Barriers of artificial intelligence implementation in the diagnosis of obstructive sleep apnea
Open accessAbstract
Background: Obstructive sleep apnea is a common clinical condition and has a significant impact on the health of patients if untreated. The current diagnostic gold standard for obstructive sleep apnea is polysomnography, which is labor intensive, requires specialists to utilize, expensive, and has accessibility challenges. There are also challenges with awareness and identification of obstructive sleep apnea in the primary care setting. Artificial intelligence systems offer the opportunity for a new diagnostic approach that addresses the limitations of polysomnography and ultimately benefits patients by streamlining the diagnostic expedition. Main body: The purpose of this project is to elucidate the barriers that exist in the implementation of artificial intelligence systems into the diagnostic context of obstructive sleep apnea. It is essential to understand these challenges in order to proactively create solutions and establish an efficient adoption of this new technology. The literature regarding the evolution of the diagnosis of obstructive sleep apnea, the role of artificial intelligence in the diagnosis, and the barriers in artificial intelligence implementation was reviewed and analyzed. Conclusion: The barriers identified were categorized into different themes including technology, data, regulation, human resources, education, and culture. Many of these challenges are ubiquitous across artificial intelligence implementation in any medical diagnostic setting. Future research directions include developing solutions to the barriers presented in this project. Graphical abstract:
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Inaccurate and underrepresentative training data sets for AI models can cause bias, misleading predictions, adverse events, and large-scale discrimination, causing barriers to adoption.
“Inaccurate and underrepresentative training data sets for AI models can cause bias, misleading predictions, adverse events, and large-scale discrimination, causing barriers to adoption [32,41,47,48].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
The Main body section on barriers says unreliable training data obstructs implementation and that dataset bias can make AI predictions less accurate. It also describes unrepresentative datasets and risks of bias affecting marginalized groups. Our copy does not establish adverse events or large-scale discrimination as consequences of those datasets. (LLM, conf=0.87)
Experimental — accuracy not yet measuredPhysicians expressed that they would be less likely to use an AI given their familiarity with the patient's condition and the value of intuition in clinical decision-making.
“From a clinician standpoint, physicians expressed that they would be less likely to use an AI, given their familiarity with the patient’s condition and the value of intuition in clinical decision-making [32,35].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
In “Barriers involved in AI implementation” (Environment/culture; page unavailable), the paper says physician intuition is often valued above technological approaches. The same section describes physicians as hesitant to adopt AI. It does not report physicians expressing that they would be less likely to use AI because they were familiar with a patient’s condition. (LLM, conf=0.84)
Experimental — accuracy not yet measuredThe presence of AI experts, perhaps a multidisciplinary team with clinical scientists, data science and subject matter experts with AI skills, an innovation manager, AI experts to provide training, and local champions within departments involved in the end-to-end process, was considered an important element for adoption.
“The presence of AI experts, perhaps a multidisciplinary team, particularly with clinical scientists, data science and subject matter experts with AI skills, an innovation manager, AI experts to provide training, and local champions within departments involved in the end-to-end process, was considered an important element for adoption [32,57,59,62,71].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
The Human resources section says AI implementation requires knowledgeable personnel to build, maintain and improve systems, as well as people to establish training programs. The Education section also identifies a need for knowledgeable teachers. Neither section specifies a multidisciplinary team, an innovation manager, or departmental champions involved throughout the end-to-end process. (LLM, conf=0.85)
Experimental — accuracy not yet measuredShortages of personnel with the required skills and the quality of IT infrastructure available for AI implementation were reported as barriers.
“Shortages of personnel with the required skills were reported as barriers, along with the quality of IT infrastructure available for AI implementation [32,35,36,40,45,50].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
The ‘Human resources’ and ‘Education’ subsections under ‘Barriers involved in AI implementation’ identify the need for knowledgeable personnel and training, supporting the skilled-personnel portion. Our copy does not identify the quality of available IT infrastructure as a barrier; searches for infrastructure and related technical resources found no such passage. (LLM, conf=0.84)
Experimental — accuracy not yet measuredPrivacy and ethics concerns were focused on maintaining confidentiality, ensuring processes are in place to obtain consent, and having informed consent with clarity on how data will be processed.
“Concerns about privacy and ethics were focused on maintaining confidentiality, ensuring processes are in place to obtain consent, and having informed consent with clarity on how the data will be processed [32,39,47,56].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
The “Privacy and confidentiality” subsection identifies confidentiality and ethical data management as concerns. The following passage says patient consent is important and may need to be a dynamic process, but it does not say that informed consent must clearly explain how data will be processed. No page numbers are available in our copy. (LLM, conf=0.84)
Experimental — accuracy not yet measured
Lack of and uncertainty surrounding funding are barriers to implementation.
“The lack of and uncertainty surrounding funding are presented as barriers to implementation [32,51,56].”
From Barriers to and facilitators of artificial intelligence adoption in health care: scoping review (2024) · cites this paper as [32]
The Main body’s “Human resources” subsection identifies finances as a resource needed to develop and implement AI systems, supporting a financial barrier. It does not say that funding is lacking or that uncertainty about funding is a barrier. (LLM, conf=0.83)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362 · cited 1× · see citing works
“In the development of a neural network designed to distinguish benign and malignant lesions, it was found that the algorithm was more likely to predict malignancy when a ruler was present on the image [51].”
BackgroundIncreased Prevalence of Sleep-Disordered Breathing in Adults
Peppard, P. E., Young, T., Barnet, J. H., Palta, M., Hagen, E. W., & Hla, K. M. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kws342 · Cited by 4,618 · cited 1× · see citing works
“Obstructive sleep apnea (OSA) is a common clinical condition, affecting 14% of men and 5% of women, characterized by a cessation of airflow due to upper airway obstruction with simultaneous respiratory effort present [1].”
BackgroundKey challenges for delivering clinical impact with artificial intelligence
Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D. (2019) · BMC Medicine · doi:10.1186/s12916-019-1426-2 · Cited by 3,136 · cited 1× · see citing works
“” [34] There is a balance between accuracy and algorithm transparency, where the most accurate models (deep learning and ANNs) are the least interpretable and the more intuitive models are less accurate (linear regression) [48].”
BackgroundThe epidemiology of adult obstructive sleep apnea
Punjabi, N. M. (2008) · Proceedings of the American Thoracic Society · doi:10.1513/pats.200709-155mg · Cited by 2,361 · cited 1× · see citing works
“Although the prevalence of OSA is relatively similar across North America, Europe, Australia and Asia, the burden of the disease is higher in subsets of the population, including overweight and obese people, racial minority groups and older people [52].”
BackgroundAgatonović-Kuštrin, S., & Beresford, R. (2000) · Journal of Pharmaceutical and Biomedical Analysis · doi:10.1016/s0731-7085(99)00272-1 · Cited by 1,987 · cited 3× · see citing works
“The weights or coefficients of the inputs are adjustable parameters and the transfer function allows for nonlinearity [29].”
BackgroundInfluentialCollop, N. A., Anderson, W. M., Boehlecke, B., Claman, D., Goldberg, R., Gottlieb, D. J., Hudgel, D., Sateia, M., Schwab, R., & Portable Monitoring Task Force of the American Academy of Sleep Medicine (2007) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.27032 · Cited by 1,550 · see citing works
Privacy in the age of medical big data
Price, W. N., & Cohen, I. G. (2019) · Nature Medicine · doi:10.1038/s41591-018-0272-7 · Cited by 1,522 · cited 1× · see citing works
“Anonymized open-access data may address the question of ownership and privacy, but as data sets proliferate, the ability to combine data sets re-identification may be possible [60].”
BackgroundThe Association between Sleep Apnea and the Risk of Traffic Accidents
Téran‐Santos, J., Jimenez-Gomez, A., & Cordero-Guevara, J. A. (1999) · New England Journal of Medicine · doi:10.1056/nejm199903183401104 · Cited by 1,252 · cited 1× · see citing works
“Significant untreated OSA is associated with increased risk of cardiovascular disease, metabolic disease, cancer, higher rates of health care use and both traffic and occupational accidents [3–5].”
BackgroundCollop, N. A., Anderson, W. M., Boehlecke, B. A., Claman, D. M., Goldberg, R., Gottlieb, D. J., Hudgel, D. W., Sateia, M. J., & Schwab, R. (2007) · PubMed Central · Cited by 1,082 · see citing works
Artificial intelligence in medicine
Ramesh, A., Kambhampati, C., Monson, J. Q., & Drew, P. J. (2004) · Annals of The Royal College of Surgeons of England · doi:10.1308/147870804290 · Cited by 1,008 · see citing works
Trends in CPAP adherence over twenty years of data collection: A flattened curve
Rotenberg, B. W., Murariu, D., & Pang, K. P. (2016) · Journal of Otolaryngology - Head and Neck Surgery · doi:10.1186/s40463-016-0156-0 · Cited by 753 · cited 1× · see citing works
“Despite machine improvements and compliance interventions, the non-adherence rates remain significantly high at 30–40% [23].”
BackgroundNagappa, M., Liao, P., Wong, J., Auckley, D., Ramachandran, S. K., Memtsoudis, S., Mokhlesi, B., & Chung, F. (2015) · PLOS ONE · doi:10.1371/journal.pone.0143697 · Cited by 684 · cited 2× · see citing works
“Current awareness and detection of OSA in the primary care setting is a major contributor to the disease process, as it is not routinely screened for and there are limitations to utilizing screening tools, like the STOP-Bang questionnaire, in the general public [2, 6].”
BackgroundMethod(2019) · Handbook of clinical neurology · doi:10.1016/b978-0-444-64032-1.00025-4 · Cited by 555 · see citing works
Metrics of sleep apnea severity: beyond the apnea-hypopnea index
Malhotra, A., Ayappa, I., Ayas, N., Collop, N., Kirsch, D., Mcardle, N., Mehra, R., Pack, A. I., Punjabi, N., White, D. P., et al. (2021) · SLEEP · doi:10.1093/sleep/zsab030 · Cited by 489 · cited 1× · see citing works
“AI offers an opportunity to consider more factors beyond AHI, which can provide more information about severity, OSA endo-phenotypes and can inform treatment selection [36, 37].”
BackgroundMyofunctional Therapy to Treat Obstructive Sleep Apnea: A Systematic Review and Meta-analysis
Camacho, M., Certal, V., Abdullatif, J. D., Zaghi, S., Ruoff, C. M., Capasso, R., & Kushida, C. A. (2015) · SLEEP · doi:10.5665/sleep.4652 · Cited by 420 · cited 1× · see citing works
“Myofunctional therapy (MT) is another treatment option which involves oro-pharyngeal exercises and has been shown to decrease AHI by 50% in adults [26, 27].”
BackgroundLung cancer classification using neural networks for CT images
Kuruvilla, J., & Gunavathi, K. (2013) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2013.10.011 · Cited by 366 · cited 1× · see citing works
“AI is currently being employed in various medical settings, including detecting ischemia in myocardial perfusion imaging, pulmonary embolism diagnosis, and classification of lung cancer with computerized tomography (CT) images [30–32].”
MethodEpidemiological aspects of obstructive sleep apnea.
Garvey, J., Pengo, M. F., Drakatos, P., & Kent, B. D. (2015) · Europe PMC (PubMed Central) · doi:10.3978/j.issn.2072-1439.2015.04.52 · Cited by 324 · cited 2× · see citing works
“Significant untreated OSA is associated with increased risk of cardiovascular disease, metabolic disease, cancer, higher rates of health care use and both traffic and occupational accidents [3–5].”
BackgroundSingh, R. P., Hom, G. L., Abramoff, M. D., Campbell, J. P., Chiang, M. F., & on behalf of the AAO Task Force on Artificial Intelligence (2020) · Translational Vision Science & Technology · doi:10.1167/tvst.9.2.45 · Cited by 259 · cited 6× · see citing works
“…medical school curriculum is an important step, yet there are barriers including the dense curriculum, the knowl-edgeable personnel required to teach such skills and the uncertainty surrounding how AI will impact the future of heath care and the scope of healthcare workers responsibilities [50].”
BackgroundInfluentialArtificial intelligence in digital pathology: a roadmap to routine use in clinical practice
Colling, R. T., Pitman, H., Oien, K. A., Rajpoot, N., Macklin, P. S., CM‐Path AI in Histopathology Working Group, Snead, D. R. J., Sackville, T., & Verrill, C. L. (2019) · The Journal of Pathology · doi:10.1002/path.5310 · Cited by 238 · cited 2× · see citing works
“Implementing AI into diagnosis and decision making is difficult when the “ground truth is subjective,” for example in Gleason grading of prostate cancer [55].”
BackgroundEl Shayeb, M., Topfer, L.-A., Stafinski, T., Pawluk, L., & Menon, D. (2014) · Canadian Medical Association Journal · doi:10.1503/cmaj.130952 · Cited by 237 · see citing works
Predictors of Continuous Positive Airway Pressure Adherence in Patients with Obstructive Sleep Apnea
Mehrtash, M., Bakker, J. P., & Ayas, N. (2019) · Lung · doi:10.1007/s00408-018-00193-1 · Cited by 231 · cited 1× · see citing works
“The factors that influence adherence should be considered in prescribing CPAP, including socio-demographic and psychosocial characteristics, disease severity and side effects [22].”
BackgroundDiagnosis of Obstructive Sleep Apnea by Peripheral Arterial Tonometry
Yalamanchali, S., Farajian, V., Hamilton, C., Pott, T. R., Samuelson, C. G., & Friedman, M. (2013) · JAMA Otolaryngology–Head & Neck Surgery · doi:10.1001/jamaoto.2013.5338 · Cited by 229 · see citing works
Artificial intelligence: opportunities and implications for the health workforce
Hazarika, I. (2020) · International Health · doi:10.1093/inthealth/ihaa007 · Cited by 203 · cited 1× · see citing works
“While AI can assist in diagnosis efficiency and accuracy, the physician’s role in the relationship remains important offering compassion and empathy [63].”
BackgroundArtificial Intelligence for the Otolaryngologist: A State of the Art Review
Bur, A. M., Shew, M. A., & New, J. (2019) · Otolaryngology · doi:10.1177/0194599819827507 · Cited by 172 · cited 4× · see citing works
“If AI systems are responsible in medical decision making, the population concerns may take precedent over individuals [49].”
BackgroundMethodInfluentialObstructive sleep apnea treatment in adults
Chang, H., Chen, Y., & Du, J. (2019) · The Kaohsiung Journal of Medical Sciences · doi:10.1002/kjm2.12130 · Cited by 146 · see citing works
Adult Obstructive Sleep Apnea/Hypopnea Syndrome: Definitions, Risk Factors, and Pathogenesis
Yaggi, H. K., & Strohl, K. P. (2010) · Clinics in Chest Medicine · doi:10.1016/j.ccm.2010.02.011 · Cited by 123 · cited 1× · see citing works
“The diagnosis of OSA is less reliant on the interpretation, observation and annotation by physicians, as it is based on clinical symptoms and elevated AHI [56].”
Clinician-Focused Overview and Developments in Polysomnography
Markun, L. C., & Sampat, A. (2020) · Current Sleep Medicine Reports · doi:10.1007/s40675-020-00197-5 · Cited by 117 · cited 1× · see citing works
“electrocardiogram (ECG), pulse oximetry, airflow and respiratory effort to report the Apnea–Hypopnea Index (AHI), which refers to the number of obstructions of airflow events divided by the sleep time [8, 9].”
BackgroundDiagnosis and Treatment of Obstructive Sleep Apnea in Adults
Semelka, M., Wilson, J., & Floyd, R. (2011) · PsycEXTRA Dataset · doi:10.1037/e676562012-001 · Cited by 114 · see citing works
Healthcare uses of artificial intelligence: Challenges and opportunities for growth
Racine, E., Boehlen, W., & Sample, M. (2019) · Healthcare Management Forum · doi:10.1177/0840470419843831 · Cited by 102 · cited 2× · see citing works
“The use of AI in healthcare challenges the ideology of privacy and confidentiality as “manageable, predictable and controllable,” as patients data is digitalized and utilized in complex algorithms [57].”
BackgroundMyofunctional therapy (oropharyngeal exercises) for obstructive sleep apnoea
Rueda, J.-R., Mugueta-Aguinaga, I., Vilaró, J., & Rueda-Etxebarria, M. (2020) · Cochrane Database of Systematic Reviews · doi:10.1002/14651858.cd013449.pub2 · Cited by 102 · see citing works
Neural Network Prediction of Obstructive Sleep Apnea From Clinical Criteria
Kirby, S. D., Danter, W. R., George, C. F., Francovic, T., Ferguson, K. A., Eng, P., & Ruby, R. R. F. (1999) · CHEST Journal · doi:10.1378/chest.116.2.409 · Cited by 101 · cited 7× · see citing works
“Unlike the two earliest ANNs developed for OSA diagnosis [33, 34], a more recent neural network developed with input from four variables obtained in the office setting, had a validation stage to determine network’s architecture [38].”
BackgroundMethodInfluentialSelf-citationHirshkowitz, M. (2016) · Sleep Medicine Clinics · doi:10.1016/j.jsmc.2016.07.002 · Cited by 100 · cited 1× · see citing works
“[ 4] Limitations associated with PSG include a significant cost, both financially and in regards to labor, facilities and specialists required [10].”
BackgroundRotenberg, B. W., Vicini, C., Pang, E. B., & Pang, K. P. (2016) · Journal of Otolaryngology - Head and Neck Surgery · doi:10.1186/s40463-016-0136-4 · Cited by 96 · cited 1× · see citing works
“A study showed how the implications of adherence may challenge CPAP as the gold-standard, and demonstrated that surgery should also be considered in the context of an individual’s anatomical features, OSA severity and treatment goals [24].”
BackgroundThe Value of Artificial Intelligence in Laboratory Medicine
Paranjape, K., Schinkel, M., Hammer, R., Schouten, B., Panday, R. S. N., Elbers, P., Kramer, M. H. H., & Nanayakkara, P. W. B. (2020) · American Journal of Clinical Pathology · doi:10.1093/ajcp/aqaa170 · Cited by 92 · cited 4× · see citing works
“Of the stakeholders surveyed, in an analysis of the value of AI in laboratory medicine, a substantial proportion of those who reported AI as valuable were unsure of reasons supporting this stance and 19% did not perceive the adoption of AI to be valuable in their organization [62].”
BackgroundInfluentialTreatment options for obstructive sleep apnea
Pavwoski, P., & Shelgikar, A. V. (2017) · Neurology Clinical Practice · doi:10.1212/cpj.0000000000000320 · Cited by 89 · see citing works
Screening and assessment for obstructive sleep apnea in primary care
Miller, J. N., & Berger, A. M. (2015) · Sleep Medicine Reviews · doi:10.1016/j.smrv.2015.09.005 · Cited by 81 · cited 1× · see citing works
“Current awareness and detection of OSA in the primary care setting is a major contributor to the disease process, as it is not routinely screened for and there are limitations to utilizing screening tools, like the STOP-Bang questionnaire, in the general public [2, 6].”
MethodBig Data in sleep apnoea: Opportunities and challenges
Pépin, J., Bailly, S., & Tamisier, R. (2019) · Respirology · doi:10.1111/resp.13669 · Cited by 79 · see citing works
Artificial Intelligence in Emergency Medicine: Surmountable Barriers With Revolutionary Potential
Grant, K. L., McParland, A., Mehta, S., & Ackery, A. (2020) · Annals of Emergency Medicine · doi:10.1016/j.annemergmed.2019.12.024 · Cited by 76 · cited 1× · see citing works
“Since AI systems change and ‘learn’ there is also a question of whether these systems require serial cycles of approval [54].”
BackgroundCritical Issues in Dental and Medical Management of Obstructive Sleep Apnea
Lavigne, G. J., Babiloni, A. H., Beetz, G., Fabbro, C. D., Sutherland, K., Huynh, N. T., & Cistulli, P. A. (2019) · Journal of Dental Research · doi:10.1177/0022034519885644 · Cited by 75 · cited 1× · see citing works
“Navigating the path from diagnosis to effective treatment is also challenging and requires a multidisciplinary approach and interprofessional collaboration, including sleep specialists, dental sleep medicine and physical therapists [7].”
BackgroundDetection of Sleep Apnea from Single-Lead ECG Signal Using a Time Window Artificial Neural Network
Wang, T., Lu, C., & Shen, G. (2019) · BioMed Research International · doi:10.1155/2019/9768072 · Cited by 75 · cited 1× · see citing works
“Physiological data from a singlelead ECG has also been utilized within an ANN to create an accessible, fast, portable method for diagnosis of sleep apnea [42].”
MethodIoachimescu, O. C., Allam, J. S., Samarghandi, A., Anand, N., Fields, B. G., Dholakia, S. A., Venkateshiah, S. B., Eisenstein, R., Ciavatta, M.-M., Collop, N. A., et al. (2020) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.8620 · Cited by 71 · cited 1× · see citing works
“WatchPAT accuracy has been improved by utilizing demographic and anthropometric data [16, 17].”
MethodPombo, N. G. C. C., García, N. M., & Bousson, K. (2017) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2017.01.001 · Cited by 62 · cited 1× · see citing works
“A limitation of this approach is “knowledge visualization,” as visualizing how the variables are interconnected and weighted within the network is virtually impossible [47].”
Cécile, P., Aude, M., Jean-Louis, D., & Catherine, R. (2020) · Studies in Health Technology and Informatics · doi:10.3233/shti200312 · Cited by 62 · cited 1× · see citing works
“” [53] Many of the ANNs that have been developed to detect OSA have required data from patient questionnaires reporting on their snoring status [38].”
BackgroundNikkonen, S., Afara, I. O., Leppänen, T., & Töyräs, J. (2019) · Scientific Reports · doi:10.1038/s41598-019-49330-7 · Cited by 58 · cited 1× · see citing works
“Another ANN solely utilized the blood oxygen saturation signal (SpO2) to predict AHI and oxygen desaturation index (ODI), classifying 90.4% and 94.4% of patient into the correct severity category based on estimated AHI and ODI respectively [41].”
MethodInfluentialValidity of Neural Network in Sleep Apnea
El-Solh, A., Mador, M. J., TenBrock, E., Shucard, D. W., Abul-Khoudoud, M. R., & Grant, B. J. B. (1999) · SLEEP · doi:10.1093/sleep/22.1.105 · Cited by 50 · cited 7× · see citing works
“Unlike logistical regressions, the relationships between the variables in ANNs are ambiguous and not easily understood [34].”
BackgroundMethodInfluentialSun, L. M., Chiu, H., Chuang, C. Y., & Liu, L. (2010) · Sleep And Breathing · doi:10.1007/s11325-010-0384-x · Cited by 50 · cited 1× · see citing works
“…developed a prediction model based on an AI system to detect sleep apnea from easily obtained information in a questionnaire without any biochemical data or use of other resources, resulting in a 81.8–88.0% sensitivity and 95–97% specificity, notably more accurate than a logistic regression [39].”
InfluentialHilmisson, H., Berman, S., & Magnúsdóttir, S. (2020) · Sleep And Breathing · doi:10.1007/s11325-020-02049-6 · Cited by 45 · see citing works
Turino, C., Benitez, I. D., Rafael-Palou, X., Mayoral, A., Lopera, A., Pascual, L., Vaca, R., Cortijo, A., Moncusí‐Moix, A., Dalmases, M., et al. (2021) · Journal of Medical Internet Research · doi:10.2196/24072 · Cited by 38 · cited 1× · see citing works
“A machine learning-based intelligent monitoring sys-tem has also been shown as a cost-efficient method to increase CPAP compliance, with an average increase of 1.14 h per day [45].”
MethodInfluentialA prediction model based on artificial neural networks for the diagnosis of obstructive sleep apnea
Karamanlı, H., Yalçinöz, T., Yalcinoz, M. A., & Yalcinoz, T. (2015) · Sleep And Breathing · doi:10.1007/s11325-015-1218-7 · Cited by 36 · cited 7× · see citing works
“However, in the diagnosis of OSA, physicians relying on their subjective intuition detects OSA with a specificity of 65% and sensitivity of 60%, drastically less than the ANNs that have been developed and discussed above [38].”
BackgroundMethodInfluentialStewart, S. A., Skomro, R., Reid, J. K., Penz, E. D., Fenton, M. E., Gjevre, J. A., & Cotton, D. J. (2015) · Canadian Respiratory Journal · doi:10.1155/2015/516580 · Cited by 35 · see citing works
Artificial intelligence in medicine: A matter of joy or concern?
Kolańska, K., Chabbert‐Buffet, N., Daraï, É., & Antoine, J.-M. (2020) · Journal of Gynecology Obstetrics and Human Reproduction · doi:10.1016/j.jogoh.2020.101962 · Cited by 33 · cited 1× · see citing works
“While the logic AI offers is superior and demonstrates high accuracy, human intuition is not replaceable but can collaborate with the logical advances AI offers [61].”
BackgroundA knowledge-based artificial neural network classifier for pulmonary embolism diagnosis
Serpen, G., Tekkedil, D. K., & Orra, M. (2007) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2007.10.001 · Cited by 32 · see citing works
Cost Minimization Using an Artificial Neural Network Sleep Apnea Prediction Tool for Sleep Studies
Teferra, R. A., Grant, B. J. B., Mindel, J. W., Siddiqi, T., IFTIKHAR, I. H., Ajaz, F., Aliling, J. P., Khan, M. S., Hoffmann, S. P., & Magalang, U. J. (2014) · Annals of the American Thoracic Society · doi:10.1513/annalsats.201404-161oc · Cited by 27 · cited 1× · see citing works
“A cost analysis study demonstrated that the ANN (OSUNet) was the most cost effective as a prediction tool compared to HSAT and other screening questionnaires [43].”
MethodNakajima, K., Matsuo, S., Wakabayashi, H., Yokoyama, K., Bunko, H., Okuda, K., Kinuya, S., Nyström, K., & Edenbrandt, L. (2015) · Circulation Journal · doi:10.1253/circj.cj-15-0079 · Cited by 26 · cited 1× · see citing works
“AI is currently being employed in various medical settings, including detecting ischemia in myocardial perfusion imaging, pulmonary embolism diagnosis, and classification of lung cancer with computerized tomography (CT) images [30–32].”
MethodDevelopment of a pediatric obstructive sleep apnea triage algorithm
Heath, D. S., El-Hakim, H., Al-Rahji, Y., Eksteen, E., Uwiera, T. C., Isaac, A., Castro‐Codesal, M. L., Gerdung, C. A., MacLean, J. E., & Mandhane, P. J. (2021) · Journal of Otolaryngology - Head and Neck Surgery · doi:10.1186/s40463-021-00528-8 · Cited by 18 · cited 1× · see citing works
“This concept of considering individual anatomy guiding treatment is also evident in a study which developed a pediatric OSA algorithm and demonstrated that tonsil size was the strongest predictor of adenotonsillectomy (AT) [25].”
BackgroundIoachimescu, O. C., Dholakia, S. A., Venkateshiah, S. B., Fields, B. G., Samarghandi, A., Anand, N., Eisenstein, R., Ciavatta, M.-M., Allam, J. S., & Collop, N. A. (2020) · Journal of Investigative Medicine · doi:10.1136/jim-2020-001448 · Cited by 17 · cited 1× · see citing works
“WatchPAT accuracy has been improved by utilizing demographic and anthropometric data [16, 17].”
MethodSeo, M. Y., Yoo, J. H., Hwang, S. J., & Lee, S. H. (2020) · Clinical and Experimental Otorhinolaryngology · doi:10.21053/ceo.2020.01984 · Cited by 12 · cited 1× · see citing works
“Cardiopulmonary coupling (CPC) is an algorithm derived from the autonomic and respiratory signals recorded by a wearable device, utilizing ECG or photo plethysmography (PPG) and has been shown to have a diagnostic accuracy comparable to PSG; 100% sensitivity and 93.63% specificity [18, 19].”
MethodInfluentialBelani, S., Tiarks, G. C., Mookerjee, N., & Rajput, V. (2021) · Cureus · doi:10.7759/cureus.18736 · Cited by 12 · cited 1× · see citing works
“Obtaining consent from patients often involves releasing their ownership of the information but the understanding of who owns and is responsible the data can be obscured as data can be transferred or bought by commercial companies [59].”
BackgroundPredicting optimal CPAP by neural network reduces titration failure: a randomized study
Solh, A. E., Akinnusi, M. E., Patel, A. N., Bhat, A., & TenBrock, R. (2009) · Sleep And Breathing · doi:10.1007/s11325-009-0247-5 · Cited by 9 · see citing works
Home Sleep Testing for the Diagnosis of Obstructive Sleep Apnea—Indications and Limitations
Cruz, S. D., Littner, M. R., & Zeidler, M. R. (2014) · Seminars in Respiratory and Critical Care Medicine · doi:10.1055/s-0034-1390066 · Cited by 9 · cited 2× · see citing works
“However it is not recommended in diagnosing patients with significant comorbid conditions or screening asymptomatic populations, and has been associated with high rates of signal loss and increased study failure [12–14].”
BackgroundAn introduction to artificial intelligence in sleep medicine
Lovejoy, C. A., Abbas, A., & Ratneswaran, D. (2021) · Journal of Thoracic Disease · doi:10.21037/jtd-21-1569 · Cited by 9 · cited 1× · see citing works
“AI offers an opportunity to consider more factors beyond AHI, which can provide more information about severity, OSA endo-phenotypes and can inform treatment selection [36, 37].”
BackgroundA Novel Artificial Neural Network Based Sleep-Disordered Breathing Screening Tool
Li, A., Quan, S. F., Silva, G. E., Perfect, M. M., & Roveda, J. M. (2018) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.7182 · Cited by 8 · cited 2× · see citing works
“, further validation in other populations would be required for screening in high risk groups [40].”
BackgroundMethodThe AHI is useful but limited: how can we do better?
Malhotra, A., & Gottlieb, D. J. (2021) · SLEEP · doi:10.1093/sleep/zsab150 · Cited by 7 · cited 1× · see citing works
“AHI is only one consideration and is limited in its ability to predict adverse effects and responsiveness to treatment [35].”
Background
Flagged first, then by citations
Cited by
Sorted by citations
Barriers to and facilitators of artificial intelligence adoption in health care: scoping review
Hassan, M., Kushniruk, A., & Borycki, E. (2024) · JMIR Human Factors · doi:10.2196/48633 · Cited by 324 · see citing works
“A governance model should have mechanisms and systems in place to facilitate changes impacting AI technologies in development or use based on cyclical changes in the technology or changes in the external landscape, such as the ones initiated by regulatory bodies [32,41,79].”
BackgroundMethodInfluentialZávodná, L. S., Überwimmer, M., & Frankus, E. (2024) · Journal of Economics and Management · doi:10.22367/jem.2024.46.13 · Cited by 61 · see citing works
“In the healthcare sector, Brennan (2022) categorized the barriers to AI implementation into several themes: technology, data, regulation, human resources, education, and culture.”
BackgroundJahrami, H., Husain, W., Trabelsi, K., Penzel, T., Hirshkowitz, M., Razjouyan, J., BaHammam, A. S., & Sharafkhaneh, A. (2025) · Sleep Medicine Reviews · doi:10.1016/j.smrv.2025.102212 · Cited by 16 · see citing works
Hoang, N. H., & Liang, Z. (2025) · Sensors · doi:10.3390/s25061698 · Cited by 15 · see citing works
“This decision is driven by the fact that, unlike deep learning, models developed using traditional machine learning techniques are often more interpretable, which makes them better suited for healthcare applications where transparency and explainability are important [27].”
Park, M. J., Choi, J. H., Kim, S. Y., & Ha, T. K. (2024) · Digital Health · doi:10.1177/20552076241291707 · Cited by 11 · see citing works
Unveiling the Influence of AI on Advancements in Respiratory Care: Narrative Review
Alqahtani, M. Μ., Alanazi, A., Algarni, S. S., Aljohani, H., Alenezi, F. K., Alotaibi, T. F., Alotaibi, M. M., Alqahtani, M. K., Alahmari, M. A., Alwadeai, K. S., et al. (2024) · Interactive Journal of Medical Research · doi:10.2196/57271 · Cited by 11 · see citing works
“Thus, addressing these barriers is crucial for enhancing future artificial intelligence applications in the field [120].”
Saba, L., Maindarkar, M. A., Khanna, N. N., Puvvula, A., Faa, G., Isenović, E. R., Johri, A. M., Fouda, M. M., Tiwari, E., Kalra, M. K., et al. (2024) · Reviews in Cardiovascular Medicine · doi:10.31083/j.rcm2512463 · Cited by 9 · see citing works
Lv, X., Li, J., & Cai, Y. (2024) · Current Psychology · doi:10.1007/s12144-024-07169-w · Cited by 8 · see citing works
“Presently, consumers are concerned and fearful of the “black box” nature of AI algorithmic decision-making and its limited interpretability (Brennan & Kirby, 2022 ).”
Artificial intelligence in primary care
Sherbini, A. E., Glicksberg, B. S., & Krittanawong, C. (2023) · Elsevier eBooks · doi:10.1016/b978-0-443-15688-5.00039-5 · Cited by 9 · see citing works
Hoang, N. H., & Liang, Z. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1510166 · Cited by 7 · see citing works
Adams, J. (2025) · AI & Society · doi:10.1007/s00146-025-02398-4 · Cited by 11 · see citing works
“Clinicians, meanwhile, require consistent and clinically relevant AI outputs because if a model’s performance fluctuates across different sleeping experiences or device types, it will be unreliable as a diagnostic tool (Brennan and Kirby 2022).”
Isaac, M., ElBeshlawy, D. M., Elsobki, A., Ahmed, D. F., & Kenawy, S. M. (2024) · Imaging Science in Dentistry · doi:10.5624/isd.20230249 · Cited by 4 · see citing works
Jouan, G., Arnardóttir, E. S., Íslind, A. S., & Óskarsdóttir, M. (2023) · European Journal of Operational Research · doi:10.1016/j.ejor.2023.09.039 · Cited by 4 · see citing works
Artificial Intelligence for Diagnosis of Obstructive Sleep Apnea
May, J., & Malkani, R. G. (2024) · Current Pulmonology Reports · doi:10.1007/s13665-024-00361-0 · Cited by 4 · see citing works
“These include ensuring access to the technology, educating clinicians and the public, maintenance of AI systems, adaptation to changing needs and treatments, addressing data security and privacy concerns, and regulation (for a review, see [16]).”
Esfandiari, E., Kalroozi, F., Mehrabi, N., & Hosseini, Y. (2024) · Jundishapur Journal of Chronic Disease Care · doi:10.5812/jjcdc-143796 · Cited by 3 · see citing works
Transforming sleep medicine: the evolving role of artificial intelligence
Banjade, P., Sharma, M., & Surani, S. (2025) · Current Opinion in Pulmonary Medicine · doi:10.1097/mcp.0000000000001210 · Cited by 2 · see citing works
Ma, Y., Li, H., Chen, H., Xu, J., Ye, Z., Li, Y., Wu, X., Chen, J., Cao, C., Zhang, P., et al. (2026) · Alimentary Pharmacology & Therapeutics · doi:10.1111/apt.70793 · Cited by 2 · see citing works
“More importantly, biomarker testing needs to be incorporated into clinical practice guidelines and integrated into existing clinical workflows [156].”
Artificial intelligence in the diagnosis and treatment of sleep apnea. First applications
Enchev, M., Georgieva, G., Milkov, M. P., & Baycheva, S. (2022) · Scripta Scientifica Medica · doi:10.14748/ssm.v54i0.9010 · Cited by 2 · see citing works
Perspective: A resident’s role in promoting safe machine-learning tools in sleep medicine
Smith, C. M., & Vendrame, M. (2023) · Journal of Clinical Sleep Medicine · doi:10.5664/jcsm.10724 · Cited by 2 · see citing works
Holm, B., Jouan, G., Hardarson, E., Sigurdardottir, S. E., Hoelke, K., Murphy, C., Arnardóttir, E. S., Óskarsdóttir, M., & Íslind, A. S. (2024) · Frontiers in Neuroinformatics · doi:10.3389/fninf.2024.1379932 · Cited by 2 · see citing works
Vigil, L., Zapata, T., Grau, A., Bonet, M., Montaña, M., & Piñar, M. (2024) · Open Respiratory Archives · doi:10.1016/j.opresp.2025.100402 · Cited by 2 · see citing works
Alber, B. R., Cheung, E. C., Russo, R., Jain, V., Schunke, K. J., Mendelowitz, D., & Kay, M. W. (2026) · Biomedicines · doi:10.3390/biomedicines14061263 · Cited by 1 · see citing works
“Regulatory considerations will also be important, especially for software as a “medical device” [137].”
Lin, W.-L., Chen, S.-H., & Fang, S.-S. (2026) · The Egyptian Journal of Otolaryngology · doi:10.1186/s43163-026-01169-2 · Cited by 1 · see citing works
“However, as Brennan and Kirby [56] noted in their analysis of barriers to AI implementation in OSA, challenges persist including the need for standardized data formats across different PSG systems, clear regulatory pathways for AI-scored sleep studies, and demonstration of cost-effectiveness before…”
Avanços na odontologia: o papel da inteligência artificial em diagnósticos precisos
Domingues, C. A. M., Ribeiro, M. E. D. R., Orlandi, L. E., Rodrigues, R., Pessanha, G. R. G., Fernandes, L. A., de Lima, D. C., & Gomes, H. D. S. (2024) · Contribuciones a las Ciencias Sociales · doi:10.55905/revconv.17n.2-296 · Cited by 1 · see citing works
Zan, X., Liu, F., Xian, X., & Pardalos, P. M. (2024) · Springer optimization and its applications · doi:10.1007/978-3-031-68263-6_1 · Cited by 1 · see citing works
Efficiently and securely handling multidimensional medical data of Obstructive Sleep Apnea
Hemrajani, P., Dhaka, V. S., & Rani, G. (2023) · International Conference on Computational Intelligence, Communication Technology and Networking (CICTN) · doi:10.1109/cictn57981.2023.10141078
Hasannejadasl, H., Offermann, C. J. W., Essink, E., Dekker, A. L. A. J., Roumen, C., & Fijten, R. (2023) · Research Square · doi:10.21203/rs.3.rs-3166678/v1
Identification of Barriers to AI Implementation in Medical Information Systems
Pilch, A., Zygała, R., Gryncewicz, W., Dyvak, M., & Melnyk, A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-66761-9_6
Al-Awami, S., Tanberg, W., Monegro, A. F., Covell, D. A., Martinot, J., & Al‐Jewair, T. (2024) · European Journal of Dentistry · doi:10.1055/s-0044-1795120
Research on Intelligent Diagnosis and Treatment of Sleep Apnea Syndrome Driven by Health Big Data
Cheng, J., & Bin, H. (2024) · International Conference on Ubiquitous Computing and Communications (IUCC) · doi:10.1109/iucc65928.2024.00117
EEG Sleep Stage Classification via Domain Similarity Detection and Trajectories in Riemannian Space
Wang, Y., & He, H. S. (2025) · Electronics · doi:10.3390/electronics14234604
Kanpa, V. N., D’Onofrio, D., Samanta, N., Tada, M., & Patel, S. (2025) · medRxiv · doi:10.64898/2025.12.10.25341937
Mallick, R., & Jain, V. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-44021-2.00006-3
Yang, Z., Wang, W. Z., Chen, Y., Delen, D., & Deveci, M. (2026) · International Journal of Production Economics · doi:10.1016/j.ijpe.2026.110146
From Algorithm to Augmentation: Redefining Human–AI Dynamics in Modern Healthcare
Gresbek, M., & Schirmer, H. (2026) · IntechOpen eBooks · doi:10.5772/intechopen.1017039
Chen, X., Li, C., Xu, H., Yin, S., & Sheng, B. (2026) · The Visual Computer · doi:10.1007/s00371-026-04586-9
Zaher, A., Prikshat, V., el aziz, J. A., & Vinot, D. (2026) · Journal of Health Organization and Management · doi:10.1108/jhom-03-2026-0364
“At the same time, effective integration of AI into workflows and collaboration between clinicians and AI systems are critical for translating adoption into practice (Brennan and Kirby, 2022; Chew and Achananuparp, 2022).”
Lenze, N. R., & Goldstein, C. A. (2026) · Sleep Medicine Clinics · doi:10.1016/j.jsmc.2026.08.020
Regulatory, Legal, and Ethical Challenges in Artificial Intelligence-powered Sleep Medicine
Trabelsi, K., Ammar, A., Pandi-Perumal, S. R., & Jahrami, H. (2026) · Sleep Medicine Clinics · doi:10.1016/j.jsmc.2026.08.021
- Authors
- Brennan, H. L., & Kirby, S. D.
- Affiliations
- Memorial University of Newfoundland (CA)
- Corresponding author
- Brennan, Hannah L.
- ORCID
- Brennan, Hannah L.
- Venue
- Journal of Otolaryngology - Head & Neck Surgery
- Publisher
- SAGE Publications
- Published
- 2022-01
- Volume
- 51
- Issue
- 1
- Pages
- 16-16
- ISSN
- 1916-0208 (ISSN-L)
- 1916-0216
- PMC
- PMC9036782
- OpenAlex
- W4224849966
- Cited by
- 39
- References
- 63
- Citations by year
- 2026: 10 · 2025: 7 · 2024: 16 · 2023: 5 · 2022: 1
- FWCI
- 1.00
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Journal of Otolaryngology - Head and Neck Surgery (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Informationszentrum Lebenswissenschaften (ZB MED) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Memorial University of Newfoundland and Labrador Summer Undergrad Research Award
- Topics
- Health Sciences › Medicine › Physiology › Obstructive Sleep Apnea Research
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- obstructive sleep apnea, artificial intelligence, AI implementation barriers, sleep apnea diagnosis, polysomnography, primary care, health data, regulation, human resources, medical education, technology adoption
- MeSH terms
- Artificial Intelligence, Humans, Polysomnography, Sleep Apnea, Obstructive
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
