Artificial intelligence-assisted airway evaluation and imaging-based decision-making in head and neck anesthesia: a narrative review

Introduction

Anesthesia for head and neck surgery, otorhinolaryngologic procedures, and oral and maxillofacial surgeries is intrinsically complicated and frequently necessitates individualized planning (1). Airway management is one of the most important issues in this situation, and it can be seriously hampered by anatomical variability, tumor-related obstruction, prior radiation therapy, or prior surgical procedures. Furthermore, the requirement for a shared airway between surgeons and anesthesiologists makes the procedure even more challenging and necessitates exact coordination (2). These factors emphasize the significance of precise preoperative assessment and strategic planning and increase the risk of airway-related complications.

Beyond airway management, regional anesthesia techniques in the cervical and craniofacial regions have gained increasing attention as adjuncts or alternatives to general anesthesia. However, these approaches require detailed anatomical knowledge and technical precision, particularly when performed under ultrasound guidance (3). Variations in anatomy, reduced tissue contrast, and operator dependency can limit reproducibility and increase the risk of complications, reinforcing the need for improved guidance and standardization (4,5).

In recent years, advances in medical imaging, especially ultrasound and computed tomography (CT), have enhanced the ability to visualize anatomical structures relevant to anesthetic planning (6,7). These modalities allow for a more objective assessment of airway morphology, soft tissue characteristics, and spatial relationships between critical structures. At the same time, the growing availability of digital health data and increased computational power have driven the emergence of artificial intelligence (AI) applications in medicine (8,9).

Machine learning and deep learning techniques have shown promising results in analyzing complex datasets, identifying patterns not readily apparent to clinicians, and supporting predictive modeling (9). In anesthesiology, these methods have been explored for tasks such as risk prediction, image interpretation, and clinical decision support (10,11). Their relevance is particularly evident in head and neck anesthesia, where the integration of imaging data with clinical variables may improve risk stratification, reduce uncertainty, and support more informed and individualized decision-making (12).

Over the past few years, the number of publications on AI in anesthesia has increased substantially, which reflects the growing interest in its clinical application. There is a large body of evidence in the fields of perioperative risk prediction, image analysis, decision support systems and ultrasound-guided regional anesthesia (10,13,14). But this trend has not been consistent across all subspecialties. Much of the available literature has been related to general anesthesia, critical care settings, or peripheral nerve blocks performed in anatomical regions other than head and neck. Consequently, many of the concepts currently discussed in head and neck anesthesia are extrapolated from general anesthetic applications, rather than supported by evidence generated specifically in oral and maxillofacial or head and neck surgical populations.

Imaging modalities such as CT and ultrasound have been increasingly integrated into anesthetic practice, especially for airway evaluation and regional techniques (4,15,16). Despite these advances, the use of AI for imaging-based decision-making in the context of head and neck anesthesia is still at the early stage of clinical translation. Much of the available evidence is extrapolated from wider anesthetic use, with a relative lack of studies specifically relating to oral and maxillofacial and head and neck procedures. Hence, a focused synthesis of the literature is required to critically examine what is known, determine clinically relevant applications and identify gaps in current knowledge. This review covers airway risk stratification, CT-based airway assessment, ultrasound-guided regional anesthesia, and imaging-assisted decision support, all of which are directly relevant to the management of anatomically complex patients. To ensure transparency and clarity of the way this review was conducted and reported, we present this article in accordance with the Narrative Review reporting checklist (available at https://joma.amegroups.com/article/view/10.21037/joma-2026-0012/rc).

Methods

This narrative review was conducted to provide a clinically oriented synthesis of current evidence on AI applications in head and neck anesthesia. A structured literature search was performed using the electronic databases PubMed, Google Scholar, Web of Science, and Scopus.

The search strategy included combinations of Medical Subject Headings (MeSH) terms and free-text keywords such as “artificial intelligence”, “machine learning”, “deep learning”, “head and neck anesthesia”, “airway management”, “computed tomography”, “ultrasound”, and “ultrasound-guided anesthesia”. Boolean operators (AND/OR) were used to refine the search and improve relevance.

The search was conducted from January 1997 to June 2025. This period was chosen to allow the gradual development of AI applications in anesthesiology, especially the increasing use of imaging-based techniques and machine learning approaches in the last decades. The period of time was considered appropriate to provide a contemporary overview of the evidence available given the rapid development of this field.

Studies were included if they were related to AI use in anesthesiology with a focus on airway assessment, imaging analysis, ultrasound-guided procedures, or clinical decision support. The primary studies and review articles were considered. Articles published in English language only were included. Studies were selected based on their relevance to the objectives of this review and their contribution to understanding the current role of AI in head and neck anesthesia.

Titles, abstracts and full texts were qualitatively evaluated for the selection process. Special emphasis was given to studies with direct implications to airway management, imaging based risk stratification, ultrasound guided procedures and perioperative decision support in patients undergoing oral and maxillofacial, head and neck or otorhinolaryngologic procedures. Studies with potential clinical applicability and studies addressing challenges, limitations and future directions of AI-assisted decision-making were preferred. Given the narrative nature of this review, no formal systematic screening process or risk of bias assessment was conducted. The summary of the search strategy is shown in Table 1.

AI for airway assessment and risk stratification

Airway management is one of the most critical and challenging aspects of anesthesia for oral and maxillofacial surgery, as well as for head and neck and otorhinolaryngologic procedures. In many of these patients, airway control can be complicated by maxillofacial trauma, restricted mouth opening, tumor-related obstruction, previous surgical procedures, radiotherapy-related changes, or significant anatomical distortion. These conditions may substantially increase the difficulty of airway management and make accurate preoperative assessment essential for safe anesthetic planning (1,6).

Traditionally, airway evaluation has relied on clinical examination and individual predictors of difficult intubation. While these methods remain fundamental in daily practice, their accuracy is often limited when used alone, particularly in patients with complex head and neck anatomy (17,18). For this reason, there has been growing interest in more objective approaches capable of integrating multiple sources of information and providing a more comprehensive assessment of airway-related risk.

AI has emerged as a promising tool in this context. Machine learning models can combine clinical variables, imaging findings, and perioperative information to identify patterns that may not be readily apparent during routine evaluation (12,19,20). Several studies have reported improved performance in predicting difficult mask ventilation and tracheal intubation when compared with traditional single-parameter assessment methods (18,20-22). In oral and maxillofacial anesthesia, this may be particularly useful in patients with anticipated airway difficulties, helping clinicians identify higher-risk cases before entering the operating room.

Imaging has become an increasingly important component of this process. CT is frequently available in patients undergoing head and neck procedures, especially those with facial trauma, benign or malignant tumors, orthognathic deformities, or previous reconstructive surgery. In addition to its diagnostic role, CT provides detailed information about airway dimensions, soft tissues, and anatomical relationships that may influence airway management (23,24). More recently, AI-based tools have been applied to CT images to automatically segment the airway and extract quantitative measurements such as airway volume, cross-sectional area, and soft tissue thickness (8,14,24). These approaches offer a more objective evaluation and may reduce the variability associated with subjective image interpretation (8,14,24).

Ultrasound has also attracted growing interest as a bedside imaging tool for airway assessment. It allows visualization of anterior neck structures and may assist in identifying anatomical landmarks associated with difficult laryngoscopy or intubation (25). Although AI-assisted ultrasound analysis remains in an earlier stage of development, preliminary studies suggest that automated landmark recognition and image interpretation may improve consistency and reduce operator dependency. At present, however, most of these applications remain investigational.

Despite the encouraging results reported in the literature, important limitations remain. Much of the available evidence is based on retrospective datasets, and external validation is still limited (12,20,22). Furthermore, relatively few studies have specifically focused on oral and maxillofacial surgical patients, making it difficult to determine how well these tools perform in real-world clinical scenarios. As a result, AI should currently be viewed as a decision-support tool capable of complementing, rather than replacing, clinical judgment.

Looking ahead, the combination of AI with CT and ultrasound imaging has the potential to improve airway risk stratification and support more individualized anesthetic planning. This may be particularly valuable in patients with distorted anatomy, trismus, post-radiotherapy changes, or other conditions commonly encountered in oral and maxillofacial practice. As more clinically focused studies become available, these technologies may play an increasingly important role in preoperative airway assessment and perioperative decision-making.

A schematic representation of the AI-assisted workflow for airway risk assessment and clinical decision-making in head and neck anesthesia is shown in Figure 2.

Figure 2 Illustration of the workflow integrating imaging findings, clinical data, and patient history for AI-assisted airway risk assessment. Machine learning and deep learning models analyze multimodal information to identify factors associated with difficult airway management, including airway anatomy, tumor characteristics, previous radiotherapy, limited mouth opening, and anatomical variations. The resulting risk stratification may support individualized anesthetic planning, improve communication among the perioperative team, and contribute to safer clinical decision-making and improved patient outcomes. AI, artificial intelligence.

Ultrasound-guided regional anesthesia and AI assistance

Ultrasound-guided regional anesthesia has become an increasingly valuable component of anesthetic practice in head and neck procedures, including cervical plexus blocks and other peripheral nerve blocks (3,4,25). The ability to directly visualize anatomical structures in real time has significantly improved the precision of needle placement and may help reduce the risk of inadvertent vascular or neural injury (3-5). In the head and neck region, where important neurovascular structures are often located in close proximity, ultrasound guidance offers a level of anatomical detail that is difficult to achieve using traditional landmark-based techniques alone (3,4).

This approach may be particularly useful in selected oral and maxillofacial procedures, including surgeries involving the cervical region, reconstructive procedures, oncologic interventions, and cases in which anatomical landmarks are altered by trauma, previous surgery, or radiotherapy (3,4,25). In these situations, direct visualization can improve procedural confidence and facilitate safer anesthetic management (3-5,25).

Despite its advantages, successful ultrasound-guided regional anesthesia still depends heavily on operator expertise. Obtaining high-quality images and correctly identifying relevant anatomical structures can be challenging, particularly in patients with distorted anatomy or altered tissue planes (26). Small differences in probe positioning, image optimization, or anatomical interpretation may significantly affect procedural success. Consequently, variability among operators remains an important limitation, and mastering these techniques often requires a considerable learning curve (27,28).

In this context, AI has emerged as a promising tool to support ultrasound-based procedures. AI algorithms, particularly those based on deep learning, have been applied to ultrasound imaging to enhance image quality, automatically identify anatomical landmarks, and assist in real-time interpretation (29). These systems can recognize structures such as nerves, vessels, and fascial planes, and may provide visual overlays or guidance cues that help standardize image interpretation and improve procedural confidence (29,30).

AI has emerged as a promising strategy to address some of these challenges. Recent advances in deep learning have enabled the development of systems capable of enhancing ultrasound image quality, automatically identifying anatomical landmarks, and assisting with real-time image interpretation (29). These tools can recognize structures such as nerves, vessels, and fascial planes and provide visual guidance that may help standardize image interpretation and reduce operator dependency (29,30).

AI-assisted technologies have also been investigated for needle tracking and trajectory guidance during ultrasound-guided procedures (31). By providing additional visual feedback during needle advancement, these systems may improve procedural accuracy and potentially increase operator confidence, particularly among less experienced clinicians (32). Such applications could contribute to wider adoption of ultrasound-guided regional anesthesia in head and neck procedures and may facilitate training in complex anatomical regions.

Nevertheless, most available evidence remains focused on technical feasibility rather than clinical effectiveness. Many studies have been conducted under experimental or controlled conditions and primarily evaluate algorithm performance rather than patient-centered outcomes (13,33). As a result, there is still limited evidence demonstrating that AI-assisted ultrasound improves procedural success rates, reduces complications, or enhances workflow efficiency in routine clinical practice (13,29).

At present, AI-assisted ultrasound should be viewed as an emerging adjunct rather than a mature clinical technology. Additional prospective studies are needed to determine whether the promising results observed in preliminary investigations translate into meaningful benefits for patients undergoing oral and maxillofacial and head and neck procedures.

CT-based airway assessment and AI applications

Patients undergoing oral and maxillofacial and head and neck procedures frequently present anatomical conditions that can complicate airway management. Maxillofacial trauma, head and neck tumors, orthognathic deformities, previous reconstructive surgery, trismus, and post-radiotherapy changes may significantly alter airway anatomy and make conventional bedside assessment less reliable (23,34). In many of these situations, CT examinations are already available as part of the routine diagnostic workup, providing anesthesiologists with detailed anatomical information that can support preoperative planning without the need for additional imaging studies.

One of the main strengths of CT is its ability to provide a comprehensive view of the airway and surrounding structures. Unlike clinical examination alone, CT allows visualization of the airway lumen, adjacent soft tissues, and the spatial relationships between critical anatomical landmarks (23,24). This information can be particularly valuable when planning airway management in patients with distorted anatomy or when a difficult intubation is anticipated.

Another important advantage of CT is the possibility of obtaining objective and reproducible measurements (35). Parameters such as airway volume, cross-sectional area, and soft tissue thickness can be quantified with high precision (6,16,35). In clinical practice, these measurements may help clinicians better understand the anatomical factors contributing to airway difficulty and support preoperative planning in patients with complex craniofacial or cervical conditions.

More recently, AI has been applied to CT imaging to expand its potential role in airway evaluation. Deep learning algorithms can automatically process CT datasets, segment the airway, and extract quantitative features in a rapid and reproducible manner (8,14,24). By reducing the need for labor-intensive manual analysis, these approaches may improve workflow efficiency and decrease the variability associated with subjective image interpretation (8,14).

The combination of CT-derived features with clinical information has also created opportunities for more advanced risk prediction models. Several AI-based approaches have been developed to identify patients at increased risk of difficult airway management by integrating anatomical and clinical variables (22,24). Such tools move beyond purely descriptive imaging and introduce a more individualized approach to airway assessment. For oral and maxillofacial anesthesia, this may be particularly relevant in patients with facial trauma, trismus, head and neck malignancies, extensive reconstructive procedures, or post-radiotherapy changes, where conventional airway assessment may underestimate procedural complexity.

Despite these promising developments, several limitations remain. Most studies have been based on retrospective datasets or controlled research environments, and relatively few have evaluated the impact of AI-assisted CT analysis on clinical outcomes (12,22,33,36). Furthermore, differences in imaging protocols, patient populations, and validation strategies make comparisons across studies difficult. As a result, although AI-assisted CT analysis shows considerable promise, many currently available tools should still be considered investigational rather than fully established clinical technologies.

At present, CT remains a valuable component of airway assessment in selected oral and maxillofacial patients, while AI serves primarily as a decision-support technology that may enhance image interpretation and risk stratification. Future studies should focus on prospective validation, external testing across diverse patient populations, and demonstration of clinically meaningful benefits before widespread implementation can be recommended (12,22,33,36).

Current applications and emerging trends

The application of AI in head and neck anesthesia is growing in several areas, such as airway risk stratification, CT-based airway assessment, ultrasound-guided regional anesthesia, and perioperative decision support. Table 2 summarizes that most of the published work has been in the area of airway applications, which reflects the central role of airway management in oral and maxillofacial and head and neck procedures (12,19,21,22,24,36). Most of these approaches aim to improve the preoperative evaluation by combining clinical information with data from imaging, particularly anatomical measurements from CT examinations. The quality of the evidence also varies considerably in the applications but the common aim is to encourage more accurate risk assessment and more individualised anaesthesia management in patients with anatomically difficult airways.

The data in Table 2 also tell us important differences in the maturity of AI applications that are available today. CT-based airway assessment and airway prediction models are the most advanced fields with several studies showing promising predictive performance and the feasibility of incorporating imaging-derived parameters in risk assessment strategies (12,19,21,22,24). On the other hand, applications based on ultrasound, such as the automated recognition of landmarks and needle guidance systems, are still in the proof-of-concept or early validation stage with little evidence for routine clinical implementation (13,29-32).

One of the most consistent trends seen in the literature is the increasing use of medical imaging as a source of quantitative data for AI models. CT-based techniques can automatically identify and segment airway structures, providing objective anatomical measurements that might help predict challenging airway situations (8,14,22,24). Ultrasound-based systems have also been employed to identify anatomical landmarks, track needles and provide procedural guidance during regional anesthesia (13,29-32). These are encouraging developments, but many are still technically focused and have yet to demonstrate clear benefits in patient-centered outcomes.

Another important point is that a significant portion of the current evidence is derived from retrospective studies performed in relatively controlled research settings (12,22,33,36). External validation is limited and there are few studies that have specifically focused on patient populations typical for oral and maxillofacial anesthesia. Available datasets often lack patients with facial trauma, trismus, head and neck tumours, post-radiotherapy anatomical changes or complex reconstructive procedures, yet they are amongst the ones most likely to benefit from improved airway assessment and decision-support tools.

All this evidence together suggests that AI should be viewed as a supportive technology rather than a substitute for clinical judgment at present. Several applications have shown promising technical performance, but further prospective validation, multicenter collaboration, and evaluation of clinically meaningful outcomes are required before widespread implementation can be recommended. Future research should be directed toward applications that are directly relevant to oral and maxillofacial anesthesia, where anatomical complexity and airway-related challenges continue to be important determinants of perioperative risk and patient safety.

Strengths and limitations of the review

There are several strengths to this review. In particular, it addresses the intersection of AI, imaging and airway management in oral and maxillofacial, head and neck, and otorhinolaryngologic anaesthesia, a field that continues to be less widely explored than broader applications of AI in anaesthesiology. In addition, the review includes evidence from various fields including airway risk stratification, CT-based airway assessment, ultrasound-guided regional anaesthesia and perioperative decision support, giving a clinically relevant perspective for the anaesthesiologist managing anatomically complex patients. In addition, the discussion focused on the potential clinical applicability of emerging technologies and their relevance to everyday anaesthetic practice rather than on technical performance metrics.

There are also some limitations that should be acknowledged. This study is a narrative review and does not adhere to the methodological rigour of a systematic review and therefore did not include a formal screening process, quantitative synthesis, or risk-of-bias assessment. The study selection was subject to some author judgement which may have led to selection bias. Moreover, a large portion of the literature consists of retrospective studies, technical feasibility studies and proof-of-concept studies, which limits the power of conclusions on real-world clinical effectiveness. Another limitation is that the literature search was conducted for studies published through June 2025. In addition, there are few studies that have specifically examined oral and maxillofacial surgical populations. Therefore, additional research in clinically relevant head and neck settings is needed.

Conclusions

AI can improve many aspects of head and neck anaesthesia, especially in airway assessment, imaging interpretation and ultrasound-guided procedures. AI-based approaches could allow for more objective risk stratification and more personalised anaesthetic planning by integrating clinical and imaging data. While many applications remain at early stages of clinical implementation, current evidence suggests that AI may become an increasingly valuable adjunct in the management of anatomically complex patients. Future studies should address prospective validation and clinically relevant applications, especially in oral and maxillofacial anaesthesia, where airway-related challenges continue to be a major determinant of perioperative risk and patient safety.

Figure 1 Schematic representation of the workflow integrating imaging acquisition, clinical information (A), AI-based analysis (B), and clinical decision support (C) in head and neck anesthesia. Ultrasound and CT provide anatomical and airway-related data that are combined with clinical information and processed using AI techniques. The resulting outputs include airway measurements, difficult airway risk stratification, and predictive metrics that may support individualized anesthetic planning and intraoperative decision-making. This framework illustrates the potential role of AI in transforming imaging and clinical data into actionable information for the management of anatomically complex patients. AI, artificial intelligence; CT, computed tomography.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://joma.amegroups.com/article/view/10.21037/joma-2026-0012/rc

Peer Review File: Available at https://joma.amegroups.com/article/view/10.21037/joma-2026-0012/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://joma.amegroups.com/article/view/10.21037/joma-2026-0012/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Apfelbaum JL, Hagberg CA, Connis RT, et al. 2022 American Society of Anesthesiologists Practice Guidelines for Management of the Difficult Airway. Anesthesiology 2022;136:31-81. [Crossref] [PubMed]
2. Cook TM. Strategies for the prevention of airway complications - a narrative review. Anaesthesia 2018;73:93-111. [Crossref] [PubMed]
3. Sites BD, Chan VW, Neal JM, et al. The American Society of Regional Anesthesia and Pain Medicine and the European Society Of Regional Anaesthesia and Pain Therapy Joint Committee recommendations for education and training in ultrasound-guided regional anesthesia. Reg Anesth Pain Med 2009;34:40-6. [Crossref] [PubMed]
4. Marhofer P, Harrop-Griffiths W, Kettner SC, et al. Fifteen years of ultrasound guidance in regional anaesthesia: part 1. Br J Anaesth 2010;104:538-46. [Crossref] [PubMed]
5. Perlas A, Chan VW, Simons M. Brachial plexus examination and localization using ultrasound and electrical stimulation: a volunteer study. Anesthesiology 2003;99:429-35. [Crossref] [PubMed]
6. de Jong PA, Müller NL, Paré PD, et al. Computed tomographic imaging of the airways: relationship to structure and function. Eur Respir J 2005;26:140-52. [Crossref] [PubMed]
7. Osorio F, Perilla M, Doyle DJ, et al. Cone beam computed tomography: an innovative tool for airway assessment. Anesth Analg 2008;106:1803-7. [Crossref] [PubMed]
8. Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal 2017;42:60-88. [Crossref] [PubMed]
9. Ker J, Wang L, Rao J, et al. Deep learning applications in medical image analysis. IEEE Access 2017;6:9375-89.
10. Hashimoto DA, Witkowski E, Gao L, et al. Artificial Intelligence in Anesthesiology: Current Techniques, Clinical Applications, and Limitations. Anesthesiology 2020;132:379-94. [Crossref] [PubMed]
11. Solanki SL. Artificial Intelligence in Airway Management and Anaesthesia. Airway 2023;6:1-3.
12. Wilk M, Pikiewicz W, Florczak K, et al. Use of Artificial Intelligence in Difficult Airway Assessment: The Current State of Knowledge. J Clin Med 2025;14:1602. [Crossref] [PubMed]
13. Harris J, Kamming D, Bowness JS. Artificial intelligence in regional anesthesia. Curr Opin Anaesthesiol 2025;38:605-10. [Crossref] [PubMed]
14. Smistad E, Falch TL, Bozorgi M, et al. Medical image segmentation on GPUs--a comprehensive review. Med Image Anal 2015;20:1-18. [Crossref] [PubMed]
15. Adnet F, Borron SW, Racine SX, et al. The intubation difficulty scale (IDS): proposal and evaluation of a new score characterizing the complexity of endotracheal intubation. Anesthesiology 1997;87:1290-7. [Crossref] [PubMed]
16. Prasad A, Yu E, Wong DT, et al. Comparison of sonography and computed tomography as imaging tools for assessment of airway structures. J Ultrasound Med 2011;30:965-72. [Crossref] [PubMed]
17. Shiga T, Wajima Z, Inoue T, et al. Predicting difficult intubation in apparently normal patients: a meta-analysis of bedside screening test performance. Anesthesiology 2005;103:429-37. [Crossref] [PubMed]
18. Nørskov AK, Rosenstock CV, Wetterslev J, et al. Diagnostic accuracy of anaesthesiologists' prediction of difficult airway management in daily clinical practice: a cohort study of 188 064 patients registered in the Danish Anaesthesia Database. Anaesthesia 2015;70:272-81. [Crossref] [PubMed]
19. Yamanaka S, Goto T, Morikawa K, et al. Machine Learning Approaches for Predicting Difficult Airway and First-Pass Success in the Emergency Department: Multicenter Prospective Observational Study. Interact J Med Res 2022;11:e28366. [Crossref] [PubMed]
20. Sezari P, Sabouri AS, Dabbagh A, et al. Applied Artificial Intelligence in Airway Management: A Narrative Review. Anesthesiol Clin 2025;43:549-61. [Crossref] [PubMed]
21. Zhou CM, Wang Y, Xue Q, et al. Predicting difficult airway intubation in thyroid surgery using multiple machine learning and deep learning algorithms. Front Public Health 2022;10:937471. [Crossref] [PubMed]
22. De Rosa S, Bignami E, Bellini V, et al. The Future of Artificial Intelligence Using Images and Clinical Assessment for Difficult Airway Management. Anesth Analg 2025;140:317-25. [Crossref] [PubMed]
23. Grimes D, MacLeod I, Taylor T, et al. Computed tomography as an aid to planning intubation in the difficult airway. Br J Oral Maxillofac Surg 2016;54:80-2. [Crossref] [PubMed]
24. Wang J, Xia M, Jiang H. Advances in studies on imaging and artificial intelligence technology-assisted difficult airway assessment. Front Oral Maxillofac Med 2021;3:66.
25. Kristensen MS, Teoh WH, Graumann O, et al. Ultrasonography for clinical decision-making and intervention in airway management: from the mouth to the lungs and pleurae. Insights Imaging 2014;5:253-79. [Crossref] [PubMed]
26. Chin KJ, Perlas A, Chan VW, et al. Needle visualization in ultrasound-guided regional anesthesia: challenges and solutions. Reg Anesth Pain Med 2008;33:532-44. [Crossref] [PubMed]
27. Sites BD, Gallagher JD, Cravero J, et al. The learning curve associated with a simulated ultrasound-guided interventional task by inexperienced anesthesia residents. Reg Anesth Pain Med 2004;29:544-8. [Crossref] [PubMed]
28. Barrington MJ, Kluger R. Ultrasound guidance reduces the risk of local anesthetic systemic toxicity following peripheral nerve blockade. Reg Anesth Pain Med 2013;38:289-99. [Crossref] [PubMed]
29. Wang S, Liu JB, Zhu Z, et al. Artificial Intelligence in Ultrasound Imaging: Current Research and Applications. Advanced Ultrasound in Diagnosis and Therapy 2019;3:53-61.
30. Noble JA, Boukerroui D. Ultrasound image segmentation: a survey. IEEE Trans Med Imaging 2006;25:987-1010. [Crossref] [PubMed]
31. Hui X, Rajendran P, Ling T, et al. Ultrasound-guided needle tracking with deep learning: A novel approach with photoacoustic ground truth. Photoacoustics 2023;34:100575. [Crossref] [PubMed]
32. Yan L, Li Q, Fu K, et al. Progress in the Application of Artificial Intelligence in Ultrasound-Assisted Medical Diagnosis. Bioengineering (Basel) 2025;12:288. [Crossref] [PubMed]
33. Esteva A, Robicquet A, Ramsundar B, et al. A guide to deep learning in healthcare. Nat Med 2019;25:24-9. [Crossref] [PubMed]
34. Bola S, Osuji J, Rivero-Bosch M, et al. Airway Management in Otolaryngology and Head and Neck Surgery: A Narrative Review of Current Techniques and Considerations. J Clin Med 2025;14:4717. [Crossref] [PubMed]
35. Mazonakis M, Damilakis J. Computed tomography: What and how does it measure? Eur J Radiol 2016;85:1499-504. [Crossref] [PubMed]
36. Alatau M, Bauer J, Sazonov V. Artificial Intelligence for Predicting Difficult Airways: A Review. J Clin Med 2025;14:8600. [Crossref] [PubMed]
37. Lundberg SM, Nair B, Vavilala MS, et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng 2018;2:749-60. [Crossref] [PubMed]
38. Obermeyer Z, Emanuel EJ. Predicting the Future - Big Data, Machine Learning, and Clinical Medicine. N Engl J Med 2016;375:1216-9. [Crossref] [PubMed]