1
|
Bechny M, Monachino G, Fiorillo L, van der Meer J, Schmidt MH, Bassetti CLA, Tzovara A, Faraci FD. Bridging AI and Clinical Practice: Integrating Automated Sleep Scoring Algorithm with Uncertainty-Guided Physician Review. Nat Sci Sleep 2024; 16:555-572. [PMID: 38827394 PMCID: PMC11143488 DOI: 10.2147/nss.s455649] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/19/2023] [Accepted: 04/18/2024] [Indexed: 06/04/2024] Open
Abstract
Purpose This study aims to enhance the clinical use of automated sleep-scoring algorithms by incorporating an uncertainty estimation approach to efficiently assist clinicians in the manual review of predicted hypnograms, a necessity due to the notable inter-scorer variability inherent in polysomnography (PSG) databases. Our efforts target the extent of review required to achieve predefined agreement levels, examining both in-domain (ID) and out-of-domain (OOD) data, and considering subjects' diagnoses. Patients and Methods A total of 19,578 PSGs from 13 open-access databases were used to train U-Sleep, a state-of-the-art sleep-scoring algorithm. We leveraged a comprehensive clinical database of an additional 8832 PSGs, covering a full spectrum of ages (0-91 years) and sleep-disorders, to refine the U-Sleep, and to evaluate different uncertainty-quantification approaches, including our novel confidence network. The ID data consisted of PSGs scored by over 50 physicians, and the two OOD sets comprised recordings each scored by a unique senior physician. Results U-Sleep demonstrated robust performance, with Cohen's kappa (K) at 76.2% on ID and 73.8-78.8% on OOD data. The confidence network excelled at identifying uncertain predictions, achieving AUROC scores of 85.7% on ID and 82.5-85.6% on OOD data. Independently of sleep-disorder status, statistical evaluations revealed significant differences in confidence scores between aligning vs discording predictions, and significant correlations of confidence scores with classification performance metrics. To achieve κ ≥ 90% with physician intervention, examining less than 29.0% of uncertain epochs was required, substantially reducing physicians' workload, and facilitating near-perfect agreement. Conclusion Inter-scorer variability limits the accuracy of the scoring algorithms to ~80%. By integrating an uncertainty estimation with U-Sleep, we enhance the review of predicted hypnograms, to align with the scoring taste of a responsible physician. Validated across ID and OOD data and various sleep-disorders, our approach offers a strategy to boost automated scoring tools' usability in clinical settings.
Collapse
Affiliation(s)
- Michal Bechny
- Institute of Computer Science, University of Bern, Bern, Switzerland
- Institute of Digital Technologies for Personalized Healthcare (Meditech), University of Applied Sciences and Arts of Southern Switzerland, Lugano, Switzerland
| | - Giuliana Monachino
- Institute of Computer Science, University of Bern, Bern, Switzerland
- Institute of Digital Technologies for Personalized Healthcare (Meditech), University of Applied Sciences and Arts of Southern Switzerland, Lugano, Switzerland
| | - Luigi Fiorillo
- Institute of Digital Technologies for Personalized Healthcare (Meditech), University of Applied Sciences and Arts of Southern Switzerland, Lugano, Switzerland
| | | | - Markus H Schmidt
- Department of Neurology, University Hospital of Bern, Bern, Switzerland
- Ohio Sleep Medicine Institute, Dublin, OH, USA
| | | | - Athina Tzovara
- Institute of Computer Science, University of Bern, Bern, Switzerland
- Department of Neurology, University Hospital of Bern, Bern, Switzerland
| | - Francesca D Faraci
- Institute of Digital Technologies for Personalized Healthcare (Meditech), University of Applied Sciences and Arts of Southern Switzerland, Lugano, Switzerland
| |
Collapse
|
2
|
Trindade PAK, Nogueira VDSN, Weber SAT. Is maxillomandibular advancement an effective treatment for obstructive sleep apnea? Systematic literature review and meta-analysis. Braz J Otorhinolaryngol 2023; 89:503-510. [PMID: 37167845 DOI: 10.1016/j.bjorl.2023.02.007] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/06/2021] [Revised: 11/29/2022] [Accepted: 02/27/2023] [Indexed: 03/16/2023] Open
Abstract
OBJECTIVES To evaluate the effectiveness of maxillomandibular advancement surgery in the treatment of Obstructive Sleep Apnea by comparing the pre- and postoperative Apnea and Hypopnea Index, in addition to classifying the degree of evidence and risk of intervention bias. METHODS A systematic review of the literature was carried out in the PUBMED, LILACS, EMBASE, SCOPUS, WEB OF SCIENCE and COCHRANE platforms, including cohort studies with polysomnographic follow-up, without other associated pharyngeal or nasal surgical procedures. The risk of study bias was assessed using the Modified Delphi technique. Pre- and postoperative Apnea and Hypopnea Index data were plotted for meta-analysis, and the quality of evidence was assessed using the GRADE system. RESULTS Of 1882 references, 32 articles were selected for full-text reading, of which four studies were included, totaling 83 adults with obstructive sleep apnea who underwent maxillomandibular advancement. The meta-analysis was in favor of the intervention (DM = -33.36, 95% CI -41.43 to -25.29, p < 0.00001), with a mean percentage reduction in the Apnea and Hypopnea Index of 79.5% after surgery, even though the level of evidence was classified as very low quality by the GRADE system. CONCLUSION The meta-analysis was in favor of the intervention, characterizing maxillomandibular advancement surgery as an effective treatment for obstructive sleep apnea in adults.
Collapse
|
3
|
Jamal BT, Ibrahim EA. Satisfaction With Facial Aesthetic Appearance Following Maxillomandibular Advancement (MMA) for Obstructive Sleep Apnea (OSA): A Meta-Analysis. Cureus 2023; 15:e35568. [PMID: 37007354 PMCID: PMC10061353 DOI: 10.7759/cureus.35568] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 02/28/2023] [Indexed: 03/06/2023] Open
Abstract
PURPOSE A large cohort of patients diagnosed with obstructive sleep apnea (OSA) require surgical intervention, sometimes in the form of maxillomandibular advancement (MMA), to correct their functional disturbance. Such a surgical procedure typically results in a slight modification of the patients' facial appearance. The purpose of the current systematic review and meta-analysis was to examine the rate of satisfaction with facial aesthetics post-MMA intervention and to assess its dependability on and relationship with other patient or treatment factors. Based on the literature currently available, and to the best of our knowledge, this is the first paper to draw on the topic analytically. METHODS A search was conducted on four electronic literature databases (Pubmed, Ovid, Science Direct, and Scholar). Using referred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), our inclusion criterion covered any case with adequate reported data pertaining to the research question up to June 2021. Three evaluator groups were utilized. Satisfaction was defined as either an obvious reported increase in fondness for facial appearance or a state of indifference to the cosmetic results of the conducted changes. Dissatisfaction was defined as a clear discontent with the post-operative esthetic results. A multivariate analysis of the data was conducted, and Chi-square tests for independence were used to detect any significant associations. A meta-analysis of proportion was employed to permit for Freeman-Tukey double arcsine transformation and stabilize the variance of each study's proportion. Cochran's Q was computed, and the significance level was gauged as a function of P value. RESULTS Meta-analyses of proportion conducted for assessment of aesthetic appraisal following surgical MMA for OSA elucidated a significantly higher predilection towards aesthetic satisfaction after surgical MMA for OSA for all evaluator groups in the encompassed studies. 94.2% of patients were satisfied with their facial esthetics postoperatively. CONCLUSION The vast majority of patients that undergo MMA for the correction of OSA report satisfaction with post-surgical facial aesthetics. The subjective assessment of this parameter by physicians and laypeople portrays an equivalently significant skew toward post-surgical appearance improvement. MMA is a generally safe procedure that substantially contributes to enhancement of both overall quality of life and perceived aesthetic appeal.
Collapse
|
4
|
Stevner ABA, Vidaurre D, Cabral J, Rapuano K, Nielsen SFV, Tagliazucchi E, Laufs H, Vuust P, Deco G, Woolrich MW, Van Someren E, Kringelbach ML. Discovery of key whole-brain transitions and dynamics during human wakefulness and non-REM sleep. Nat Commun 2019; 10:1035. [PMID: 30833560 PMCID: PMC6399232 DOI: 10.1038/s41467-019-08934-3] [Citation(s) in RCA: 105] [Impact Index Per Article: 21.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/23/2018] [Accepted: 02/11/2019] [Indexed: 12/02/2022] Open
Abstract
The modern understanding of sleep is based on the classification of sleep into stages defined by their electroencephalography (EEG) signatures, but the underlying brain dynamics remain unclear. Here we aimed to move significantly beyond the current state-of-the-art description of sleep, and in particular to characterise the spatiotemporal complexity of whole-brain networks and state transitions during sleep. In order to obtain the most unbiased estimate of how whole-brain network states evolve through the human sleep cycle, we used a Markovian data-driven analysis of continuous neuroimaging data from 57 healthy participants falling asleep during simultaneous functional magnetic resonance imaging (fMRI) and EEG. This Hidden Markov Model (HMM) facilitated discovery of the dynamic choreography between different whole-brain networks across the wake-non-REM sleep cycle. Notably, our results reveal key trajectories to switch within and between EEG-based sleep stages, while highlighting the heterogeneities of stage N1 sleep and wakefulness before and after sleep. Sleep is composed of a number of different stages, each associated with a different pattern of brain activity. Here, using a data-driven Hidden Markov Model (HMM) of fMRI data, the authors discover a more complex set of neural activity states underlying the conventional stages of non-REM sleep.
Collapse
Affiliation(s)
- A B A Stevner
- Department of Psychiatry, University of Oxford, Warneford Hospital, OX3 7JX, Oxford, UK. .,Center of Functionally Integrative Neuroscience (CFIN), Aarhus University, 8000, Aarhus, Denmark. .,Center for Music in the Brain (MIB), Department of Clinical Medicine, Aarhus University, 8000, Aarhus, Denmark.
| | - D Vidaurre
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Human Brain Activity (OHBA), University of Oxford, Warneford Hospital, OX3 7JX, Oxford, UK
| | - J Cabral
- Department of Psychiatry, University of Oxford, Warneford Hospital, OX3 7JX, Oxford, UK.,Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, 4710-057, Braga, Portugal
| | - K Rapuano
- Department of Psychological and Brain Sciences, Dartmouth College, 03755, Hanover, NH, USA
| | - S F V Nielsen
- Department of Applied Mathematics and Computer Science, Technical University of Denmark, 2800 Kgs., Lyngby, Denmark
| | - E Tagliazucchi
- Netherlands Institute for Neuroscience, 1105 BA, Amsterdam, The Netherlands.,Department of Neurology, University Hospital Schleswig Holstein, Christian-Alrbrechts-Universität, 24105, Kiel, Germany.,Department of Neurology and Brain Imaging Center, Goethe University, 60528, Frankfurt am Main, Germany
| | - H Laufs
- Department of Neurology, University Hospital Schleswig Holstein, Christian-Alrbrechts-Universität, 24105, Kiel, Germany.,Department of Neurology and Brain Imaging Center, Goethe University, 60528, Frankfurt am Main, Germany
| | - P Vuust
- Center for Music in the Brain (MIB), Department of Clinical Medicine, Aarhus University, 8000, Aarhus, Denmark
| | - G Deco
- Center for Brain and Cognition, Computational Neuroscience Group, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Roc Boronat 138, Barcelona, 08018, Spain.,Institució Catalana de la Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys 23, Barcelona, 08010, Spain.,Department of Neuropsychology, Max Planck Institute for Human Cognitive and Brain Sciences, 04103, Leipzig, Germany.,School of Psychological Sciences, Monash University, Melbourne, Clayton, VIC, 3800, Australia
| | - M W Woolrich
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Human Brain Activity (OHBA), University of Oxford, Warneford Hospital, OX3 7JX, Oxford, UK
| | - E Van Someren
- Netherlands Institute for Neuroscience, 1105 BA, Amsterdam, The Netherlands.,Departments of Integrative Neurophysiology and Psychiatry GGZ-InGeest, Amsterdam Neuroscience, VU University and Medical Center, 1081 HV, Amsterdam, The Netherlands
| | - M L Kringelbach
- Department of Psychiatry, University of Oxford, Warneford Hospital, OX3 7JX, Oxford, UK.,Center of Functionally Integrative Neuroscience (CFIN), Aarhus University, 8000, Aarhus, Denmark.,Center for Music in the Brain (MIB), Department of Clinical Medicine, Aarhus University, 8000, Aarhus, Denmark.,Life and Health Sciences Research Institute (ICVS), School of Health Sciences, University of Minho, 4710-057, Braga, Portugal
| |
Collapse
|