Breadcrumb

  1. Home
  2. Research
  3. Challenges
  4. DARPA Triage Challenge
  5. Challenge Events
  6. Data Competition | Event 1

Data Competition | Event 1

Unlocking Trauma Data’s Potential 
 

Algorithms for Early Prediction

dtc-data-competition-logo

In Challenge Event 1, teams were provided real-world trauma center datasets containing thousands of cases with >500 variables. This data was collected from pre-hospital (medevac, ambulance) until 4 hours after hospital admission. Teams developed algorithms to predict the need for life-saving interventions during this time.

In Challenge Event 2, the data size will significantly increase in size. Additional data types will add to the training data’s complexity. To win a prize, teams’ performance must exceed the specified baseline algorithm as well as specificity, sensitivity and lead time thresholds

Accordion item
Life Saving Interventions
Body

Life Saving Interventions (LSIs) are a medical care resource, activity, or procedure used to sustain life. LSIs have been grouped by type of injury and/or mechanism of treatment:

  • Airway & respiration: Intubation, Cricothyroidotomy, Laryngoscopy, Mechanic Ventilation
  • Vascular access & monitoring: Peripheral IV, Central Line Maintenance, Arterial Line Maintenance
  • Bleeding control: Tourniquet, Pelvic Binder
  • Vaso/cardioactive medications: Epinephrine, Dopamine, Phenylephrine, Vasopressin, Hydrocortisone
  • Crystalloid products: Normal Saline (>500cc), Lactated Ringers (>500cc)
  • Blood products: Packed Red Blood Cells, Plasma, Whole Blood Transfusion
  • Cardiovascular procedures: CPR, Defibrillation/Cardioversion, Pericardiocentesis
  • Chest decompression: Needle Decompression, Chest Tube Management
  • Neurologic products & procedures: Craniotomy/Craniectomy, Hypertonic Saline, Antiepileptic Medications
  • RSI sedation medications: Ketamine, Versed, Etomidate, Ativan, Propofol
  • Limb salvage: Amputation, Fasciotomy
  • Damage control procedures: Thoracotomy, Exploratory Laparotomy, IR Embolization
Evaluation Criteria
Body

Performance measured for each case using two metrics:

Jaccard Index (JI): How accurate were the predictions?
 

Jaccard



Prediction Lead Time (PLT): How early were the correct predictions?

Prediction Lead Time

 

Teams
Body

These innovative data teams developed cutting-edge algorithms to predict life-saving interventions from complex trauma data and ultimately improve medical decision-making.

  • AI TEMPO | Alert for Intervention using Timeseries EMergency Physiological Observations | DARPA-funded
  • ALICE | AI Life Saving Intervention Compute Engine | DARPA-funded
  • AUSTERE | AI User Supporting Triage and Evacuation Recommendation
  • CAMA | Center for Advanced Medical Analytics
  • CNA | Center for Naval Analyses
  • Coordinated Robotics
  • CRITIC | Continuous Review and Intervention for Timely Care | DARPA-funded
  • LENS | LSI Early Notification System | DARPA-funded
  • MGB-Harvard
  • MSAI
  • Robotika
  • TrueFit.AI

 

See all teams  |  Team qualifications guide 

Research Infrastructure for Trauma with Medical Observations
Leaderboard

Scoring criteria: Accuracy of LSI prediction (for any LSI and for more specific classifications, such as hemorrhage or airway interventions)

 TeamsScore 12
Coordinated Robotics*625434191
MSAI*555364191
CAMA*532368164
LENS310129181
AI TEMPO 26891177
CRITIC 18921168
TrueFit.AI*17937142
Robotika*14327116
ALICE1091891
AUSTERE 614912
MGB-Harvard*441331
CNA*35827


* Self-funded

darpa-challenge-dtc-data-competition
Source: DARPA | Jahyra Catala
DTC heartbeat right

Overview
 

Challenge Event 1

2024

Competitions
Systems  |  Data |  Virtual

Challenge Event 2

2025

Competitions
Systems  |  Data

Challenge Event 3

2026

Competitions
Systems  |  Data

 

DTC heartbeat right
DARPA Triage Challenge 
Resources  |   FAQs  |  Terms and Conditions  |  YouTube Playlist  |  Contact

Contact