Research Abstract
My research focuses on developing computational methods and tools that advance cancer detection, diagnosis, and treatment through biomedical imaging, machine learning, and scientific visualization. As a computer science faculty member specializing in biomedical imaging, I am committed to bridging technological innovation and clinical practice to improve outcomes in cancer care.
A primary focus of my work is melanoma detection. Motivated by personal experience with this disease, I have developed machine learning models for automated skin lesion classification using uncertainty quantification to enhance diagnostic confidence. This work was recently published at the IEEE International Symposium on Biomedical Imaging (ISBI 2025). My lab is currently translating these models into low-cost, patient-accessible skin change tracking tools to support melanoma screening and early detection.
In addition to melanoma research, I serve as Principal Investigator of the Oregon-Massachusetts Mammography Database (OMAMA-DB), the world’s largest publicly available breast cancer imaging dataset, hosted at the Harvard Dataverse (https://dataverse.harvard.edu/dataverse/omama). OMAMA-DB includes over 165,000 2D mammograms and over 67,000 3D tomosyntheses with cancer annotations. This project was funded by the Massachusetts Life Sciences Center through a Bits to Bytes grant.
Additionally, my group has contributed open-source tools for interpretable lesion detection and image retrieval using self-supervised learning, helping clinicians explore similar cases and improve diagnostic workflows.
Beyond these projects, I direct the AI Core Facility at UMass Boston, providing computational support for biomedical research and fostering interdisciplinary collaboration. I am committed to developing accessible, transparent, and reproducible computational tools that empower cancer researchers and clinicians. Through my research and leadership, I aim to contribute to DF/HCC’s mission by advancing data-driven approaches to cancer detection and diagnosis while mentoring the next generation of interdisciplinary scientists.
* Selected References
Kim, S., Gaibor, E., Matejek, B., Haehn, D. Melanoma Detection with Uncertainty Quantification. IEEE International Symposium on Biomedical Imaging (ISBI), 2025. danielhaehn.com/papers/
Zurrin, R., Goyal, N., Bendiksen, P., Manocha, M., Simovici, D., Haspel, N., Pomplun, M., Haehn, D. Outlier Detection for Mammograms. International Conference on Medical Imaging with Deep Learning (MIDL), 2023. danielhaehn.com/papers/
Qi, K., Cheng, J., Haehn, D. Lesion Search with Self-supervised Learning. International Conference on Learning Representations (ICLR), 2023. danielhaehn.com/papers/