ISSN: 2320-5407 Int. J. Adv. Res.

12(09), 1687-1697

Journal Homepage: -www.journalijar.com

Article DOI:10.21474/IJAR01/19608
DOI URL: http://dx.doi.org/10.21474/IJAR01/19608

RESEARCH ARTICLE
EVOLUTION OF ARTIFICIAL INTELLIGENCE IN ANESTHESIA AND ITS PATH FORWARD

Sameer Kapoor, Arushi Arvind, Ghassan Kloub, Shrutika Mahendra Parekh, Susan Khalafallah and
Hekmat Hammam
……………………………………………………………………………………………………....
Manuscript Info Abstract
……………………. ………………………………………………………………
Manuscript History Artificial intelligence (AI) has brought a paradigm shift in patient
Received: 31 July 2024 management in the modern healthcare industry. This narrative literature
Final Accepted: 31 August 2024 review aims to explore the evolution of AI in the field of anesthesia and
Published: September 2024 discuss its potential path forward. The increased demand for
personalized care, smart decision-making tools, precision in practice,
Key words:-
Anesthesia, Artificial Intelligence, and growth in telemedicine, particularly in the realm of perioperative
Machine Learning, Deep Learning, care, has brought AI to the forefront. By examining the existing
Perioperative Applications, Clinical research and applications of AI in anesthesia, this review seeks to
Decision Support Systems, Ethics
provide insights into the current state of AI technology, its applications,
implications for practicing clinicians, and the challenges and limitations
that need to be addressed for its successful integration into
perioperative care. This review encourages interdisciplinary teamwork
between anesthesiologists, technologists, engineers, and technicians to
simplify the complexities of AI implementation, address ethical issues,
mitigate risks, and ensure safe perioperative care.

Copyright, IJAR, 2024,. All rights reserved.

……………………………………………………………………………………………………....
Introduction:-
Artificial intelligence (AI) has emerged as a transformative technology with the potential to revolutionize various
industries, including healthcare [1]. In the field of anesthesia, AI holds great promise in enhancing patient care,
improving clinical decision-making, and optimizing workflow efficiency [2]. This review aims to provide a
comprehensive overview of the principles and evolution of AI in anesthesia, its multifaceted roles involving novel
approaches to patient care, recent guidelines on its use, and shed light on its perioperative applications.

Principles of artificial intelligence:

AI is a rapidly evolving technology to research and develop theories, methods, and clinical applications for
simulating and extending human intelligence to carry out tasks via algorithms [2]. AI technologies include machine
learning (ML), deep learning (DL), natural language processing (NLP), speech and image recognition, and expert
systems including robotics. ML is a subset of AI that works using sample data to train computer programs to
recognize patterns based on algorithms. DL refers to the core of ML, incorporating computer systems designed to
imitate neurons in the brain. [Figure 1] illustrates the relationship of DL as compared to ML and AI [3].

The term “deep” in the DL methodology refers to the concept of multiple levels through which data is processed to
build a data-driven model. These multi-stage processes involve multi-layer neural networks, referred to as Artificial
neural networks (ANNs). However, unlike ML modeling, feature extraction in DL is huge and automated rather than
manual. Large amounts of data can be processed using DL modeling, which enhances its performance significantly
and exponentially [Figure 2].

1687
Corresponding Author:-Sameer Kapoor
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

Figure 1: An approach to illustrate the relationship between Artificial Intelligence (AI), Machine Learning (ML),
and Deep Learning (DL).

Figure 2:- Comparison of performance of Deep Learning and traditional Machine Learning models considering the
amount of data.

1688
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

A DL model typically follows three processing stages, such as data understanding and preprocessing, model
building and training, and validation and interpretation. DL accomplishes the task using convolutional neural
networks (CNNs), recurrent neural networks (RNNs), autoencoder (AE), and deep belief networks (DBN)
mimicking the complex neural structures of the human brain [Figure 3]. CNNs utilize filters to extract data, followed
by correlating features in different layers by utilizing multiple convolutions and pooling layers. CNNs are
specifically meant to deal with a variety of 2D shapes and are thus widely employed in image processing, computer
vision, and pattern recognition. The RNNs are specialized ANNs where connections between nodes can create a
cycle, as the output relies on prior elements in the sequence, allowing it to exhibit temporal dynamic behavior.
RNNs tend to have „memory‟ that allows them to process the task using information from previous inputs. Machine
translation, robot control, speech recognition, and speech synthesis are some of the applications of RNNs [3].

Figure 3:- A typical Deep Learning workflow model. Source (with permission). Deep learning: A Comprehensive
Overview on Techniques, Taxonomy, Applications and Research Directions. SN Comput Sci. 2021;2(6):420.

Evolution and current applications of AI in Anesthesia:

The early attempts at automation in anesthesia focused on the development of systems for monitoring and control.
However, recent advancements in AI, particularly in ML and DL algorithms, have opened up new possibilities for
the integration of AI into anesthesia practice [Figure 4]. AI-based systems are being developed relentlessly for
various applications of perioperative medicine to tailor anesthesia to individual patient needs, minimize risks, and
„enhance recovery after surgery‟ (ERAS) [4].

The journey of a surgical patient during the perioperative period is carved here, using the two dynamic, yet
intersecting independent fields - anesthesia and AI.

Pre-anesthesia check-up (PAC)

The preoperative phase is undoubtedly the quintessential aspect of perioperative care. The patient's safety begins
with the complete medical history and examination by an anesthesiologist. NLP, one of the applications of AI, can
be used to extract relevant information from unstructured massive text data. A retrospective study depicted the
ability of NLP to extract meaningful data from unstructured electronic medical records (EMR) and generate an
automated pre-anesthetic evaluation report. The data produced by NLP was compared with the information collected
by an anesthesiologist. NLP could pick up relevant conditions missed by anesthesia professionals in 16.57% of the
cases and missed important data procured by anesthetists in only 2.19% of the cases [5].

Assessment of the airway is a routine step in PAC, accomplished by history and physical examination of the patient.
AI models enable anesthesiologists to guide difficult airways objectively, using predictive features like body mass
index, thyroid distance, computerized facial analysis, and photographs [6,7].

For specific patients, preoperative assessment can be extremely challenging even for an expert in view of
identifying risk factors, patient-specific prediction, explaining the risks involved, and designing the real informed

1689
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

consent. With the application of ML algorithms, the AI models process multifaceted data, multiple patterns, and
relationships to generate optimized and individualized approaches to patient care.

Figure 4:- Healthcare solutions powered by artificial intelligence to bridge Deep Learning techniques and
perioperative care.

Perioperative risk stratification

The counseling of a presurgical patient and obtaining valid well-informed consent requires proper risk stratification.
ML methods have been quite successful in grading the risk of patients undergoing high-risk surgeries.
Mysurgeryrisk is an ML algorithm that identifies the risk scores for eight perioperative complications (acute kidney
injury, sepsis, venous thromboembolism, intensive care unit [ICU] admission for more than 48 hours, mechanical
ventilation for more than 48 hours, neurologic, and cardiovascular events) and predicts the mortality risk up to 2
years postoperatively [8].

The American Society of Anesthesiologist's Physical Status (ASA-PS) classification has been used by professionals
for more than half a century to explain risk levels to preoperative patients.

ML algorithms have been used to help classify the risk involved in complicated scenarios [9]. Further, Machine
Learning Physical Status (ML-PS) is developed to stratify high-risk categories of preoperative patients, independent
of the clinician‟s decision [10]. The professionals utilize the ML methods to predict postoperative postsurgical
mortality and ICU admission, from data, such as patient demographics, medical history, surgical procedures,
medications, vital signs, laboratory values, and other clinical factors [11].

Perioperative optimization
The proactive approach of the optimization of a surgical patient has played a significant role in successful outcomes
and reduced mortality. AI-driven applications develop tailor-made instructions for a specific patient undergoing a
unique kind of surgery, that includes the patient's nutrition, physiotherapy, incentive spirometry, medication
adjustments, glucose control, and prophylaxis of deep venous thrombosis to optimize clinical scenarios and mitigate
the risks [12]. This personalized and adaptive approach to perioperative care has brought a paradigm shift in the
preparation of patients for surgery.

Clinical Decision Support Systems (CDSS)

CDSS refers to a computer-based program that assists anesthesia professionals in clinical decision-making. This
methodology assimilates the plethora of data and enables the anesthesiologists to devise patient-specific anesthesia

1690
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

care plans. The care provider uses system-generated advice to make necessary interventions, such as drug dosing,
fluid management, ventilatory parameters, and blood transfusion, after analyzing the patient‟s comorbidities,
allergies, and laboratory values [13].

Recently, the application of Anesthesia information management systems (AIMS) has enabled real-time CDSS to
provide immediate feedback and advice, in case of any deviation from best medical practices. AIMS was initially
introduced for automatic recording of data on a patient's monitor and anesthesia machine parameters for
intraoperative documentation. The automated record keeping is analyzed in real-time during surgery involving
rapidly changing physiological states of the patient. CDSS, along with AIMS and EMR, has emerged as a beacon of
innovation with a promising approach to detecting clinical issues, and alerting anesthesiologists to take necessary
steps that ensure patient safety [14].

Depth of Anesthesia (DoA) monitoring

In addition to basic monitoring advised by ASA, it is prudent to always monitor the depth of anesthesia. The
incidence of postoperative mortality and adverse events increases with „too deep‟ anesthesia [15]. On the contrary,
„too shallow‟ anesthesia is associated with a high incidence of intraoperative awareness. The widely used bispectral
index (BIS), based on analysis of EEG signals provides a crude reading and it lacks the correlation with nociceptive
stimulation.

AI promises the DoA monitoring system, based on real-time EEG and DL algorithms, that has the capacity to ensure
the ideal anesthetic depth, where the risk of recall is as low as possible, and where the blood pressure and heart rate
of the patient are kept optimal for an individual patient. CNN and RNN applications ensure the near-ideal, accurate
prediction of DoA within 20ms, with a performance index significantly better than BIS [16].

Automatic drug administration systems

The synergism between AI and medicine has brought a dramatic move towards automation in the field of anesthesia.
The closed-loop drug delivery system, especially target-controlled infusion (TCI) is a well-practiced technique. The
benefits of volatile anesthesia include a better environmental profile, rapid cognitive recovery, ERAS, and lesser
cost. These machines use ML algorithms to aid automatic drug delivery, and hence, are referred to as
pharmacological robots [17].

An ideal automatic drug delivery system, powered by AI shall allow the sedation without using a mechanical
airway, as well as, too profound loss of consciousness in a closed-loop general anesthesia. The close loop control
technology of McSleepy offers management of all stages of anesthesia, hypnosis, analgesia, and muscle relaxation
[18].

Operating Room Logistics

By leveraging ML algorithms and predictive analytics, AI systems can analyze vast amounts of data to identify
patterns, trends, and potential bottlenecks in the operating room workflow. This can lead to better resource
allocation, improved patient outcomes, and enhanced overall efficiency [19]. One of the most compelling benefits of
AI in operating room logistics and anesthesia management is its ability to optimize staff allocations. AI algorithms
can analyze patient data, such as medical histories, preoperative assessments, and surgical complexity, to predict the
optimal team composition for each case. This can result in more efficient use of resources and reduce the risk of
over- or under-staffing. For instance, an AI system could recommend that a certain surgical team be assigned to a
high-risk surgery, ensuring that the necessary expertise and experience are available to minimize complications [20].

Postoperative monitoring and follow-up

The postoperative period is equally crucial in the journey of a surgical patient to be uneventful.

AI technologies can guide the clinician to make a decision to discharge a vulnerable patient from the recovery room,
after analysis of multiple factors simultaneously. The accuracy of predictive models is directly proportional to the
amount of data added to the system during the perioperative period. AI techniques identify patterns, and anticipate
issues after developing personalized risk profiles, thus preventing postoperative complications significantly [21].

The effective control of pain in the postoperative period reduces complications, decreases the incidence of chronic
pain, and increases patient satisfaction. Effective pain assessment at regular intervals and patient-controlled

1691
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

analgesia (PCA) in the postoperative period are keys to success. However, the current PCA is not intelligent enough
as information feedback is lacking, and patients suffer from inadequate analgesia, frequent clinical complaints, and a
high call-back rate [22].

To increase the efficiency of Acute Pain services (APS), Ai-PCA is introduced, which integrates the Internet Of
Things (IoT) technology and AI operation. Ai-PCA sends warning alarms of „poor analgesia‟ to APS doctors by
analyzing the repeated pressing by the patient. APS team members are alerted timely to modify infusion rates of Ai-
PCA, to avoid the occurrence of severe pain. Similarly, a „blocking alarm‟ is meant for decrease in infusion rates, in
case a patient has nausea, vomiting, itching, or dizziness, to avoid life-threatening complications. The real-time
alarm system of Ai-PCA uses a central analgesia monitoring platform, which may be reviewed by APS team
members remotely and at repeated intervals [23].

Tele-anesthesia
Automated anesthesia delivery systems can be controlled remotely and have the potential to provide anesthesia at a
distance, which is referred to as „tele-anesthesia‟. The essential components of tele-anesthesia include continuous
audio-video communication with local healthcare providers, a continuous video feed of important monitoring
systems, and remote control of a local anesthesia system. It is pertinent to have complete preoperative check-up of
the patient using audio-visual communication. Transcontinental anesthesia could help to overcome the shortage of
specialists in remote areas, potentially reduce travel costs, and improve patients‟ accessibility to professional
treatments[24].

Streamlining the anesthesia process remotely, the Kepler Intubation System (KIS) is a dependable technique. The
system consists of a joystick, a computer as the „electronic brain,‟ a carbon fiber robotic arm, and a bespoke piece
that allows attachment of the video laryngoscope.

These are referred to as mechanical robots. Complete robotic anesthesia seems feasible using the combination of
pharmacological (McSleepy) and mechanical robots (KIS) [17]. An automated airway intubation robot, REALITI is
developed on the principle of real-time image recognition function. AI-based active visual tracking technology
appreciates the first anatomical sign, switches itself from manual to automatic mode, and the endoscope moves
towards the glottis. Once the glottis is entered, the structural image of the tracheal rings is ready for manual
confirmation [25].

“Quintuple Aim” and AI

The Institute for Healthcare Improvement introduced the idea of Improved patient experience, better population
health, and lower costs. Later, the clinician's well-being was included as a bundle to address the burnout of
professionals. To deliver high-quality, biomedical ethics, and safe patient care for populations, „Health Equity‟ was
proposed as the fifth aim recently [26] [Figure. 5].

Figure 5:- Evolution of the Quintuple Aim in healthcare delivery.

1692
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

AI can play a crucial role in perioperative patient safety through the lens of the Quintuple Aim. The anesthesia
professional needs to adopt the applications of AI in the perioperative management of surgical patients, within the
framework of the Quintuple Aim to improve safety and quality [Figure 6]. The application of AI tools shall help the
anesthesiologist rapidly navigate data from disparate sources, and by actively guiding the individual in making
better, scientific, and evidence-based medical decisions within the intricate healthcare system [27].

Figure 6:- Framework applying Quintuple Aim in applications of artificial intelligence aiding patient safety across
the perioperative continuum. Source (with permission). Tan JM, Cannesson MP. Artificial intelligence, patient
safety, and achieving quintuple aim in anesthesiology. APSF newsletter 2023;38(1);1-4.

Implications for Practicing Clinicians:

It is crucial for practicing clinicians to have a solid understanding of the basics of AI to effectively utilize and
interpret the outputs of AI systems. The integration of AI in clinical practice is a complex process that requires
careful consideration and planning. Clinicians will need to assess the feasibility of incorporating AI technologies
into their daily workflows, taking into account factors such as software compatibility, user-friendliness, and data
security [28].

As AI systems become more prevalent in healthcare, clinicians have the responsibility to ensure that patient data is
not just entered correctly but also remains safe and inaccessible to anyone outside the concerned healthcare team.
Clinicians must also cross-check all functionings, results, monitoring, and outputs themselves and avoid blind
reliance.

Challenges, limitations, and path forward:

While the potential benefits of AI in anesthesia are significant, there are several challenges and limitations that need
to be addressed. As an anesthesiologist, incorporating AI into clinical practice presents both strengths and
weaknesses, even threats.

1693
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

One major challenge is the lack of data standardization and interoperability within healthcare systems. In order for
AI algorithms to effectively analyze and interpret data from EMR, imaging studies, and monitoring devices,
consistent data formats and protocols need to be established across different platforms [29].

Fixing the limitations in the accuracy and reliability of AI algorithms in clinical decision-making is a major
consideration. To date, there are a limited number of DL programs with external validation in medicine. The
incomplete, biased, or incorrect data, fed to ML shall produce false results without an alarm, leading clinicians to
make irrational decisions [16]. One of the primary ethical concerns with using AI in anesthesiology is the issue of
accountability. When errors occur in anesthesia administration, who is ultimately responsible – the human
anesthesiologist or the AI system? This question becomes even more complex when considering that AI systems are
programmed by humans and can potentially contain biases or errors themselves. Anesthesiologists must not have
overconfidence in the recommendations provided by AI systems, as any errors or misinterpretations could have
serious consequences for patient safety [30].

Furthermore, there are ethical considerations surrounding the use of AI in anesthesia. Anesthesiologists must
grapple with questions of autonomy, accountability, and transparency when relying on AI systems to assist in
clinical decision-making. In order to hold algorithms accountable, the internal bias should be addressed. However, it
may become impossible as ML cannot answer how the results are generated, referred to as the “black box”
phenomenon [31].

Additionally, there may be concerns about data privacy and security when sensitive patient information is stored and
processed by AI algorithms. Unethical hacking is another massive hurdle in data processing with artificial
intelligence that can create huge data leaks and breaches which results in multidimensional catastrophes for both
patients and healthcare providers [30].

In response to the growing need to responsibly manage the rapid rise of AI health technologies, the World Health
Organization (WHO) outlines the following six areas for regulation of AI for health [32].
1. To foster trust, the importance of transparency and documentation, such as through documenting the entire
product lifecycle and tracking development processes is stressed.
2. For risk management, the issues like „intended use‟, „continuous learning‟, human interventions, training
models, and cybersecurity threats, must all be comprehensively addressed, with models made as simple as
possible.
3. Externally validating data and being clear about the intended use of AI helps assure safety and facilitate
regulation.
4. A commitment to data quality, such as through rigorously evaluating systems pre-release, is vital to ensure that
the systems do not amplify biases and errors.
5. The challenges posed by important, complex regulations are addressed with an emphasis on understanding the
scope of jurisdiction and consent requirements, in service of privacy and data protection.
6. Fostering collaboration between regulatory bodies, patients, healthcare professionals, industry representatives,
and government partners, can help ensure products and services stay compliant with regulation throughout their
life cycles.

To harness the full potential of AI in anesthesia, collaboration between clinicians, researchers, and industry experts
is essential. Clinicians should actively engage in the development and implementation of AI technologies, ensuring
that they align with the needs and priorities of clinical practice. Continued research and innovation in AI algorithms,
data collection, and integration with existing clinical systems are crucial for the successful integration of AI into
anesthesia practice. Furthermore, ongoing education and training programs should be established to equip clinicians
with the necessary skills to effectively utilize AI technologies [Figure 7].

1694
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

Figure 7:- SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis of artificial intelligence in anesthesia.

Conclusion:-
The evolution of AI in anesthesia holds immense potential for improving patient care, enhancing clinical decision-
making, and optimizing workflow efficiency. While there are challenges and limitations that need to be addressed,
the path forward involves collaboration, research, and education. By embracing AI technology and actively
participating in its development, practicing clinicians can shape the future of AI in anesthesia and ensure its
successful integration into clinical practice.

References:-
1. Stanfill MH, Marc DT. Health Information Management: Implications of Artificial Intelligence on Healthcare
Data and Information Management. Yearb Med Inform. 2019 Aug;28(1):56-64. doi: 10.1055/s-0039-1677913.
Epub 2019 Aug 16. PMID: 31419816; PMCID: PMC6697524.
2. Bellini V, Valente M, Gaddi AV, Pelosi P, Bignami E. Artificial intelligence and telemedicine in anesthesia:
potential and problems. Minerva Anestesiologica. 2022 Sep;88(9):729-734. DOI: 10.23736/s0375-
9393.21.16241-8. PMID: 35164492.
3. Sarker IH. Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research
Directions. SN Comput Sci. 2021;2(6):420. doi: 10.1007/s42979-021-00815-1. Epub 2021 Aug 18. PMID:
34426802; PMCID: PMC8372231.
4. Zain Z, Almadhoun MKIK, Alsadoun L, Bokhari SFH. Leveraging Artificial Intelligence and Machine
Learning to Optimize Enhanced Recovery After Surgery (ERAS) Protocols. Cureus. 2024 Mar
21;16(3):e56668. doi: 10.7759/cureus.56668. PMID: 38646209; PMCID: PMC11032416.
5. Suh HS, Tully JL, Meineke MN, et al. Identification of preanesthetic history elements by a natural language
processing engine [published online ahead of print, 2022 Jul 15]. Anesth Analg. 2022 Dec 1;135:1162–1171.
PMID: 35841317
6. Moustafa MA, El-Metainy S, Mahar K, et al. Defining difficult laryngoscopy findings by using multiple
parameters: A machine learning approach. Egypt J Anaesth. 2017;33:153-58
7. Cuendet GL, Schoettker P, Yüce A, et al. Facial image analysis for fully automatic prediction of difficult
endotracheal intubation. IEEE Trans Biomed Eng. 2016;63:328-39
8. Bihorac A, Ozrazgat-Baslanti T, Ebadi A, Motaei A, Madkour M, Pardalos PM, et al. MySurgeryRisk:
Development and Validation of a Machine-learning Risk Algorithm for Major Complications and Death After
Surgery. Ann Surg 2019;269:652–62.
9. Böhmer A, Defosse J, Geldner G, Rossaint R, Zacharowski K, Zwissler B et al: The updated ASA
classification. Anästh Intensivmed 2021;62:223–227. DOI: 10.19224/ai2021.223

1695
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

10. Wongtangman K, Aasman B, Garg S, Witt AS, Harandi AA, Azimaraghi O, Mirhaji P, Soby S, Anand P,
Himes CP, Smith RV, Santer P, Freda J, Eikermann M, Ramaswamy P. Development and validation of a
machine learning ASA-score to identify candidates for comprehensive preoperative screening and risk
stratification. J Clin Anesth. 2023 Aug;87:111103. Doi: 10.1016/j.jclinane.2023.111103. Epub 2023 Mar 8.
PMID: 36898279.
11. Chiew, C.J.; Liu, N.; Wong, T.H.; Sim, Y.E.; Abdullah, H.R. Utilizing Machine Learning Methods for
Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission. Ann. Surg. 2020, 272,
1133–1139. [CrossRef] [PubMed]
12. Solanki SL, Pandrowala S, Nayak A, Bhandare M, Ambulkar RP, Shrikhande SV. Artificial intelligence in
perioperative management of major gastrointestinal surgeries. World J Gastroenterol 2021; 27(21): 2758-2770
[PMID: 34135552 DOI: 10.3748/wjg.v27.i21.2758]
13. Cascella, M.; Tracey, M.C.; Petrucci, E.; Bignami, E.G. Exploring Artificial Intelligence in Anesthesia: A
Primer on Ethics, and Clinical Applications. Surgeries 2023, 4, 264–274. https://doi.org/10.3390/
surgeries4020027
14. Nair, Bala G. PhD*; Gabel, Eilon MD†; Hofer, Ira MD†; Schwid, Howard A. MD‡; Cannesson, Maxime MD,
PhD†. Intraoperative Clinical Decision Support for Anesthesia: A Narrative Review of Available Systems.
Anesthesia & Analgesia 124(2):p 603-617, February 2017. | DOI: 10.1213/ANE.0000000000001636
15. Anaesthetic depth and complications after major surgery: an international, randomized controlled trial Short,
Timothy GShulman, M et al. The Lancet, Volume 394, Issue 10212, 1907 - 1914
16. Song B. et al: Necessity and importance of developing AI in anesthesia © Med Sci Monit, 2023; 29: e938835.
DOI: 10.12659/MSM.938835
17. Hemmerling, Thomas M. MD, MSc, DEAA*,†; Jeffries, Sean D. MSc*. Robotic Anesthesia: A Vision for
2050. Anesthesia & Analgesia 138(2):p 239-251, February 2024. | DOI: 10.1213/ANE.0000000000006835
18. Zaouter C, Hemmerling TM, Lanchon R, et al. The feasibility of a completely automated total IV anesthesia
drug delivery system for cardiac surgery. Anesth Analg. 2016;123:885–893.
19. Bellini, V., Russo, M., Domenichetti, T. et al. Artificial Intelligence in Operating Room Management. J Med
Syst 48, 19 (2024). https://doi.org/10.1007/s10916-024-02038-2
20. Daniel A. Hashimoto, Elan Witkowski, Lei Gao, Ozanan Meireles, Guy Rosman; Artificial Intelligence in
Anesthesiology: Current Techniques, Clinical Applications, and Limitations. Anesthesiology 2020; 132:379–
394 doi: https://doi.org/10.1097/ALN.0000000000002960
21. Lei VJ, Luong T, Shan E, Chen X, Neuman MD, En- eanya ND, et al. Risk Stratification for Postoperative
Acute Kidney Injury in Major Noncardiac Surgery Using Preoperative and Intraoperative Data. JAMA Netw
Open 2019;2:e1916921.
22. She S, Huang Y, Review of patient controlled analgesia and clinical strategy in China: Journal of Practical
Pain, 2018; 4(14); 247-49
23. Wang S, Duan N, Li X, Effect of intelligent patient-controlled analgesia on adverse reactions and satisfaction
of postoperative analgesia patients: Guangdong Medicine, 2020; 41(11); 1097-110
24. Transcontinental anaesthesia: a pilot study
Hemmerling, T.M. et al. British Journal of Anaesthesia, Volume 110, Issue 5, 758 - 763. .
doi:10.1093/bja/aes498
25. Biro P, Hofmann P, Gage D, Automated tracheal intubation in an airway manikin using a robotic endoscope:
A proof of concept study: Anesthesia, 2020; 75(7); 881-86.
26. Nundy S, Cooper LA, Mate KS. The Quintuple Aim for Health Care Improvement: A New Imperative to
Advance Health Equity. JAMA. 2022;327(6):521–522. doi:10.1001/jama.2021.25181
27. Tan JM, Cannesson MP. Artificial intelligence, patient safety, and achieving quintuple aim in anesthesiology.
Anesthesia Patient Safety Foundation (APSF) newsletter; 2023;1 (38);1-4.
28. Paiste, H.J., Godwin, R.C., Smith, A.D., Berkowitz, D.E. and Melvin, R.L. (2024). Strengths-weaknesses-
opportunities-threats analysis of artificial intelligence in anesthesiology and perioperative medicine. Frontiers in
digital health, 6. doi:https://doi.org/10.3389/fdgth.2024.1316931.
29. Goldfarb, A. and Teodoridis, F. (2022). Why is AI adoption in health care lagging? [online] Brookings.
Available at: https://www.brookings.edu/articles/why-is-ai-adoption-in-health-care-lagging/.
30. Khan, B., Fatima, H., Qureshi, A. et al. Drawbacks of Artificial Intelligence and Their Potential Solutions in
the Healthcare Sector. Biomedical Materials & Devices 1, 731–738 (2023). https://doi.org/10.1007/s44174-023-
00063-2

1696
ISSN: 2320-5407 Int. J. Adv. Res. 12(09), 1687-1697

31. Gerke, S., Minssen, T. and Cohen, G. (2020). Ethical and Legal Challenges of Artificial intelligence-driven
Healthcare. Artificial Intelligence in Healthcare, 1(1), pp.295–336. doi:https://doi.org/10.1016/B978-0-12-
818438-7.00012-5.
32. World Health Organization. News release. Geneva, 2023. https://www.who.int/news/item/19-10-2023-who-
outlines-considerations-for-regulation-of-artificial-intelligence-for-health.

1697