A leading provider of digital and computational pathology solutions, Proscia has released new study results on a technology that could have big implications in healthcare. The new technology leverages artificial intelligence (AI) to automatically detect melanoma with high accuracy. Melanoma, which is the deadliest form of skin cancer, stands to benefit greatly from AI technologies, which can also result in faster diagnoses, improved patient outcomes, and optimized lab economics.
The new study was conducted at Thomas Jefferson University and the University of Florida, and it involved real-world AI performance on an uncurated set of 1,422 sequential skin biopsies. According to the results, the AI was able to correctly identify invasive melanoma and melanoma in situ with a sensitivity of 93% and specificity of 91%.
The study validated a multi-site retrospective study of 12,784 images, which is set to be presented at the 2021 International Conference on Computer Vision.
Dr. Kiran Motaparthi is Director of Dermatopathology and Clinical Associate Professor of Dermatology at the University of Florida.
“The performance of Proscia’s technology in detecting melanoma and other malignant skin diseases is impressive,” said Dr. Motaparthi. “This is an exciting development as pathologists increasingly look to unlock new sources of value from artificial intelligence.”
Additional Research on AI in Dermatopathology
Proscia is also conducting further research into the possible benefits of AI in dermatology, including how it can deliver faster results to patients. AI can be designed to automatically identify melanoma and alert pathologists to high-risk cases, which could lead to earlier diagnosis. Technology like this could help ensure certain patients are prioritized when necessary, helping begin treatment quicker.
AI could also create more consistent diagnosis of difficult melanoma cases, given the disease is one of the most challenging to diagnose. Because of this difficulty, it is important for AI to be able to distinguish melanoma from benign mimickers. AI could aid the pathologists while increasing diagnostic accuracy and improving patient outcomes.
The third additional area of research is how AI can optimize laboratory productivity to enhance profitability. AI can classify and distinguish melanoma and non-melanoma skin cancer, which is crucial given that there are more than 15 million biopsies taken each year in the United States. This would enable labs to optimize case distribution among specialists and non-specialists, which would help improve efficiency gains and lead to more cases being able to be processed.
Julianna Ianni, Ph.D., is Vice President of AI Research & Development at Proscia.
“Proscia’s technology represents a significant advancement in our work on skin pathology,” said Ianni. “Our AI not only identifies melanoma, a difficult diagnosis, but also accounts for the high degree of variation in disease to push the boundaries of deep learning in medicine. In doing so, it holds great promise to help pathologists deliver faster, more consistent diagnoses and improve patient outcomes.”
Proscia DermAI Application
These new developments build upon the company’s DermAI application, which is available on the Concentriq digital pathology platform. The application provides AI-based classification for every skin case, which helps drive efficiency and quality gains.
In one of the most comprehensive studies ever carried out, DermAI demonstrated impressive performance and is continually being validated and deployed. Proscia is working with some of the top academic and commercial laboratories such as LabPON, Johns Hopkins School of Medicine, Unilabs, University Medical Center Utrecht, and University of California, San Francisco.
Julianna Ianni and collaborators Dr. Kiran Motaparthi and Dr. Jason Lee will be hosting a webinar on November 9th.
- Nanoantenna Enables Advanced Quantum Communication and Data Storage
- Human Brain Project Releases New Paper on Exascale Computing Power
- AI Researchers Estimate 97% Of EU Websites Fail GDPR Privacy Requirements- Especially User Profiling
- 5 Best Machine Learning & AI Books of All Time
- Earkick Raises $1M for Real-Time Mental Health Tracker