The role of artificial intelligence in the treatment of obstructive sleep apnea

Application of Artificial Intelligence in Cone-Beam Computed Tomography for Airway Analysis: A Narrative Review

Cone-beam computed tomography (CBCT) has emerged as a promising tool for the analysis of the upper airway, leveraging on its ability to provide three-dimensional information, minimal radiation exposure, affordability, and widespread accessibility. The integration of artificial intelligence (AI) in CBCT for airway analysis has shown improvements in the accuracy and efficiency of diagnosing and managing airway-related conditions. This review aims to explore the current applications of AI in CBCT for airway analysis, highlighting its components and processes, applications, benefits, challenges, and potential future directions. A comprehensive literature review was conducted, focusing on studies published in the last decade that discuss AI applications in CBCT airway analysis. Many studies reported the significant improvement in segmentation and measurement of airway volumes from CBCT using AI, thereby facilitating accurate diagnosis of airway-related conditions. In addition, these AI models demonstrated high accuracy and consistency in their application for airway analysis through automated segmentation tasks, volume measurement, and 3D reconstruction, which enhanced the diagnostic accuracy and allowed predictive treatment outcomes. Despite these advancements, challenges remain in the integration of AI into clinical workflows. Furthermore, variability in AI performance across different populations and imaging settings necessitates further validation studies. Continued research and development are essential to overcome current challenges and fully realize the potential of AI in airway analysis.

Automatic prediction of obstructive sleep apnea in patients with temporomandibular disorder based on multidata and machine learning

Obstructive sleep apnea (OSA) is closely associated with the development and chronicity of temporomandibular disorder (TMD). Given the intricate pathophysiology of both OSA and TMD, comprehensive diagnostic approaches are crucial. This study aimed to develop an automatic prediction model utilizing multimodal data to diagnose OSA among TMD patients. We collected a range of multimodal data, including clinical characteristics, portable polysomnography, X-ray, and MRI data, from 55 TMD patients who reported sleep problems. This data was then analyzed using advanced machine learning techniques. Three-dimensional VGG16 and logistic regression models were used to identify significant predictors. Approximately 53% (29 out of 55) of TMD patients had OSA. Performance accuracy was evaluated using logistic regression, multilayer perceptron, and area under the curve (AUC) scores. OSA prediction accuracy in TMD patients was 80.00-91.43%. When MRI data were added to the algorithm, the AUC score increased to 1.00, indicating excellent capability. Only the obstructive apnea index was statistically significant in predicting OSA in TMD patients, with a threshold of 4.25 events/h. The learned features of the convolutional neural network were visualized as a heatmap using a gradient-weighted class activation mapping algorithm, revealing that it focuses on differential anatomical parameters depending on the absence or presence of OSA. In OSA-positive cases, the nasopharynx, oropharynx, uvula, larynx, epiglottis, and brain region were recognized, whereas in OSA-negative cases, the tongue, nose, nasal turbinate, and hyoid bone were recognized. Prediction accuracy and heat map analyses support the plausibility and usefulness of this artificial intelligence-based OSA diagnosis and prediction model in TMD patients, providing a deeper understanding of regions distinguishing between OSA and non-OSA.

An introduction to machine learning and generative artificial intelligence for otolaryngologists-head and neck surgeons: a narrative review

PURPOSE

Despite the robust expansion of research surrounding artificial intelligence (AI) and machine learning (ML) and their applications to medicine, these methodologies often remain opaque and inaccessible to many otolaryngologists. Especially, with the increasing ubiquity of large-language models (LLMs), such as ChatGPT and their potential implementation in clinical practice, clinicians may benefit from a baseline understanding of some aspects of AI. In this narrative review, we seek to clarify underlying concepts, illustrate applications to otolaryngology, and highlight future directions and limitations of these tools.

METHODS

Recent literature regarding AI principles and otolaryngologic applications of ML and LLMs was reviewed via search in PubMed and Google Scholar.

RESULTS

Significant recent strides have been made in otolaryngology research utilizing AI and ML, across all subspecialties, including neurotology, head and neck oncology, laryngology, rhinology, and sleep surgery. Potential applications suggested by recent publications include screening and diagnosis, predictive tools, clinical decision support, and clinical workflow improvement via LLMs. Ongoing concerns regarding AI in medicine include ethical concerns around bias and data sharing, as well as the "black box" problem and limitations in explainability.

CONCLUSIONS

Potential implementations of AI in otolaryngology are rapidly expanding. While implementation in clinical practice remains theoretical for most of these tools, their potential power to influence the practice of otolaryngology is substantial.

LEVEL OF EVIDENCE: 4

Comparative Performance of ChatGPT 3.5 and GPT4 on Rhinology Standardized Board Examination Questions

Objective

Advances in deep learning and artificial intelligence (AI) have led to the emergence of large language models (LLM) like ChatGPT from OpenAI. The study aimed to evaluate the performance of ChatGPT 3.5 and GPT4 on Otolaryngology (Rhinology) Standardized Board Examination questions in comparison to Otolaryngology residents.

Methods

This study selected all 127 rhinology standardized questions from www.boardvitals.com, a commonly used study tool by otolaryngology residents preparing for board exams. Ninety-three text-based questions were administered to ChatGPT 3.5 and GPT4, and their answers were compared with the average results of the question bank (used primarily by otolaryngology residents). Thirty-four image-based questions were provided to GPT4 and underwent the same analysis. Based on the findings of an earlier study, a pass-fail cutoff was set at the 10th percentile.

Results

On text-based questions, ChatGPT 3.5 answered correctly 45.2% of the time (8th percentile) (P = .0001), while GPT4 achieved 86.0% (66th percentile) (P = .001). GPT4 answered image-based questions correctly 64.7% of the time. Projections suggest that ChatGPT 3.5 might not pass the American Board of Otolaryngology Written Question Exam (ABOto WQE), whereas GPT4 stands a strong chance of passing.

Discussion

The older LLM, ChatGPT 3.5, is unlikely to pass the ABOto WQE. However, the advanced GPT4 model exhibits a much higher likelihood of success. This rapid progression in AI indicates its potential future role in otolaryngology education.

Implications for Practice

As AI technology rapidly advances, it may be that AI-assisted medical education, diagnosis, and treatment planning become commonplace in the medical and surgical landscape.

Level of Evidence

Level 5.

Achieving Better Understanding of Obstructive Sleep Apnea Treatment Effects on Cardiovascular Disease Outcomes through Machine Learning Approaches: A Narrative Review

Obstructive sleep apnea (OSA) affects almost a billion people worldwide and is associated with a myriad of adverse health outcomes. Among the most prevalent and morbid are cardiovascular diseases (CVDs). Nonetheless, randomized controlled trials (RCTs) of OSA treatment have failed to show improvements in CVD outcomes. A major limitation in our field is the lack of precision in defining OSA and specifically subgroups with the potential to benefit from therapy. Further, this has called into question the validity of using the time-honored apnea-hypopnea index as the ultimate defining criteria for OSA. Recent applications of advanced statistical methods and machine learning have brought to light a variety of OSA endotypes and phenotypes. These methods also provide an opportunity to understand the interaction between OSA and comorbid diseases for better CVD risk stratification. Lastly, machine learning and specifically heterogeneous treatment effects modeling can help uncover subgroups with differential outcomes after treatment initiation. In an era of data sharing and big data, these techniques will be at the forefront of OSA research. Advanced data science methods, such as machine-learning analyses and artificial intelligence, will improve our ability to determine the unique influence of OSA on CVD outcomes and ultimately allow us to better determine precision medicine approaches in OSA patients for CVD risk reduction. In this narrative review, we will highlight how team science via machine learning and artificial intelligence applied to existing clinical data, polysomnography, proteomics, and imaging can do just that.

Artificial intelligence in sleep medicine: Present and future

Artificial intelligence (AI) has impacted many areas of healthcare. AI in healthcare uses machine learning, deep learning, and natural language processing to analyze copious amounts of healthcare data and yield valuable outcomes. In the sleep medicine field, a large amount of physiological data is gathered compared to other branches of medicine. This field is primed for innovations with the help of AI. A good quality of sleep is crucial for optimal health. About one billion people are estimated to have obstructive sleep apnea worldwide, but it is difficult to diagnose and treat all the people with limited resources. Sleep apnea is one of the major contributors to poor health. Most of the sleep apnea patients remain undiagnosed. Those diagnosed with sleep apnea have difficulty getting it optimally treated due to several factors, and AI can help in this situation. AI can also help in the diagnosis and management of other sleep disorders such as insomnia, hypersomnia, parasomnia, narcolepsy, shift work sleep disorders, periodic leg movement disorders, etc. In this manuscript, we aim to address three critical issues about the use of AI in sleep medicine: (1) How can AI help in diagnosing and treating sleep disorders? (2) How can AI fill the gap in the care of sleep disorders? and (3) What are the ethical and legal considerations of using AI in sleep medicine?

The Contribution of Sleep Texture in the Characterization of Sleep Apnea

Obstructive sleep apnea (OSA) is multi-faceted world-wide-distributed disorder exerting deep effects on the sleeping brain. In the latest years, strong efforts have been dedicated to finding novel measures assessing the real impact and severity of the pathology, traditionally trivialized by the simplistic apnea/hypopnea index. Due to the unavoidable connection between OSA and sleep, we reviewed the key aspects linking the breathing disorder with sleep pathophysiology, focusing on the role of cyclic alternating pattern (CAP). Sleep structure, reflecting the degree of apnea-induced sleep instability, may provide topical information to stratify OSA severity and foresee some of its dangerous consequences such as excessive daytime sleepiness and cognitive deterioration. Machine learning approaches may reinforce our understanding of this complex multi-level pathology, supporting patients' phenotypization and easing in a more tailored approach for sleep apnea.