Developing Artificially Intelligent Agents to Support Earth Independent Medical Capabilities During Human Exploration-Class Space Missions

Authors: William R. Buras, David C. Hilmers

Abstract

Artificially-intelligent agents are being developed to support NASA crewmembers during exploration missions where Earth-based medical support is unfeasible. This article outlines the need for such agents, provides a functional overview, and presents a concept of operations demonstrating how they could support crew health. Key desirable characteristics to support crew medical officers are listed. This technology also has potential applications in remote terrestrial environments where expert consultation is inaccessible. Exploration-class missions to the Moon, Mars, asteroids, and beyond present unique medical challenges not encountered during low Earth orbit (LEO) missions. On the International Space Station (ISS), astronauts maintain constant real-time contact with Mission Control Center (MCC), enabling immediate medical consultation, diagnosis, treatment, resupply, and evacuation if needed. However, during deep space missions, communication delays (up to 45 minutes round-trip to Mars), lack of resupply, and evacuation impossibility require crews to be increasingly autonomous in medical care — a paradigm known as Earth Independent Medical Operations (EIMO). To enable EIMO, crews must substitute onboard expertise for that currently provided by MCC. Limited pre-flight medical training and skill degradation during missions compound the challenge. Cognitive impairments such as “space fog” further complicate medical decision-making. Advances in artificial intelligence (AI) offer promising tools to augment crew capacity in training, diagnosis, treatment, and medical inventory management. These include mixed reality for just-in-time training, AI-driven chatbots using large language models (LLMs) to emulate ground-based expertise, and onboard reference materials. Imagine a crew medical officer (CMO) on a Mars mission faced with a crewmember experiencing chest pain. With a 45-minute communication delay to Earth, an AI-based system could assist in diagnosis and management by asking relevant questions, guiding examinations, and recommending interventions based on available resources.

Key Qualities for AI Medical Support Systems:

• Acts like a ground-based medical consultant, asking questions and making recommendations.
• Functions independently without real-time communication or internet.
• Complies with strict mass, volume, power, and data constraints.
• Contains a comprehensive medical database covering common and rare conditions.
• Conducts symptom-based dialogs, guiding history-taking, examination, testing, and management.
• Optimizes question flow logically and succinctly.
• Recognizes emergent conditions requiring immediate treatment.
• Provides procedural guidance and just-in-time refresher training.
• Adapts communication complexity to the operator’s expertise level.
• Continuously monitors vital signs and alerts crew to medical events.
• Responds instantly to medical information queries.
• Supports hands-free voice interaction.
• Maintains detailed knowledge of crewmembers’ medical histories and onboard resources.

Methods

Agents in Development

Our team is developing AI tools to meet EIMO needs, focusing on a localized Large Language Model called Space Medicine GPT (SGPT). SGPT is designed to fulfill most of the key qualities above. Unlike generic chatbots, SGPT uses novel prompt engineering to create an interactive diagnostic dialog that mimics a clinician’s thought process, employing Bayesian logic to prioritize questions and tests. SGPT operates offline using a distilled LLM (<50 billion parameters) suitable for laptops, tablets, and smartphones, addressing spaceflight constraints. It integrates vector databases for rapid retrieval of evidence-based medical information. Recorded physician-patient conversations train SGPT to respond as an expert consultant. When a symptom like chest pain arises, SGPT guides the CMO through history-taking, physical examination, and testing, prioritizing emergent conditions such as acute coronary syndrome or tension pneumothorax. It adapts its communication to the user’s expertise and continuously monitors vital signs via wearables, alerting to dangerous trends.

Augmented Reality and Procedural Guidance

The Intelligent Medical Crew Agent (IMCA) uses augmented reality (AR) to provide step-by-step guidance for medical procedures. AR devices project imaging, diagrams, checklists, and videos in the user’s field of view. IMCA supports procedures like pericardiocentesis, chest tube placement, and abscess drainage, tailored to novice or expert users. A key IMCA component is the Visual Ultrasound Learning, Control and Analysis Network (VULCAN), which monitors ultrasound procedure execution and provides real-time corrective feedback using AI, computer vision, and telemetry. This “closed loop” guidance ensures proper technique and outcomes.

Challenges and Aspirations

While promising, significant challenges remain. The complexity of clinical reasoning is difficult to emulate optimally. Training must rely on best medical practices to avoid errors. Offline operation requires distillation of large medical databases to fit spaceflight constraints. Rigorous testing in Earth analog environments (e.g., Antarctica, wilderness, military deployments) is essential. These AI tools represent only part of a comprehensive clinical decision support system needed for space missions, which also includes pharmacy, supply management, health maintenance, and environmental control. Looking forward, advances in hardware will enable AI systems like SGPT to run on portable devices, providing expert-level medical support not only in space but also in remote terrestrial settings, disaster zones, and conflict areas lacking communication infrastructure.

Data availability

All data presented in this manuscript is available in the presented figures.

Code availability

The code used in this study is proprietary and cannot be publicly shared due to confidentiality agreements. Researchers interested in accessing the code should contact william.buras@tietronix.com to discuss potential collaborations under a non-disclosure agreement (NDA).

References

(References omitted for brevity; see original article for full list.)

Acknowledgements

We thank NASA and TRISH for their research support. This work was supported by NASA SBIR funding under grants NNX16CC522P, NNX17CC12C, 80NSSC120C0541, 80NSSC21C0578, 80NSSC23PB612.

Author information

• William R. Buras, Tietronix Software Inc., Houston, TX, USA
• David C. Hilmers, Baylor College of Medicine, Translational Research Institute for Space Health (TRISH), Houston, TX, USA

Ethics declarations

The authors declare no competing interests.

---

Source: Developing artificially intelligent agents to support earth independent medical capabilities during human exploration-class space missions, npj Microgravity, 2025.

---

Frequently Asked Questions (FAQ)

AI Agents for Medical Support in Space

Q: What is the primary challenge addressed by developing AI agents for space missions? A: The primary challenge is the need for Earth Independent Medical Operations (EIMO) during exploration-class space missions, where communication delays and the impossibility of evacuation necessitate autonomous medical capabilities for the crew. Q: What are "Earth Independent Medical Operations" (EIMO)? A: EIMO refers to the capability for crews to provide their own medical care without real-time support from Earth-based Mission Control Centers, due to factors like communication delays or lack of access. Q: How do AI agents like SGPT assist in medical diagnosis in space? A: SGPT, a localized Large Language Model, mimics a clinician's thought process by asking diagnostic questions, guiding examinations, and recommending treatments based on a comprehensive medical database and the specific symptoms presented by a crew member. Q: What is the role of Augmented Reality (AR) in the IMCA system? A: The Intelligent Medical Crew Agent (IMCA) uses AR to project step-by-step guidance, diagrams, and videos directly into the user's field of view, aiding in complex medical procedures. Q: How does VULCAN contribute to medical support in space missions? A: VULCAN, a component of IMCA, uses AI and computer vision to monitor ultrasound procedures in real-time, providing corrective feedback to ensure proper technique and successful outcomes. Q: What are some of the key desirable characteristics for AI medical support systems in space? A: Key characteristics include functioning independently without real-time communication, complying with strict mass/volume/power constraints, possessing a comprehensive medical database, conducting logical symptom-based dialogues, recognizing emergent conditions, and providing procedural guidance. Q: Can these AI medical support systems function without an internet connection or real-time communication with Earth? A: Yes, a crucial requirement for these systems, like SGPT, is to function offline to support missions with significant communication delays. Q: What are the potential terrestrial applications for this AI medical technology? A: This technology has potential applications in remote terrestrial environments, disaster zones, and conflict areas where access to expert medical consultation is limited or impossible.

Crypto Market AI's Take

The development of sophisticated AI agents like Space Medicine GPT (SGPT) and the Intelligent Medical Crew Agent (IMCA) highlights a broader trend in AI application: the creation of specialized, autonomous systems capable of operating in environments with limited connectivity and resources. This mirrors the challenges faced in decentralized finance (DeFi) and cryptocurrency markets, where automated trading bots and AI-driven analytics platforms are increasingly vital for navigating volatile conditions and executing strategies without constant human oversight. Just as these space-bound AI agents must be robust and self-sufficient, so too must AI tools in the crypto space be reliable and data-driven to manage risk and identify opportunities. Our platform leverages AI for precisely these purposes, providing insights and tools to help users navigate the complexities of the digital asset landscape. You can learn more about how AI is transforming financial markets and discover strategies for leveraging AI in your investment journey on our AI Agents hub.

More to Read:

• AI-Driven Crypto Trading Tools Reshape Market Strategies in 2025
• The Future of Finance: How AI is Revolutionizing Investment
• Understanding Blockchain: A Comprehensive Beginner-Friendly Guide