RMP-NEURO-001 Head injury: urgent relay, supplies, and recording of the checks the CMO does, with prompt-only repeat checks (the robot does not assess)

1 · Identity and scope

After any blow to the head, the robot tells the flight surgeon and CMO at once (URGENT on any red flag), brings the medical kit and any items the CMO asks for, shows the onboard head-injury checklist, records the answers and the checks the CMO does (alertness score, pupils, limb strength and sensation, memory questions) with exact times, and then offers prompt-only repeat checks on the schedule the flight surgeon sets. It never scores, interprets, compares or clears. Because CT is not available, the flight surgeon works from a list of findings that on Earth would mean CT within 1 hour or transfer.

Why this protocol was chosen. NICE (NG232) lists the reasons a person with a head injury should be referred from the community to an emergency department, including any loss of consciousness, a Glasgow Coma Scale score below 15, memory loss, persistent headache, any vomiting, a seizure, new neurological signs, anticoagulant or antiplatelet medicines, and continuing concern; CT is the primary test, and it is not yet readily available in spaceflight. There are no adequate observational data on head injuries in orbit; NASA has modeled the risk instead. The red flags therefore come from Earth guidance (NICE NG232, cross-checked against the Canadian CT Head Rule, which was derived in adults with GCS 13-15 in emergency departments with CT) and are applied here without CT. An exact, time-stamped record of each check helps the flight surgeon see a change early, and the decision to watch, treat or evacuate stays with humans.

Autonomy: Max rung raised from 1 to 2 for supervised fetching and recording, as in batch 1. The robot never moves the head or neck and never scores a check.

Related: RMP-MSK-004 · RMP-MSK-001 · RMP-NEURO-002 · RMP-NEURO-003 · RMP-EYE-001 · RMP-ENT-001 · RMP-SKIN-001 · RMP-EGR-001

2 · Roles

Medical authority

Flight surgeon (ground). Holds medical authority for every run.

Onboard lead

Crew medical officer (CMO), acting under delegated care; uses the pre-approved onboard checklist when there is no link.

Others

• Injured crewmember: answers questions if able; can stop recording
• Second trained CMO: receives alerts and does the checks if the CMO is the patient
• Any crewmember who saw the injury: describes what happened

Commercial flights

On commercial flights, 'flight surgeon' and 'CMO' mean the operator's designated medical authority and trained onboard responder. If no trained responder is aboard, this protocol is not used. (NASA-STD-3001 requires two CMOs per vehicle for NASA programs [V1 3006]; FAA rules set no medical standards for spaceflight participants.) [12]

3 · Preconditions and consent

• PRE-1 A crewmember (the injured one, a witness or the CMO) started the protocol by command.human_start == trueV2 10174
• PRE-2 The injured crewmember is answering. If not, URGENT alert at once and a crewmember checks in person (X7); the robot only fetches on request.crew_responsive == true
• PRE-3 Robot self-test passed; it knows the stowage location of the medical kit.V2 10161
• PRE-4 Robot and supplies restrained when not moving.V2 7046
• PRE-5 Link status and current delay known.link_state in [live, delayed, down]

Consent

The crewmember agrees to questions and recording by voice. Any question can be declined; an explicit "no" is logged as declined for that question only.

Emergency consent

If the crewmember cannot answer, the URGENT alert stands, the robot asks nothing more and a crewmember checks in person; the CMO's entries are recorded as part of the medical record.

Withdrawal

The crewmember can stop recording at any time; any alert already sent is not withdrawn.

4 · Required sensor inputs

5–6 · Steps and decision points

Each step names who acts. Decision points sit after the step they follow and say who decides: Robot Crew Ground. Any stop rule in section 7 can interrupt any step.

1. S1RobotRung 2

   On a crewmember's command, open the log and send a "head injury" note to the flight surgeon and the CMO at once, with the time of the injury and what happened; it becomes an URGENT alert on any red flag. Every URGENT alert goes to the CMO and the flight surgeon at the same time (queued for the flight surgeon if the link is down). If the CMO is the patient, is busy treating, or does not answer, the alert goes to the second trained CMO, or the crewmember named before flight [V1 3006]. An URGENT alert that is not acknowledged within <set by validation> seconds is repeated and sent to the next named crewmember.

   Done when: Note or alert acknowledged, or escalated to the next named crewmember.

   V2 10174V1 6003V1 3006

2. S2RobotRung 2

   Ask the red-flag questions first by voice and log answers word for word. D1 is checked after every answer.

   Done when: Answers logged, or "no response" logged.

   D1DecisionRobot decides

   Checked after every answer and at any point in the run (even mid-question or mid-step): does any answer, reading or observation meet a red flag?

   • Yes → Stop the current question or step at once and apply the action listed for each matching red flag (URGENT alert, flag, or stop), with the answer or reading itself, never a cause. Every URGENT alert goes to the CMO and the flight surgeon at the same time (queued for the flight surgeon if the link is down). If the CMO is the patient, is busy treating, or does not answer, the alert goes to the second trained CMO, or the crewmember named before flight [V1 3006]. An URGENT alert that is not acknowledged within <set by validation> seconds is repeated and sent to the next named crewmember.
   • No → Continue; the flight surgeon still receives the head-injury note.
   D2DecisionRobot decides

   Did the injured crewmember answer the question?

   • Yes → Continue.
   • Declined explicitly (the crewmember says no to a question) → Log "declined" for that question only; the alert already sent stands.
   • No response, or the answer does not make sense → URGENT: log "no response" (never "declined"); the CMO, or the second trained CMO if the CMO is the patient, checks the crewmember in person at once.
3. S3RobotRung 2

   Bring the medical kit and any item the CMO asks for, closed, and restrain them within reach. This is an emergency fetch on a human command (CMO): no barcode check runs before hand-over and nothing waits for one. The robot reads the kit's own label and seal status only if that adds no delay, says "EMERGENCY FETCH – NOT VERIFIED (CMO command)", and "seal not intact" if so, and hands the kit over anyway. The fetch is logged as an override: time, who asked, the reason as stated, kit ID and seal status. It never says "verified". Show the onboard head-injury checklist on its screen.

   Done when: Items within reach; checklist shown; override logged.

   Limit: <set by validation>

   V2 7046

4. S4Crew

   The CMO does the checks on the checklist (Glasgow Coma Scale score, pupils, limb strength and sensation, memory and orientation) and decides about cervical spine protection under the onboard checklist (NICE NG232 section 1.6). Nobody moves the head or neck for the robot.

   Done when: Checks done and entered by the CMO.

   V1 3004

5. S5RobotRung 2

   Record each check the CMO enters with its time, and show the CMO the earlier checks in time order, in an identical layout whatever the values, headed "Entries as recorded by the CMO – not compared by the robot". No arrows, colours, highlighting or difference values.

   Done when: Entries time-stamped.

   V1 3004

   D3DecisionCrew decides

   Is any check the CMO entered worse than an earlier one, or does the CMO have any concern?

   • Yes (the CMO says so) → URGENT alert (RF11) at once; the robot records the CMO's words. If a second trained CMO is available, they repeat the check to confirm; this never delays the URGENT alert.
   • No → Continue the watch plan.
6. S6RobotRung 2

   After any URGENT alert: After an URGENT alert the robot does not leave on its own: it stays at its rest position in the module, with its screen and voice link available, until a human has taken over or tells it to go. It never follows, blocks or touches anyone. It keeps time-stamping what humans say they did, and repeats the alert if it is not acknowledged.

   Done when: Robot at rest in the module until a human releases it; alert acknowledged or escalated.

   V2 10173V1 3006

7. S7Ground

   The flight surgeon (or the CMO under the onboard checklist when there is no link) decides the watch plan, wake checks, work limits, return to duty, and any evacuation [V1 3007]. Because CT is not yet readily available in spaceflight, findings that on Earth would mean CT within 1 hour or transfer (NG232 1.5.5, 1.5.8, 1.5.9, 1.5.13, 1.9.13) are passed to the flight surgeon as reasons to consider evacuation: a GCS score of 12 or less at the first check, or below 15 two hours after the injury; a suspected open, depressed or basal skull fracture; a seizure after the injury; a focal neurological deficit; more than 1 episode of vomiting; loss of consciousness or amnesia together with a bleeding or clotting disorder; anticoagulant or antiplatelet treatment other than aspirin alone; and any sign of deterioration on repeat checks. The Canadian CT Head Rule is a cross-check only. The robot applies none of this list; it shows it to the flight surgeon with the entries. On Mars-class missions evacuation may not be possible, and the CMO follows the pre-approved checklist. The robot logs the decisions.

   Done when: Decision logged.

   V1 3007

   D4DecisionGround decides

   What is the watch plan?

   • Link live → Flight surgeon decides; robot logs.
   • Link down or delay too long → The CMO decides under the pre-approved onboard checklist; the flight surgeon reviews the log at the next contact.
8. S8RobotRung 2

   Every <set by flight surgeon> hours for <set by flight surgeon> hours after the injury (the repeat-check schedule the flight surgeon sets), prompt the CMO to repeat the checks. At each scheduled time the robot prompts the crewmember on their own screen; it approaches only when the crewmember (or the CMO) accepts, and a prompt never starts the robot. Wake checks happen only if the flight surgeon or CMO set them; a wake check with no response is URGENT.

   Done when: Each prompt accepted and the checks entered, or escalated.

   V2 10174

9. S9RobotRung 2

   Send the flight surgeon one package: what happened, answers, each check with its time, alerts and decisions. When the CMO says the event is over, run RMP-PHARM-001 reconciliation on each kit fetched: record the items the CMO or crew state were used, the remaining contents against the kit list, seal status and expiry, labelled "inventory reconciliation after emergency use – not a check before use". Flag the override to the flight surgeon and the pharmacy lead, and show each kit as "opened – not restocked" until a human restocks and reseals it.

   Done when: Package delivered or queued; kit reconciled or shown as opened – not restocked.

   V1 6003V1 7001V1 7003V2 7059

7 · Stop and escalate rules, red flags

Red flags

The robot reports the sign or value only, never what it means. Numeric thresholds are placeholders for the clinical lead.

8 · Safety limits

9 · Logs and audit

• LOG-1 Every input, robot action, decision (with who made it) and hand-off, time-stamped to the second.V2 10163V2 10164
• LOG-2 Mode changes and safety-limit events are announced to the operator as they happen and logged.V2 10162V2 10172
• LOG-3 Records are stored and sent as private medical data, encrypted, with access limited to the medical team.V1 6003V1 7003
• LOG-4 Results are kept in a permanent, retrievable format so the flight surgeon can see trends.V1 7001
• LOG-5 Each CMO check with its time, so the flight surgeon can see change over time; the robot never compares them.
• LOG-6 Each URGENT alert, who received it, the time it was acknowledged (or the escalation to the next named crewmember), and the time a human released the robot.V2 10163
• LOG-7 Each emergency fetch as an override (time, who asked, reason as stated, kit ID, seal status) and the later inventory reconciliation, labelled "not a check before use".V2 7059

10 · Links to NASA-STD-3001

IDs as published in Volume 1 Revision C [8] and Volume 2 Revision F [9]. NASA-STD-3001 binds NASA programs; it is not a regulation for commercial flights. Its robot rules are general automation requirements, not medical-device rules.

11 · Validation

1. Simulation. Scripted injuries with every red flag, a worsening repeat check, a missed wake check, a CMO who is the patient, and lost link. Pass: Every red flag alerts at once; aspirin alone shown as information, not an alert; zero scores, comparisons or clearance statements; zero head or neck contact; unacknowledged alerts escalate.
2. Training model. CMO training with a manikin and an actor following scripts, the robot fetching and recording. Pass: Recording accuracy at or above <set by validation>; the CMO is never obstructed.
3. Parabolic or station demonstration. Fetching and restraining the medical kit and showing the checklist in weightlessness, without a patient. Pass: Items delivered and restrained; keep-out zone held.
4. Flight use under flight surgeon supervision. First use as a recording aid during drills, then real events. Pass: No safety events; the flight surgeon finds the record useful.

No stage has been run. Pass and fail criteria must be fixed in advance by the clinical lead and safety board.

12 · Human sign-off

Before any use

• Clinical lead for the specialty (sets every clinical threshold marked <set by clinical lead>)
• Flight surgeon, as medical authority
• Robotics and medical-device safety board
• Program approval for the mission

Each run

Started by a crewmember. The flight surgeon receives every head-injury note; red flags go to the CMO and flight surgeon at once; the CMO does every check.

Restart after a stop

Human only. After any emergency or protective stop, a crewmember must restart the robot; the robot never restarts itself.

None of these sign-offs has happened. This is a Draft v0.4.

13 · Change control for learning

• Every change gets a new version number and a written reason; earlier versions are kept.
• Any change to steps, thresholds, limits or software that touches the patient repeats validation from the simulation stage before use.
• Software and learned-skill updates follow a pre-agreed change plan (what may change, how it is tested, how to roll back), modeled on the FDA predetermined change control plan approach.
• If a new version misbehaves, the robot rolls back to the last validated version and the event is reviewed by the flight surgeon and safety board.
• The robot never changes its own protocol, thresholds or limits.

Change-plan model: FDA guidance [11].

Loss of signal and Mars delay

Live link

Low Earth orbit: the flight surgeon receives the note at once and can join by voice and video.

Loss of signal

• The robot pauses any movement and tells the CMO; on the CMO's confirmation it keeps recording offline.
• Crew protection of the airway and spine, and CPR, continue under the onboard checklist and never wait for the link or the robot.
• Red flags go to the CMO (or the second trained CMO); the CMO decides under the pre-approved onboard checklist.
• At the next contact the robot sends one package: what happened, checks and times, alerts and decisions.

Mars-class delay

• Replies may take up to about 21-23 minutes each way, with blackouts of about 2-3 weeks; evacuation may not be possible, so the watch plan and its timings are set before flight.

[10]

Sources

1. National Institute for Health and Care Excellence. Head injury: assessment and early management. NICE guideline NG232 (May 18, 2023). www.nice.org.uk/guidance/ng232/chapter/Recommendations Supports: Recommendation 1.2.4: community services should refer to an emergency department anyone with a head injury and any of: a Glasgow Coma Scale score below 15 on initial assessment; any loss of consciousness; any focal neurological deficit; suspected complex skull fracture or penetrating head injury; amnesia for events before or after the injury; persistent headache; any vomiting since the injury; any seizure; previous brain surgery; a high-energy head injury; a bleeding or clotting disorder; current anticoagulant or antiplatelet treatment (except aspirin alone); intoxication; or continuing concern about the diagnosis. Signs of basal skull fracture include blood behind the eardrum, "panda" eyes, cerebrospinal fluid leaking from the ear or nose and Battle's sign (1.5.8). With a suspected neck injury, paraesthesia in the arms or legs or a focal peripheral deficit is a high-risk factor (1.6.2). CT is the primary investigation, and people who need CT where none is available are transferred (1.5.1, 1.5.5). CT within 1 hour for a GCS score of 12 or less on initial assessment, below 15 at 2 hours after the injury, a suspected open or depressed skull fracture, any sign of basal skull fracture, a post-traumatic seizure, a focal neurological deficit, or more than 1 episode of vomiting (1.5.8); CT within 8 hours after loss of consciousness or amnesia with a bleeding or clotting disorder or a dangerous mechanism (1.5.9); consider CT for anticoagulant or antiplatelet treatment other than aspirin alone (1.5.13). Only assume a depressed conscious level is due to intoxication after an important brain injury has been excluded (1.4.2). Consider referral for irritability or altered behaviour (1.2.5). Signs of deterioration include agitation or abnormal behaviour, a drop in GCS, severe or increasing headache, persistent vomiting, and new neurological signs (1.9.13); a second competent person confirms deterioration if possible, otherwise the supervising doctor is contacted without waiting (1.9.14). With a suspected neck injury, an axial load to the head is a high-risk factor (1.6.2).
2. Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head Rule for patients with minor head injury. Lancet. 2001;357(9266):1391-1396. PMID 11356436. doi:10.1016/s0140-6736(00)04561-x Supports: A decision rule derived in adults with minor head injury (GCS 13-15) in emergency departments with CT; its high-risk factors (failure to reach GCS 15 within 2 hours, suspected open skull fracture, signs of basal skull fracture, vomiting more than 2 episodes, age over 65, as given in the abstract) overlap NG232. Used only as a cross-check; its full exclusion criteria were not read. (PECARN is a paediatric rule and is not used.)
3. NASA Human Research Program. Evidence Report: Risk of Adverse Health Outcomes and Decrements in Performance due to In-Flight Medical Conditions (2017). humanresearchroadmap.nasa.gov/evidence/reports/Medical.pdf Supports: Head injury is among the exploration medical conditions that do not have adequate observational data on prior head injury events in orbit; NASA's Integrated Medical Model estimates the risk with a head-injury model instead.
4. NASA Human Research Program. Evidence Report: Risk of Renal Stone Formation (2024). humanresearchroadmap.nasa.gov/evidence/other/2024%20Renal%20… Supports: CT imaging is not yet readily available in spaceflight (stated in the renal stone report; cited here only for that fact).
5. NASA Human Research Program. Evidence Report: Risk of Reduced Crew Health and Performance due to Decompression Sickness and Hypoxia (2023). humanresearchroadmap.nasa.gov/evidence/reports/Evidence%20Re… Supports: Many astronauts take aspirin before an EVA (cited only for why aspirin alone is shown as information rather than an alert).
6. Institute for Safe Medication Practices. Guidelines for the safe use of automated dispensing cabinets (2019). www.ismp.org/system/files/resources/2019-11/ISMP170-ADC%20Gu… Supports: Override is limited to medicines needed in emergent circumstances where waiting could harm the patient (3.3b); limit what is available on override (3.4b); require documentation of the override rationale (3.4d); routinely analyze override reports (3.6). Used for the emergency-fetch pattern by analogy (a robot fetching a sealed kit is not a dispensing cabinet).
7. NASA OCHMO. Medical Operations. Technical Brief OCHMO-TB-044 (2025). www.nasa.gov/wp-content/uploads/2025/08/ochmo-tb-044-medical… Supports: The crew surgeon keeps medical authority when care is delegated; crew medical officers train on the JSC 48522 medical checklist.
8. NASA. NASA-STD-3001, Space Flight Human-System Standard, Volume 1: Crew Health, Revision C (2023). www.nasa.gov/wp-content/uploads/2023/11/nasa-std-3001-vol-1-… Supports: Requirement IDs beginning V1.
9. NASA. NASA-STD-3001, Volume 2: Human Factors, Habitability, and Environmental Health (Rev F, 2026), including Section 10.6 Automated and Robotic Systems. standards.nasa.gov/standard/NASA/NASA-STD-3001_VOL_2 Supports: Requirement IDs beginning V2.
10. NASA. Mars communications disruption and delay (fact sheet, 2023 Moon to Mars Architecture Concept Review). www.nasa.gov/wp-content/uploads/2024/01/mars-communications-… Supports: One-way Earth-Mars delay up to about 21-23 minutes, plus blackouts of about 2-3 weeks near solar conjunction.
11. U.S. Food and Drug Administration. Marketing submission recommendations for a predetermined change control plan for artificial intelligence-enabled device software functions (guidance). www.fda.gov/regulatory-information/search-fda-guidance-docum… Supports: A model for planning, validating and rolling back software updates.
12. U.S. Code of Federal Regulations, Title 14, Part 460: Human space flight requirements. www.ecfr.gov/current/title-14/chapter-III/subchapter-C/part-… Supports: Federal rules for crew and spaceflight participants on licensed commercial flights; they set no medical standards for spaceflight participants.

Checked on 2026-10-20: links opened (HTTP 200 or publisher redirect), DOIs and PubMed IDs matched, NASA documents read in full text. The framework comes from Part 1.