Skin Strata Delineation in Reflectance Confocal Microscopy Images Using Recurrent Convolutional Networks with Attention
Jun 14, 2021·
,,,,,,·
0 min read
Alican Bozkurt
Kivanc Kose
Jaume Coll-Font
Christi Alessi-Fox
Dana H. Brooks
Jennifer G. Dy
Milind Rajadhyaksha
Abstract
Reflectance confocal microscopy (RCM) is an effective, non-invasive pre-screening tool for cancer diagnosis. However, acquiring and reading RCM images requires extensive training and experience, and novice clinicians exhibit high variance in diagnostic accuracy. Consequently, there is a compelling need for quantitative tools to standardize image acquisition and analysis. In this work, we use recurrent convolutional neural networks with attention mechanisms to delineate skin strata in stacks of RCM images collected at consecutive depths. We introduce a Toeplitz attention mechanism that constrains the attention map to respect the ordered structure of image stacks. Testing our model on an expert-labeled dataset of 504 RCM stacks, we achieve 88.07% image-wise classification accuracy, which represents the state of the art for this task.
Type
Publication
In Scientific Reports 11(1), 12576 (2021)
Authors
AI Scientist
I am an AI Scientist at Paige AI. I did my Ph.D. with Jennifer Dy, Dana Brooks, and Jan-Willem van de Meent at Northeastern University. My main research interests are machine learning with emphasis on probabilistic programming, deep neural networks, and their applications in biomedical image processing. I am one of the developers of Probabilistic Torch, a library for deep generative models that extends PyTorch. I am also one of the maintainers of the PyTorch distributions module.