Identifying Pediatric Vascular Anomalies With Deep Learning
This work addresses a critical problem for pediatricians and primary care physicians in diagnosing vascular anomalies, potentially reducing delayed or incorrect referrals, with incremental improvements in accuracy using deep learning.
The researchers tackled the challenge of accurately diagnosing pediatric vascular anomalies, which are often misdiagnosed due to subtle visual differences, by developing a convolutional neural network (CNN) that achieved an average AUC of 0.9731 on a 12-class taxonomy and increased pediatricians' diagnostic accuracy from 73.10% to 91.67% when used as an aid.
Vascular anomalies, more colloquially known as birthmarks, affect up to 1 in 10 infants. Though many of these lesions self-resolve, some types can result in medical complications or disfigurement without proper diagnosis or management. Accurately diagnosing vascular anomalies is challenging for pediatricians and primary care physicians due to subtle visual differences and similarity to other pediatric dermatologic conditions. This can result in delayed or incorrect referrals for treatment. To address this problem, we developed a convolutional neural network (CNN) to automatically classify images of vascular anomalies and other pediatric skin conditions to aid physicians with diagnosis. We constructed a dataset of 21,681 clinical images, including data collected between 2002-2018 at Seattle Children's hospital as well as five dermatologist-curated online repositories, and built a taxonomy over vascular anomalies and other common pediatric skin lesions. The CNN achieved an average AUC of 0.9731 when ten-fold cross-validation was performed across a taxonomy of 12 classes. The classifier's average AUC and weighted F1 score was 0.9889 and 0.9732 respectively when evaluated on a previously unseen test set of six of these classes. Further, when used as an aid by pediatricians (n = 7), the classifier increased their average visual diagnostic accuracy from 73.10% to 91.67%. The classifier runs in real-time on a smartphone and has the potential to improve diagnosis of these conditions, particularly in resource-limited areas.