Spectral AI Shares Burn Validation Study Results for the DeepView System

Spectral AI’s DeepView System outperforms burn physicians in identifying non-healing tissue, according to the company’s Burn Validation Study results.

Spectral AI’s DeepView System is being developed as a predictive device to offer clinicians an immediate and objective assessment of a burn wound’s healing potential before treatment or other medical intervention. The image processing algorithm employed by the DeepView System uses multispectral imaging that is trained and tested against a proprietary database of more than 340 billion clinically validated data points. The DeepView System is non-invasive and mobile (cart-based).

The Burn Validation Study—initiated in January 2024 and completed in March 2025—represented one of the largest U.S. burn trials ever conducted. The study enrolled and analyzed data obtained from 164 total patients comprised of 115 adult and 49 pediatric patients in U.S. burn centers and emergency departments. The trial’s goal was to further demonstrate the innovative and versatile nature of Spectral AI’s DeepView technology, as well as its ability to predict burn wound healing potential on the first day of injury with greater performance and speed than current methods. The Burn Validation Study results, together with rigorous statistical analysis, show the DeepView System outperforms burn physicians’ clinical judgment by a large margin.

Related: Spectral AI to Scrutinize Burn Imaging Data From E.R.s

The Burn Validation Study results assessed the sensitivity, specificity and Dice score of the DeepView System at both a pixel-level and image-wise level against burn physicians’ clinical judgment. Key findings included the following:

• Sensitivity: The DeepView System demonstrated a statistically significant improvement in identifying non-healing tissue compared to burn physicians, as judged on sensitivity. At the image-wise level, the DeepView System scored 86.6% while clinical judgment annotation (CJA) of burn physicians scored 40.8%. At the pixel-wise level, the DeepView System scored 81.9% and CJA of burn physicians scored 38.8%.
• Dice Score: The DeepView System achieved statistically significant higher Dice Scores when compared to those derived from burn physicians’ CJA, representing the improved pixel-wise evaluation between predicated and true segmented wound areas with the DeepView System performing at 68.5% and burn physician’s CJA at 39.2%.
• Specificity: The Deepview System significantly outperformed the anticipated results for image-wise specificity in segmenting non-healing wound areas with a result of 61.2% (vs. an anticipated result of 36%) and CJA of 79.1% reflecting burn physician’s conversative assessment of burn areas.

“We believe these are excellent results and we are thrilled with the analysis of our DeepView System in our Burn Validation Study,” said Dr J. Michael DiMaio, M.D. “The DeepView System exceeded our expectations in terms of predictive performance. Following the FDA’s review, if authorized by the agency, our hope is that this tool will provide an objective and immediate prediction of non-healing burn tissue to expedite patient care and reduce system costs across the board. I would like to thank Jeffrey C. Carter, M.D., the principal investigator for the Burn Validation Study, our entire team at Spectral AI, and especially the patients and esteemed clinicians who participated in this study. We look forward to bringing this predictive diagnostic tool to the United States marketplace as soon as possible.”

Spectral AI plans to submit these results to the U.S. Food and Drug Administration (FDA) as part of its regulatory submission by the end of Q2 2025 and hopes to win De Novo classification for the DeepView System early next year. Based on its current manufacturing arrangements, Spectral AI will be prepared to commercialize the DeepView System in the United States, if ultimately approved, as soon as approval is obtained.  

This project is being supported in whole or in part with federal funds from the U.S. Department of Health and Human Services; Administration for Strategic Preparedness and Response; BARDA, under contract number 75A50123C00049.

Spectral AI Inc. is a Dallas-based predictive AI company focused on medical diagnostics for faster and more accurate treatment decisions in wound care, with initial applications involving patients with burns. The company is working to revolutionize wound care management by “Seeing the Unknown” with its DeepView System, a predictive device that gives clinicians an objective and immediate assessment of a burn wound’s healing potential before treatment or other medical intervention. With algorithm-driven results and a goal of exceeding the current standard of care in the future, the DeepView System is expected to provide fast and accurate treatment insight towards value care by improving patient outcomes and reducing healthcare costs.