Predicting Exploration Crew Medical Officer Training Needs: Applying Evidence-Based Predictive Analytics to Space...

Area of Science:

• Space Medicine
• Medical Training
• Predictive Analytics

Background:

• Identifying medical training needs for deep space missions is crucial.
• Exploration-class medical officers require specialized skillsets.
• Predictive analytics offers a novel approach to curriculum design.

Purpose of the Study:

• To evaluate the utility of predictive analytics in designing medical curricula for space exploration missions.
• To assess the Medical Extensible Database Probabilistic Risk Assessment Tool (MEDPRAT) for curriculum development.
• To identify common and mission-specific medical training requirements.

Main Methods:

• Utilized a preliminary version of NASA's MEDPRAT tool.
• Applied predictive analytics to 5 distinct Design Reference Mission (DRM) profiles.
• Employed a leave-one-out analysis with a 5% risk increase threshold to identify curriculum elements.

Main Results:

• Between 4-32 curriculum elements (partial treatment) and 13-126 (full treatment) met the risk threshold across DRM profiles.
• Identified 13 core medical capabilities applicable to at least 3 of the 5 DRM profiles.
• Demonstrated varying coverage of skillsets for different mission types (e.g., 100% for Starship orbital, 41% for Mars).

Conclusions:

• Predictive analytics can efficiently generate evidence-based, mission-specific medical curricula for space exploration.
• This approach supports a human-machine team strategy for optimizing medical training.
• The technique has the potential to enhance astronaut health outcomes and streamline training development.