Pixelwise tissue segmentation for precise local in-vivo dose response assessment in patient-derived xenografts

Lucas Ewing; Sebastian W. Ahn; Oliver H. Jonas; Nobuhiko Hata

doi:10.1117/12.2513080

8 March 2019 Pixelwise tissue segmentation for precise local in-vivo dose response assessment in patient-derived xenografts

Lucas Ewing, Sebastian W. Ahn, Oliver H. Jonas, Nobuhiko Hata

Author Affiliations +

Proceedings Volume 10951, Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling; 1095133 (2019) https://doi.org/10.1117/12.2513080
Event: SPIE Medical Imaging, 2019, San Diego, California, United States

Abstract

Patient-specific dose response against chemotherapeutics can be assessed through local in situ release of drugs at sub-therapeutic concentrations. Such controlled release can be performed in patient-derived xenografts (PDXs), which offer pre-clinical methods for mimicking the tumor microenvironment. However, the prolonged co-existence of intermingled human and mouse tissues poses a number of challenges for histological image analysis. Manual annotation of regions of human tissue is labor-intensive and lacks reproducibility and scalability, complicating the investigation of multiplexed local drug effects near drug-dispensing microdevices. To this end, we apply a random forest algorithm for segmenting histological images to obtain binary masks for refined image analysis. Region-of-interest masks obtained using this supervised learning approach allow for a spatially refined assessment of the dose response in heterogeneous tissue8. We achieved a Dice similarity coefficient score (DSC) of 0.56 with the random forest classifier.

Citation Download Citation

Lucas Ewing, Sebastian W. Ahn, Oliver H. Jonas, and Nobuhiko Hata "Pixelwise tissue segmentation for precise local in-vivo dose response assessment in patient-derived xenografts", Proc. SPIE 10951, Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling, 1095133 (8 March 2019); https://doi.org/10.1117/12.2513080

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available