Immunofluorescence in Human Epidermal Growth Factor Receptor 2–Positive and Triple-Negative Breast Cancers

Journal of Clinical Oncology (2019)

AI-based TIL quantification in HER2-positive and TNBC demonstrating that computational TIL scoring is reproducible, prognostically valid, and scalable for clinical trial biomarker assessment.

[3]

Training Deep Neural Networks on Noisy Labels with Bootstrapping

ICLR Workshop (2015)

Methodological contribution to training deep learning models on imperfectly-labeled data, directly applicable to computational pathology where annotation quality varies across sites and pathologists.

[4]

Artificial Intelligence to Identify Genetic Mutations in Cancer

Journal of Pathology (2020)

Review and validation of AI approaches for predicting genetic mutations including MSI, KRAS, and EGFR from H&E pathology images, surveying methodology and clinical translation pathways.

🏆Awards & Recognition 奖项与荣誉

🏆Harvard Medical School Young Investigator Award
🏆USCAP Stowell-Orbison Award for Outstanding Research in Pathology
🏆Forbes Healthcare AI 50 List
🏆MedTech Breakthrough Award for AI-Based Pathology Innovation
🏆AACR Outstanding Abstract Award in Computational Oncology

📄Data Sources 数据来源

Last updated: 2026-04-06 | All information from publicly available academic sources

关注 安德鲁·贝克 的研究动态

Follow Andrew H. Beck's research updates

留下邮箱,当我们发布与 Andrew H. Beck(PathAI; Harvard Medical School (formerly))相关的新研究或访谈时,我们会通知你。

我们不会泄露你的信息,也不会发送无关内容。随时可以退订。

Explore More Experts

Discover the researchers shaping the future of cancer treatment