Artificial Intelligence Applications in Space Medicine

INTRODUCTION:During future interplanetary space missions, a number of health conditions may arise, owing to the hostile environment of space and the myriad of stressors experienced by the crew. When managing these conditions, crews will be required to make accurate, timely clinical decisions at a high level of autonomy, as telecommunication delays and increasing distances restrict real-time support from the ground. On Earth, artificial intelligence (AI) has proven successful in healthcare, augmenting expert clinical decision-making or enhancing medical knowledge where it is lacking. Similarly, deploying AI tools in the context of a space mission could improve crew self-reliance and healthcare delivery.METHODS: We conducted a narrative review to discuss existing AI applications that could improve the prevention, recognition, evaluation, and management of the most mission-critical conditions, including psychological and mental health, acute radiation sickness, surgical emergencies, spaceflight-associated neuro-ocular syndrome, infections, and cardiovascular deconditioning.RESULTS: Some examples of the applications we identified include AI chatbots designed to prevent and mitigate psychological and mental health conditions, automated medical imaging analysis, and closed-loop systems for hemodynamic optimization. We also discuss at length gaps in current technologies, as well as the key challenges and limitations of developing and deploying AI for space medicine to inform future research and innovation. Indeed, shifts in patient cohorts, space-induced physiological changes, limited size and breadth of space biomedical datasets, and changes in disease characteristics may render the models invalid when transferred from ground settings into space.Cheung HC, De Louche C, Komorowski M. Artificial intelligence applications in space medicine. Aerosp Med Hum Perform. 2023; 94(8):610-622.

Anesthetic Management Using Multiple Closed-loop Systems and Delayed Neurocognitive Recovery: A Randomized Controlled Trial

BACKGROUND

Cognitive changes after anesthesia and surgery represent a significant public health concern. We tested the hypothesis that, in patients 60 yr or older scheduled for noncardiac surgery, automated management of anesthetic depth, cardiac blood flow, and protective lung ventilation using three independent controllers would outperform manual control of these variables. Additionally, as a result of the improved management, patients in the automated group would experience less postoperative neurocognitive impairment compared to patients having standard, manually adjusted anesthesia.

METHODS

In this single-center, patient-and-evaluator-blinded, two-arm, parallel, randomized controlled, superiority study, 90 patients having noncardiac surgery under general anesthesia were randomly assigned to one of two groups. In the control group, anesthesia management was performed manually while in the closed-loop group, the titration of anesthesia, analgesia, fluids, and ventilation was performed by three independent controllers. The primary outcome was a change in a cognition score (the 30-item Montreal Cognitive Assessment) from preoperative values to those measures 1 week postsurgery. Secondary outcomes included a battery of neurocognitive tests completed at both 1 week and 3 months postsurgery as well as 30-day postsurgical outcomes.

RESULTS

Forty-three controls and 44 closed-loop patients were assessed for the primary outcome. There was a difference in the cognition score compared to baseline in the control group versus the closed-loop group 1 week postsurgery (-1 [-2 to 0] vs. 0 [-1 to 1]; difference 1 [95% CI, 0 to 3], P = 0.033). Patients in the closed-loop group spent less time during surgery with a Bispectral Index less than 40, had less end-tidal hypocapnia, and had a lower fluid balance compared to the control group.

CONCLUSIONS

Automated anesthetic management using the combination of three controllers outperforms manual control and may have an impact on delayed neurocognitive recovery. However, given the study design, it is not possible to determine the relative contribution of each controller on the cognition score.

Closed-loop administration of analgesic drugs based on heart rate variability analysis

General anesthesia is based on the use of hypnotic, muscle relaxant and analgesic drugs in order to render the patient unresponsive to the surgical procedure. The difficulty for anesthesiologists is then to determinate the minimum efficient dose to avoid any risk of under or over dosing. For several years, monitoring systems were developed in order to measure depth of hypnosis, myorelaxation and analgesia. As soon as all these monitoring systems became available, several teams worked on the closed-loop administration of anesthetic agents. We have developed a closed-loop system allowing the automatic administration of analgesic drugs. This system is based on the analysis of a heart rate variability based index: the ANI (Analgesia Nociception Index). In this paper, we describe this device and demonstrate its efficiency, repeatability and safety in a simulation environment.

Pilot study of closed-loop anaesthesia for liver transplantation

BACKGROUND

Automated titration of propofol and remifentanil guided by the bispectral index (BIS) has been used for numerous surgical procedures. Orthotopic liver transplantation (OLT) uniquely combines major changes in circulating volume, an anhepatic phase, and ischaemia-reperfusion syndrome. We assessed the behaviour of this automated controller during OLT.

METHODS

Adult patients undergoing OLT were included in this pilot study. Consumption of propofol and remifentanil was calculated for each surgery period (dissection, anhepatic, and liver reperfusion phases). Arterial blood samples were collected at several time points to allow comparison of actual with calculated propofol and remifentanil concentrations. Data are presented as median [25th and 75th percentiles] or percentage (95% confidence interval).

RESULTS

Thirteen patients were studied. System performance, defined as the percentage of time with BIS in the range 40-60, was 88% (86-94) of the total duration of anaesthesia. Propofol requirement was decreased during the anhepatic phase compared with the dissection phase (2.9 [1.9-5.0] mg kg(-1) h(-1) and 4.6 [3.5-8.1] mg kg(-1) h(-1); P<0.03) while remifentanil consumption was unchanged (0.11 [0.09-0.19] µg kg- (1) min(-1)). Bland-Altman analysis showed a weak concordance for propofol (bias of 0.7 µg ml(-1) and limits of agreement of -2.2 to +3.7 µg ml(-1)) and remifentanil (bias of 1.3 ng ml(-1) and limits of agreement -4.3 to +6.8 ng ml(-1)). No adverse events were reported during anaesthesia.

CONCLUSIONS

This pilot study indicates that automated titration of propofol and remifentanil guided by the BIS is feasible during OLT.

Bispectral index to guide induction of anesthesia: a randomized controlled study

Background

It is unknown to what extent hypotension frequently observed following administration of propofol for induction of general anesthesia is caused by overdosing propofol. Unlike clinical signs, electroencephalon-based cerebral monitoring allows to detect and quantify an overdose of hypnotics. Therefore, we tested whether the use of an electroencephalon-based cerebral monitoring will cause less hypotension following induction with propofol.

Methods

Subjects were randomly assigned to a bispectral index (BIS)-guided (target range 40–60) or to a weight-related (2 mg.kg^− 1) manual administration of propofol for induction of general anesthesia. The primary endpoint was the incidence of hypotension following the administration of propofol. Secondary endpoints included the degree of hypotension and correlations between BIS and drop in mean arterial pressure (MAP). Incidences were analyzed with Fisher’s Exact-test.

Results

Of the 240 patients enrolled into this study, 235 predominantly non-geriatric (median 48 years, 25th – 75th percentile 35–61 years) patients without severe concomitant disease (88% American Society of Anesthesiology physical status 1–2) undergoing ear, nose and throat surgery, ophthalmic surgery, and dermatologic surgery were analyzed. Patients who were manually administered propofol guided by BIS (n = 120) compared to those who were given propofol by weight (n = 115) did not differ concerning the incidence of hypotension (44% vs. 45%; p = 0.87). Study groups were also similar regarding the maximal drop in MAP compared to baseline (33% vs. 30%) and the proportion of hypotensive events related to all measurements (17% vs. 19%). Final propofol induction doses in BIS group and NON-BIS group were similar (1.93 mg/kg vs. 2 mg/kg). There was no linear correlation between BIS and the drop in MAP at all times (r < 0.2 for all) except for a weak one at 6 min (r = 0.221).

Conclusion

Results of our study suggest that a BIS-guided compared to a weight-adjusted manual administration of propofol for induction of general anesthesia in non-geriatric patients will not lower the incidence and degree of arterial hypotension.

Trial registration

German Registry of Clinical Trials (DRKS00010544), retrospectively registered on August 4, 2016.

Electronic supplementary material

The online version of this article (10.1186/s12871-018-0522-8) contains supplementary material, which is available to authorized users.

Model-based drug administration: current status of target-controlled infusion and closed-loop control

PURPOSE OF REVIEW

Drug administration might be optimized by incorporating pharmacokinetic-dynamic (PK/PD) principles and control engineering theories. This review gives an update of the actual status of target-controlled infusion (TCI) and closed-loop computer-controlled drug administration and the ongoing research in the field.

RECENT FINDINGS

TCI is becoming mature technology clinically used in many countries nowadays with proven safety. Nevertheless, changing populations might require adapting the established PK/PD models. As TCI requires accurate PK/PD models, new models have been developed which should now be incorporated into the pumps to allow more general use of this technology. Closed-loop administration of hypnotic drugs using an electro-encephalographic-derived-controlled variable has been well studied and has been shown to outperform manual administration. Computer administration for other drugs and fluids have been studied recently. Feasibility has been shown for systems controlling multiple components of anaesthesia, but more work is required to show clinical safety and efficiency.

SUMMARY

Evidence in the literature is increasing that TCI and closed-loop technology could assist the anaesthetists to optimize drug administration during anaesthesia.

Nociceptive stimulation during Macintosh direct laryngoscopy compared with McGrath Mac videolaryngoscopy: A randomized trial using indirect evaluation using an automated administration of propofol and remifentanil

BACKGROUND

Decrease of the nociceptive stimulation induced by laryngoscopy could be an advantage for patients without risk of difficult intubation. The present study aimed to compare the difference in nociceptive stimulation between the use of a conventional laryngoscope or of a videolaryngoscope. Amount of nociception was assessed indirectly using the peak remifentanil concentration determined by a closed-loop administration of propofol and remifentanil with bispectral index (BIS) as the input signal (target 50).

METHODS

A prospective single-center randomized study was performed including surgical patients without predictable risk of difficult mask ventilation or of difficult tracheal intubation. Forty consecutive surgery patients were randomly assigned to CL group (conventional laryngoscope) or VL group (McGrath Mac videolaryngoscope). Induction of anesthesia was performed automatically using the closed-loop system and myorelaxation with atracurium. The allocation was revealed just before tracheal intubation. The primary outcome was the peak plasma remifentanil concentration observed during the 5-minute period which followed intubation.

RESULTS

Sixteen patients in the CL group and 11 in the VL group were analyzed. Plasmatic remifentanil and propofol concentrations were similar in both groups either before tracheal intubation or during the 5 minutes following intubation. There was a nonsignificant between-group difference (P = .09) for the peak concentration of remifentanil. A comparable result was observed for other outcomes except for the heart rate which increased in the CL group.

CONCLUSION

Use of the videolaryngoscope McGrath Mac did not reduce the nociceptive stimulation induced during intubation as evaluated by the automatically administered remifentanil concentration.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT02245789.

Closed-loop Administration of General Anaesthesia: From Sensor to Medical Device

Closed-loop administration devices for general anaesthesia have become a common subject of clinical research over the last decade and appear more and more acceptable in clinical practice. They encompass various therapeutic needs of the anesthetized patient, e. g. fluid administration, hypnotic and analgesic drug administration, myorelaxation. Multiple clinical trials involving closed-loop devices have underscored their safety, but data concerning their clinical benefit to the patient are still lacking. As the marketing of various devices increases, clinicians need to understand how comparisons between these devices can be made: the measure of performance error and wobble are technical but have also a clinical meaning, to which clinical outcomes can be added, such as drug consumption and maintenance of hemodynamic parameters (e. g. heart rate and blood pressure) within predefined ranges. Clinicians using closed-loop devices need especially to understand how various physiological signals lead to specific drug adaptations, which means that they switch from decision making to supervision of general anaesthesia.

Fully Automated Anesthesia and Fluid Management Using Multiple Physiologic Closed-Loop Systems in a Patient Undergoing High-Risk Surgery

Automated delivery of anesthesia guided by processed electroencephalogram monitoring using a closed-loop system is no longer a novel concept. However, combining multiple independent physiologic closed-loop systems together has never been documented before. The purpose of this case report was to evaluate the feasibility of automated anesthesia and fluid management based on a combination of physiological variables (bispectral index, stroke volume, and stroke volume variations) using 2 independent closed-loop systems.

Hypnosis closed loop TCI systems in outpatient surgery

OBJECTIVE

Determine the influence of general anaesthesia with closed-loop systems in the results of outpatient varicose vein surgery.

PATIENTS AND METHODS

Retrospective observational study including data from 270 outpatients between 2014 and 2015. The patients were divided into 2 groups according to the type of general anaesthesia used. The CL Group included patients who received propofol in closed-loop guided by BIS and remifentanil using TCI, and the C Group received non-closed-loop anaesthesia. Age, sex, surgical time, discharge time and failure of outpatient surgery were recorded. Quantitative data were checked for normal distribution by the method of Kolmogorov-Smirnov-Lilliefors. Differences between groups were analysed by a Student-t-test or Mann-Whitney-Wilcoxon test, depending on their distribution. Categorical data were analysed by a Chi-squared test. We used Kaplan-Meier estimator and the effect size (calculated by Cohen's d) to study the discharge time. Statistical analysis was performed using R 3.2.3 binary for Mac OS X 10.9.

RESULTS

There were no significant differences in age, sex and surgical time and failure of outpatient surgery. Discharge time was different in both groups: 200 (100) vs. 180 (82.5) minutes, C Group and CL Group, respectively (data are median and interquartile rank); P=.005.

CONCLUSION

The use of closed-loop devices for the hypnotic component of anaesthesia hastens discharge time. However, for this effect to be clinically significant, some improvements still need to be made in our outpatient surgery units.

Sedation options for the morbidly obese intensive care unit patient: a concise survey and an agenda for development

Background

We offer some perspectives and commentary on the sedation of obese patients in the intensive care unit (ICU).

Discussion

Sedation in morbidly obese patients should conform to the same broad principles now current in ICU practice. These include a general presumption against benzodiazepines as first-line agents. Opioids should be avoided in any situation where spontaneous breathing is required. Remifentanil is the preferred agent where continuous stable opioid levels using an infusion are required, because of its lack of context-sensitive accumulation. Volatile anaesthetics may be an option for the same reason but there are no substantial, controlled demonstrations of effectiveness/safety in short-term use in the ICU setting. Propofol is a valuable resource in the morbidly obese patients but the duration of continuous sedation should not exceed 6 days, in order to avoid propofol infusion syndrome. Alpha-2 agonists offer a range of theoretically positive features for the sedation of morbidly obese patients, but at present there is a lack of pharmacokinetic data and a critical mass of high-grade clinical data. Dexmedetomidine has the attraction of not causing respiratory depression or obstructive breathing during sedation and its sympatholytic effects should help deliver stable blood pressure and heart rate. Ketamine has a poor tolerability profile in adults so its use in the ICU context is largely confined to paediatrics.

Conclusion

None of the agents currently available is ideal for every situation encountered in the management of morbidly obese patients. This article identifies additional research needed to place sedation practice of obese patients on a more systematic footing.