The Department of Veterans Affairs (VA) announced the finalists selected to move forward in its AI Tech Sprint competition on March 13, which targets the development of AI solutions to address burnout among healthcare workers.
The 25 finalists will now prepare their designs for a panel of judges to determine what portion of up to $1 million in collective prize money each finalist will receive in May.
According to the VA, the winning solutions will focus on reducing administrative workload by helping clinicians take notes during medical appointments or integrate patients’ medical records.
“Reducing burnout among health care workers is a top priority for VA, especially at a time when VA is delivering more care and more benefits to more Veterans than ever before,” the VA said.
The Tech Sprint – announced in October – is a part of President Biden’s executive order on AI and VA’s efforts to use trustworthy AI solutions to improve health care and benefits for veterans, their families, caregivers, and survivors.
“As we deliver more care to veterans than ever before, our AI Tech Sprints are tackling an important challenge: healthcare worker burnout,” said Under Secretary for Health Shereef Elnahal. “We’re leveraging our expertise in health care with the ingenuity of private sector – and we will drive real results for Veterans and improve the work environment for our employees. And at every turn, we will adhere to President Biden’s Trustworthy AI principles.”
The competition is designed to reduce the immense administrative workload of healthcare workers, which is frequently cited as a top cause of burnout. There are two tracks of this sprint:
- The “Ambient Dictation for Clinical Encounters” track focuses on speech-to-text solutions that can be used to automatically take notes for both the doctor and their patient and upload key excerpts to medical files. The finalists for this track are Abridge; Althea Health; ARETUM; Cognosante Military and Veterans Health; Commure; Contrast AI; Credence Management Solutions; DeepScribe; TranscribeMD AI; Knowtex; Nuance; QuantumTechIT; Sourceree; Tali AI; and Veterans EZ Info.
- The “Community Care Document Processing” track looks for an AI tool that can reduce the time needed to integrate non-VA medical records into patients’ VA records in an easy to access way. The finalists for this track are Abstractive Health; Booz Allen Hamilton; Commure; Healow and eClinicalWorks; Instabase; Joint Venture iCatalyst and Semiring; TranscribeMD AI; Palantir Technologies; Wellovate; and Wired Informatics.
VA received more than 200 applications, which were scored by set criteria, resulting in 100 teams from diverse backgrounds being selected for the next phase. Sixty participants then advanced to an output challenge. After weeks of putting their AI programs to the test, 25 finalists were selected.
“We’re working to identify the winners at the end of May. So, it’s at that point that we’ve worked on bringing out pilots to the rest of VA,” Kimberly McManus, deputy chief technology officer for AI at VA, said during a press briefing with reporters on March 13. “But we’ve seen really good feedback from other health systems. And so, we’re really optimistic about those, and we hope to make progress on that in the next year.”
