Kivanc Kose, Ph.D.
Assistant Lab Member (Assistant Professor), Dermatology Service, Memorial Sloan Kettering Cancer Center
Adjunct Assistant Professor, Columbia University
I’m an electrical engineer who builds machine learning and computer vision tools for noninvasive cancer imaging: reflectance confocal microscopy, dermoscopy, smartphone photography and 3D total-body photography. I work across the Optical Imaging Lab and the Dermatology Imaging Informatics Lab at MSK, and I’m co-PI of the NIH-funded International Skin Imaging Collaboration (ISIC) Archive, the leading public skin imaging repository for AI research.
Email · Google Scholar · ORCID · CV (PDF)
About
My goal is to make new imaging technologies easier to adopt in the clinic, so patients can get rapid, minimally invasive diagnosis and treatment. I care about tools that survive clinical deployment: two of my segmentation algorithms for reflectance confocal microscopy are integrated into a research RCM device at MSK for prospective evaluation.
I believe every patient should have access to high-quality care, and that AI should be part of continuing medical education so physicians can use it effectively and responsibly. I mentor medical fellows, students and colleagues at MSK in machine learning and computer vision, and I lecture and advise engineering and arts graduate students at Columbia and Northeastern.
Through ISIC I help build and analyze large, open dermatology datasets and coordinate international machine-learning challenges, so the wider research community can work on medical image analysis. I have published more than 100 collaborative papers with clinicians and engineers.
I received my B.Sc., M.Sc. and Ph.D. in Electrical and Electronics Engineering from Bilkent University, Turkey, and was a postdoctoral fellow at MSK from 2012 to 2016.
News
- Nov 2025Our paper on automated triage of cancer-suspicious skin lesions with 3D total-body photography is out in npj Digital Medicine.
- Oct 2024Invited talk on the ISIC 2024 Skin Cancer Detection with 3D-TBP challenge at the ISIC Workshop, MICCAI 2024, Morocco.
- Aug 2024The SLICE-3D dataset — 400,000 skin lesion crops from 3D total-body photography — is published in Scientific Data.
- 2024Appointed Assistant Lab Member (Assistant Professor) in the Dermatology Service at MSK.
- Dec 2023SmoothHess: ReLU network feature interactions via Stein’s lemma, with Max Torop and Jennifer Dy, presented at NeurIPS 2023.
Selected publications
- Noninvasive diagnosis of melanoma using machine learning and reflectance confocal microscopy
Journal of Investigative Dermatology 146(1), 274–276.e1, 2026 - Automated triage of cancer-suspicious skin lesions with 3D total-body photography
npj Digital Medicine 8(1), 708, 2025 - Artificial intelligence in the non-invasive detection of melanoma
Life 14(12), 1602, 2024 - The SLICE-3D dataset: 400,000 skin lesion image crops extracted from 3D TBP for skin cancer detection
Scientific Data 11(1), 884, 2024 - Analyzing the spatial randomness in the distribution of acquired melanocytic neoplasms
Journal of Investigative Dermatology 142(12), 3274–3281, 2022 - In vivo tumor immune microenvironment phenotypes correlate with inflammation and vasculature to predict immunotherapy response
Nature Communications 13(1), 5312, 2022 - Skin strata delineation in reflectance confocal microscopy images using recurrent convolutional networks with attention
Scientific Reports 11(1), 12576, 2021 - Segmentation of cellular patterns in confocal images of melanocytic lesions in vivo via a multiscale encoder-decoder network (MED-Net)
Medical Image Analysis 67, 101841, 2021 - Correlation of handheld reflectance confocal microscopy with radial video mosaicing for margin mapping of lentigo maligna and lentigo maligna melanoma
JAMA Dermatology 153(12), 1278–1284, 2017