We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!
Spectral MD, a Dallas, Texas-based startup developing an AI-powered burn and wound imaging system, today announced that it has secured $27 million from the Biomedical Advanced Research and Development Authority (BARDA), part of the Office of the Assistant Secretary for Preparedness and Response at the U.S. Department of Health and Human Services. The department will dole out payments over the next 22 months for an expanded proof-of-concept clinical trial that targets burn care assessment in pediatric populations, with options for additional funding of up to $92 million.
Spectral MD CEO Dr. J. Michael DiMaio said the fresh capital, which builds on the $26 million Spectral MD received from BARDA between 2013 and 2018 to prove out its initial product plan, will be used accelerate development of DeepView. DiMaio added that the funds will also bolster clinical tests of its system with other wound types, including amputations, critical limb ischemia, and diabetic foot ulcers.
“DeepView is an advanced technology that analyzes tissue characteristics to identify the boundaries of dead, damaged, and healthy tissue without even touching the patient,” said DiMaio. “[It’s] portable and is intended to be positioned over any area of the body to provide rapid, real-time results [that] will allow more accurate determination than is possible with human assessment. We are elated that Spectral MD has been selected for this additional critical support and recognize that we are in very exclusive company in receiving this funding.”
DeepView — a recipient of the Federal Food and Drug Administration’s Breakthrough designation, which indicates it might demonstrate clinically significant improvement over available therapy — is a portable device that taps computer vision to spot dead tissue. The analysis is conducted on IBM Cloud, expediting the processing of wound data by up to 90%.
DeepView characterizes chemical structures based on the light they reflect (including non-visible wavelengths) using a technique called multispectral imaging, enabling it to differentiate pathologies from healthy skin in a way that’s impossible for the naked eye. Patient scans are ingested by machine learning algorithms trained on thousands of multispectral images of annotated burns, which suss out the interactions between photons and the tissue beneath the skin’s surface to determine whether a wound is serious enough to warrant surgery.
Spectral MD cites studies estimating that 30% of burns are misdiagnosed, which it claims DeepView reduces to 5%. Furthermore, the companys says automated diagnoses can reduce the size and amount of skin grafting performed during surgery, potentially improving surgical outcomes and shortening hospital stays.
“We are honored to receive this second round of support from BARDA to assist in furthering the development of our DeepView system,” said Spectral MD executive vice president and CTO Wensheng Fan.
Spectral MD is one of several startups applying computer vision to health care challenges. Paige.ai recently raised $25 million to continue its work in cancer diagnosis with AI models trained using clinical imaging data, and Aidoc in April nabbed $27 million for AI-assisted head, chest, abdomenal, and spinal exams. Healthy.io taps machine learning to conduct urinalysis, and Sight Diagnostics — which raked in $27.8 million in funding in February — leverages a family of algorithms to perform point-of-care complete blood count (CBC) tests with no more than a pinprick of blood.
Perhaps it’s no wonder the computer vision in health care market segment is anticipated to be worth $1.45 billion by 2023, according to Markets and Markets.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn more about membership.