Evidence map›Paper›PMID 41836285›Full record

ArticleNEJM AI2025

A Case Study of AI-Enabled Software as a Medical Device Cleared by the FDA for Assessing Hemorrhage Risk Index (APPRAISE-HRI) after Trauma.

Andrew Frock, Jeffrey T Robbins, Francisco G Vital-Lopez, Valmik Desai, Gheorghe Doros, Barry E Sands, Arunkumar Prabhakaran, Christopher Nemeth, Gregory T Rule, Jason L Sperry and 16 more

Abstract read
In one paragraph

Article in NEJM AI, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 3 papers.

0numbers the graph read from it
0cells of the map it votes in
3citing papers in PubMed
–field-weighted citation impact
1 · What the graph read from it

What it found

Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.

The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.

2 · The registry

The trial behind it

Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.

Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.

3 · Its place in the literature

Who cites it

3 citing papers in PubMed.

  1. Development and external validation of a novel prediction model for the TraumaTriage App.European journal of trauma and emergency surgery : official publication of the European Trauma Society · 2026
    Article
  2. Review
  3. Review
4 · The record

Corrections and comments

PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.

5 · Who and what money

Authors and funding

26 authors.

Andrew FrockDepartment of Defense Biotechnology High Performance Computing Software Applications Institute, Defense Health Agency Research and Development, Medical Research and Development Command, Fort Detrick, MD.ORCID 0000-0002-9569-867X
Jeffrey T RobbinsDepartment of Defense Biotechnology High Performance Computing Software Applications Institute, Defense Health Agency Research and Development, Medical Research and Development Command, Fort Detrick, MD.ORCID 0009-0004-8130-3376
Francisco G Vital-LopezDepartment of Defense Biotechnology High Performance Computing Software Applications Institute, Defense Health Agency Research and Development, Medical Research and Development Command, Fort Detrick, MD.ORCID 000-0001-5314-8373
Valmik DesaiDepartment of Defense Biotechnology High Performance Computing Software Applications Institute, Defense Health Agency Research and Development, Medical Research and Development Command, Fort Detrick, MD.ORCID 0009-0004-5447-6052
Gheorghe DorosRQMIS, Amesbury, MA.ORCID 0000-0001-5524-4721
Barry E SandsRQMIS, Amesbury, MA.ORCID 0009-0000-2818-898X
Arunkumar PrabhakaranRQMIS, Amesbury, MA.ORCID 0009-0008-1269-3197
Christopher NemethApplied Research Associates, Albuquerque, NM.ORCID 0000-0001-5990-0175
Gregory T RuleApplied Research Associates, Albuquerque, NM.ORCID 0009-0008-0514-7366
Jason L SperryDepartment of Surgery, University of Pittsburgh, Pittsburgh.ORCID 0000-0003-3917-6479
Francis X GuyetteDepartment of Emergency Medicine, University of Pittsburgh, Pittsburgh.ORCID 0000-0002-9151-4896
Stephen R WisniewskiSchool of Public Health, University of Pittsburgh, Pittsburgh.ORCID 000-0002-3877-9860
Ernest E MooreDepartment of Surgery, Ernest E. Moore Shock Trauma Center at Denver Health, Denver.ORCID 0000-0002-0419-4848
Martin SchreiberDonald D. Trunkey Center for Civilian and Combat Casualty Care, Oregon Health and Science University, Portland.ORCID 0000-0002-4430-6779
Bellal JosephDepartment of Surgery, Division of Trauma, Surgical Critical Care, Burns, and Acute Care Surgery, University of Arizona, Tucson.ORCID 0000-0002-2205-3061
Chad T WilsonDepartment of Surgery, Baylor College of Medicine, Houston.ORCID 0009-0007-4932-0258
Bryan CottonDepartment of Surgery, McGovern Medical School at the University of Texas Health Science Center, Houston.ORCID 0000-0003-4184-6742
Daniel OstermayerDepartment of Emergency Medicine, McGovern Medical School at the University of Texas Health Science Center, Houston.ORCID 0000-0003-3205-4161
Brian G HarbrechtDepartment of Surgery, University of Louisville, Louisville, KY.ORCID 0000-0002-6469-5265
Mayur B PatelSection of Surgical Sciences, Department of Surgery, Division of Acute Care Surgery, Vanderbilt University Medical Center, Nashville.ORCID 0000-0001-5230-0871
Suzanne TamangDepartment of Veterans Affairs, Palo Alto Health Care System, Menlo Park, CA.ORCID 0000-0003-2077-4620
Sanjay MalunjkarDepartment of Medicine, Stanford University School of Medicine, Stanford, CA.ORCID 0009-0001-9269-0939
David A SpainDepartment of Surgery, Stanford University School of Medicine, Stanford, CA.ORCID 0000-0002-0477-7613
Andrew T ReisnerDepartment of Emergency Medicine, Massachusetts General Hospital, Boston.ORCID 0000-0001-6450-5639
Jonathan D StallingsJoint Trauma System, Defense Health Agency, Fort Sam Houston, San Antonio, TX.ORCID 0000-0002-6430-5888
Jaques ReifmanDepartment of Defense Biotechnology High Performance Computing Software Applications Institute, Defense Health Agency Research and Development, Medical Research and Development Command, Fort Detrick, MD.ORCID 0000-0001-7292-2029

Funding

Intramural VA VA999999
6 · The paper itself

Abstract

Hemorrhage is the leading cause of preventable death on the battlefield, yet combat medics lack clinical decision support systems to help stratify hemorrhage risk in trauma casualties. We previously trained the Automated Processing of the Physiological Registry for Assessment of Injury Severity - Hemorrhage Risk Index (APPRAISE-HRI) software to associate patterns in vital signs (heart rate and blood pressure) collected from trauma patients with three HRI levels: I (low), II (average), or III (high). To independently validate APPRAISE-HRI and obtain U.S. Food and Drug Administration (FDA) clearance, we collected trauma registry and continuous vital sign data from 5895 trauma patients (543 with hemorrhagic injuries and 5352 controls) in an emergency department or during prehospital transport to one of eight medical centers. The study outcome was hemorrhagic injury, defined by documented injuries and blood transfusion. Using the likelihood ratio to assess the ability of APPRAISE-HRI to stratify hemorrhage risk, we found that hemorrhagic patients were 6.88 times as likely as controls to be at level III, strongly suggesting the presence of hemorrhage at this level. Similarly, hemorrhagic patients were 0.18 times as likely as controls to be at level I, suggesting the absence of hemorrhage at this level. Hemorrhagic patients were almost as likely as controls to be at level II (0.70 times as likely). Subsequently, the U.S. Department of Defense obtained FDA 510(k) clearance for the artificial intelligence-enabled APPRAISE-HRI Class II device (K233249), the first software as a medical device approved for assessing hemorrhage risk in trauma patients, allowing for triage and identification of casualties who need immediate attention and evacuation. (Funded by the U.S. Army Medical Materiel Development Activity and the Combat Casualty Care Program Area Directorate (CCCPAD) of the U.S. Army Medical Research and Development Command (USAMRDC), Fort Detrick, MD and others.).

Identifiers

PMID41836285
PMCPMC12981239

What OpenQuestion holds

Textmetadata
LicenceTDM
Read underepoch 390

Registered trials

None linked

Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.