Yulder Daniel Angarita Marmolejo, PhD

As an Assistant Professor in the Computational Pathology & AI Center of Excellence (CPACE) within the Department of Pathology at the University of Pittsburgh School of Medicine (UPSOM), Dr. Angarita focuses his research on developing multimodal artificial intelligence approaches that integrate computer vision, biomedical imaging, omics, and physiological data. His work bridges computational analysis and pathology practice through interpretable deep learning frameworks designed to support cellular and tissue-level prediction, diagnosis, and translational applications.

Dr. Angarita develops multitask deep learning models that combine CNN–Transformer architectures with large language models (LLMs) to jointly analyze heterogeneous biomedical inputs. These include brightfield and single-cell microscopy images, histopathological slide images, bulk RNA sequencing profiles, and case-level physiological information. The goal of this integrative approach is to predict cellular and tissue states, including normal versus diseased conditions and functionally differentiated phenotypes, while advancing less invasive, non-destructive strategies for assessment and diagnosis.

His doctoral research in Bioengineering at the University of Pittsburgh established foundational methods for label-free assessment of stem-cell–derived cellular systems using deep learning. Building on that training, his current research program at CPACE extends these approaches toward computational pathology platforms that can support tissue and cell diagnosis and prognosis, quality control in organoid manufacturing, AI-enabled drug discovery, and decision-support tools for physicians practicing in remote or resource-limited settings.

Dr. Angarita is committed to interdisciplinary collaboration and trainee mentorship. He has supervised undergraduate and graduate researchers across engineering, computer science, and computational biology, and has served as a teaching assistant for undergraduate statistics in bioengineering. He brings complementary experience in software development and biomedical laboratory research, including prior work in skin cancer and systemic sclerosis research at the University of Pittsburgh.

Through these efforts, Dr. Angarita aims to help transform how imaging, molecular, and physiological data are integrated in pathology, ultimately enhancing patient care through more precise, scalable, and accessible computational diagnostics.

Focus Areas

• Multimodal Deep Learning & Vision–Language Models. Dr. Angarita develops multitask deep learning architectures that combine convolutional, transformer, and language-model components for biomedical prediction tasks spanning classification and regression.
• Biomedical & Pathology Image Analysis. His research integrates brightfield microscopy, single-cell imaging, and histopathological slide images with computational pipelines for automated phenotyping, state prediction, and interpretable image-based analysis.
• Multimodal Integration of Imaging, Omics, and Physiology. He builds AI systems that jointly analyze imaging, bulk RNA sequencing, and physiological case-level data to connect visual phenotypes with molecular and clinical context.
• Translational AI Platforms for Diagnosis, Manufacturing, and Drug Discovery. A central goal of his program is to establish AI platforms that support tissue and cell diagnosis and prognosis, organoid manufacturing quality control, drug discovery, and decision support for physicians in remote clinical settings.

Research Significance to Patient Care

Dr. Angarita’s research is directed toward integrating artificial intelligence into pathology and related biomedical workflows to improve how cellular and tissue states are assessed. By combining imaging with omics and physiological information, his frameworks aim to improve diagnostic and prognostic precision, reduce dependence on more invasive or destructive assays where feasible, and accelerate biologically informed decision-making. Applications in organoid quality control and drug discovery further support translational pipelines that connect experimental models with clinical needs. Decision-support tools designed for remote settings are intended to expand access to computational pathology expertise beyond major medical centers, helping physicians make more timely and informed assessments of normal and diseased tissues and cells.

Appointments

• Assistant Professor, Department of Pathology, University of Pittsburgh School of Medicine
• Faculty Member, Computational Pathology & AI Center of Excellence (CPACE), Department of Pathology, University of Pittsburgh School of Medicine

Education

• Ph.D., Bioengineering (2021–2026). University of Pittsburgh, Pittsburgh, PA. Thesis: Deep Learning for Label-Free Assessment of iPSC-Derived Pancreatic β Cells: From CNN Classification to Multimodal Interpretability.
• M.Sc., Biomedical Engineering (2018–2020). Duquesne University, Pittsburgh, PA. Thesis: Bacteria Analysis by Using a Supervised Machine Learning Algorithm Based on Droplet Microfluidics.
• B.Sc., Computer Science (2013–2016). Duquesne University, Pittsburgh, PA.
• M.Sc. coursework, Biomedical Sciences (2004–2008). Universidad del Valle, Cali, Colombia. Coursework completed.
• B.Sc., Chemistry (1995–2001). Universidad del Valle, Cali, Colombia.

Professional Highlights

• Patent co-inventor, Rapid Enumeration of Microorganisms (U.S. Application No. 63184447 / WO2022235284A1), May 2021
• Cellular Approaches to Tissue Engineering and Regeneration (CATER) T32 Training Program, National Institute of Biomedical Imaging and Bioengineering (NIBIB), NIH, 2023
• URM Ethnic Minority and Disadvantaged Travel Award, AIMBE Public Policy Institute, American Institute for Medical and Biological Engineering, 2023
• PITT STRIVE Scholarship, National Science Foundation AGEP-KAT Program, 2021

Selected Prior Experience

• Software Developer, CGI Technologies and Solutions Inc. (2021). Developed software solutions implementing algorithms and system functionalities for robust application development.
• Research Technician III, University of Pittsburgh (2011–2012). Conducted experimental research in skin cancer and systemic sclerosis using Western blot, DNA electrophoresis, RNA extraction, and immunohistochemistry.
• Adjunct Professor, Escuela Nacional del Deporte, Cali, Colombia (2006–2008). Taught undergraduate biochemistry and designed laboratory experiments for physical therapy and sport science students.

Mentoring and Teaching

Dr. Angarita has mentored undergraduate and graduate trainees in engineering, computer science, and computational biology, including students who have advanced to graduate programs in biomedical engineering and the natural sciences. He has also served as a teaching assistant for Statistics for Bioengineering at the University of Pittsburgh, where he led laboratory sessions, graded coursework, and supported student learning through office hours.