Maintenance window scheduled to begin at February 14th 2200 est. until 0400 est. February 15th

(e.g. yourname@email.com)

Forgot Password?

    Defense Visual Information Distribution Service Logo

    USMEPCOM Invests in AI to Aide Prescreen Process

    USMEPCOM Invests in AI to Aide Prescreen Process

    Photo By Derrik Noack | Graphic by Derrik Noack read more read more

    NORTH CHICAGO, ILLINOIS, UNITED STATES

    10.04.2023

    Story by Derrik Noack 

    U.S. Military Entrance Processing Command

    The 2022 rollout of MHS GENESIS across USMEPCOM brought access to authoritative medical history for applicants to military service. This access has allowed for better medical decision making, but it has also meant an increased workload with more medical information to review for more applicants.

    To adjust to this new paradigm, USMEPCOM has updated guidance and policies, utilized virtual capabilities, and even added personnel just to execute prescreens. Currently, it can take up to two hours for a provider to medically prescreen an applicant, but a new capability is on the horizon to significantly decrease that time: Artificial Intelligence (AI).
    Led by the Command Advisory Group (CAG), concept development for the AI project began in summer 2022. The objective was to develop and implement an AI capability that would increase efficiency, effectiveness, validity and speed of military applicant medical screening. To achieve that objective, they partnered with the Department of Defense’s Chief Digital and Artificial Intelligence Office (CDAO), which will build the software.

    “Right now, we’re heading into the execution phase of the project where they’re developing the prototype,” said Randi Mcrobbie, CAG program analyst serving as AI project coordinator. “That in itself has a follow-up phase of testing and then that goes into full operating capability.”

    Initial operating capability could come possibly as early as September 2024, based on timing of AI access to MHS GENESIS data. While still in “phase zero,” plans are for the AI to screen an applicant’s medical history and place them into different bins based on the complexity of their medical history. For applicants flagged for disqualifying conditions, the AI will highlight those issues to streamline medical provider review.

    “The review of an individual’s medical history can be quite extensive,” said Anthony Cerda, Innovation Facilitation Team (IFT) lead for this project. “AI will review and extract only the necessary data needed for the provider to make a decision. Prescreen review is currently the longest part of the medical process. By giving the providers only what they need, they can more quickly and efficiently make their decision.”

    The AI will only act as a tool and will not have the capabilities to make decisions or determine an applicant’s ability to serve on the All-Volunteer Force.

    “This will only be part of the first stage of the prescreen process,” said McRobbie. “This AI capability is limited to determining the significance of the applicant’s medical history. It will sort through the data for easier review by medical prescreen personnel to inform medical provider’s decisions. A human still makes that determination to schedule a medical exam or if further information needs to be found.”

    While still in the first phases of development, this first-time partnership between USMEPCOM and CDAO could lead to even more efficient ways to evaluate applicants and thereby sustain the All-Volunteer Force.

    “What we hope is that this project allows us to explore other areas that we can use AI,” said McRobbie. “This is really groundbreaking for us. It’s the first time we’ve partnered with this type of organization. We are going to look to what it can lead to for the future, not just solving the problem at hand with medical prescreening, but what we can explore later.”

    LEAVE A COMMENT

    NEWS INFO

    Date Taken: 10.04.2023
    Date Posted: 10.05.2023 16:09
    Story ID: 455131
    Location: NORTH CHICAGO, ILLINOIS, US

    Web Views: 91
    Downloads: 0

    PUBLIC DOMAIN