If you can’t build AI that generalizes to new sites, you can’t build AI that makes an impact. Beyond potentially improving melanoma detection, our work highlights the importance of key aspects of training AI that often go overlooked. - Julianna Ianni, Ph.D., Proscia’s VP of AI Research & Development
PHILADELPHIA (PRWEB) November 15, 2022
Proscia®, a leading provider of digital and computational pathology solutions, today announced new research on improving the generalization of an artificial intelligence (AI) classification model with contrastive self-supervised learning. The results, which include a 56.3% increase in melanoma detection sensitivity, will be presented at the Conference on Neural Information Processing Systems (NeurIPS) 2022.
Proscia’s retrospective study “Learning SimCLR Representations for Improving Melanoma Whole Slide Images Classification Model Generalization” was conducted with data from three sites. The study investigated the impact of extended training time and different augmentations on an AI model that leverages the SimCLR framework of contrastive self-supervised learning to classify melanoma cases.
The results show that optimizing these factors can improve the generalization of an AI model trained with data from two sites, demonstrating a 56.3% increase in melanoma detection sensitivity when evaluated on images from the third site. As melanoma is the deadliest form of skin cancer and often challenging to diagnose, the findings illustrate the promise of contrastive self-supervised learning to help lower the misdiagnosis rate.
“Very often, if you can’t build AI that generalizes to new sites, you can’t build AI that makes an impact,” said Julianna Ianni, Ph.D., Proscia’s Vice President of AI Research and Development. “Beyond potentially improving melanoma detection, our work highlights the importance of key aspects of training AI that often go overlooked.”
AI development in pathology has been limited by the difficulty of obtaining annotated images to train models. Self-supervised learning has emerged as an approach for overcoming this challenge by leveraging unannotated data. While it has resulted in performance improvements for some models, they often fail to generalize to new sites due to the variability of data in routine practice. Contrastive learning, when applied to self-supervised learning, has yielded promising results for model generalization as indicated by Proscia’s study. Such results show its potential to increase the volume of images available for training highly-generalizable models.
Proscia’s study marks its second publication in the past two months demonstrating its expertise in AI for skin pathology. The company recently presented findings on AI that predicts diagnostic agreement for melanoma and other diseases with low pathologist concordance. Proscia also previously announced research results on AI that detects melanoma with a high degree of accuracy.
Details on Proscia’s poster presentation are as follows:
Title: “Learning SimCLR Representations for Improving Melanoma Whole Slide Images Classification Model Generalization”
Date: December 2, 2022 from 10:30-11:10 AM and 3:20-4:20 PM CT
Location: New Orleans Convention Center
About Proscia
Proscia is a software company that is accelerating pathology’s digital transformation to change the way we understand diseases like cancer. Its Concentriq digital pathology platform and powerful AI applications are advancing the 150-year-old standard of research and diagnosis towards a data-driven discipline, unlocking new insights that accelerate discovery, improve patient outcomes, and fulfill the promise of precision care. Leading diagnostic laboratories and over 10 of the top 20 pharmaceutical companies rely on Proscia’s software each day. For more information, visit proscia.com.