Artificial intelligence-enabled detection and assessment of Parkinson's disease using nocturnal breathing signals
Yuzhe Yang, Yuan Yuan, Guo Zhang, Hao Wang, Ying-Cong Chen, Yingcheng Liu, Christopher G Tarolli, Daniel Crepeau, Jan Bukartyk, Mithri R Junna, Aleksandar Videnovic, Terry D Ellis, Melissa C Lipford, Ray Dorsey, Dina Katabi, Yuzhe Yang, Yuan Yuan, Guo Zhang, Hao Wang, Ying-Cong Chen, Yingcheng Liu, Christopher G Tarolli, Daniel Crepeau, Jan Bukartyk, Mithri R Junna, Aleksandar Videnovic, Terry D Ellis, Melissa C Lipford, Ray Dorsey, Dina Katabi
Abstract
There are currently no effective biomarkers for diagnosing Parkinson's disease (PD) or tracking its progression. Here, we developed an artificial intelligence (AI) model to detect PD and track its progression from nocturnal breathing signals. The model was evaluated on a large dataset comprising 7,671 individuals, using data from several hospitals in the United States, as well as multiple public datasets. The AI model can detect PD with an area-under-the-curve of 0.90 and 0.85 on held-out and external test sets, respectively. The AI model can also estimate PD severity and progression in accordance with the Movement Disorder Society Unified Parkinson's Disease Rating Scale (R = 0.94, P = 3.6 × 10-25). The AI model uses an attention layer that allows for interpreting its predictions with respect to sleep and electroencephalogram. Moreover, the model can assess PD in the home setting in a touchless manner, by extracting breathing from radio waves that bounce off a person's body during sleep. Our study demonstrates the feasibility of objective, noninvasive, at-home assessment of PD, and also provides initial evidence that this AI model may be useful for risk assessment before clinical diagnosis.
Conflict of interest statement
C.G.T. has received research support from NINDS and Biosensics. T.D.E. has received research support from the National Institutes of Health, Michael J. Fox Foundation and MedRhythms Inc. T.D.E. has received honorarium for speaking engagements, educational programming and/or outreach activities through the American Parkinson Disease Association, the Parkinson’s Foundation, the Movement Disorders Society and the American Physical Therapy Association. R.D. has received honoraria for speaking at American Academy of Neurology, American Neurological Association, Excellus BlueCross BlueShield, International Parkinson’s and Movement Disorders Society, National Multiple Sclerosis Society, Northwestern University, Physicians Education Resource, LLC, Stanford University, Texas Neurological Society and Weill Cornell; received compensation for consulting services from Abbott, Abbvie, Acadia, Acorda, Alzheimer’s Drug Discovery Foundation, Ascension Health Alliance, Bial-Biotech Investments, Inc., Biogen, BluePrint Orphan, California Pacific Medical Center, Caraway Therapeutics, Clintrex, Curasen Therapeutics, DeciBio, Denali Therapeutics, Eli Lilly, Grand Rounds, Huntington Study Group, medical-legal services, Mediflix, Medopad, Medrhythms, Michael J. Fox Foundation, MJH Holding LLC, NACCME, Neurocrine, NeuroDerm, Olson Research Group, Origent Data Sciences, Otsuka, Pear Therapeutic, Praxis, Prilenia, Roche, Sanofi, Seminal Healthcare, Spark, Springer Healthcare, Sunovion Pharma, Sutter Bay Hospitals, Theravance, University of California Irvine and WebMD; research support from Abbvie, Acadia Pharmaceuticals, Biogen, Biosensics, Burroughs Wellcome Fund, CuraSen, Greater Rochester Health Foundation, Huntington Study Group, Michael J. Fox Foundation, National Institutes of Health, Patient-Centered Outcomes Research Institute, Pfizer, PhotoPharmics, Safra Foundation and Wave Life Sciences; editorial services for Karger Publications; and ownership interests with Grand Rounds (second opinion service). D.K. received research funding from NIH and the Michael J. Fox Foundation. D.K. is a cofounder of Emerald Innovations, Inc., and serves on the scientific advisory board of Janssen and the data and analytics advisory board of Amgen. The remaining authors declare no competing interests.
© 2022. The Author(s).
Figures
References
- Dorsey ER, Sherer T, Okun MS, Bloem BR. The emerging evidence of the Parkinson pandemic. J. Parkinsons Dis. 2018;8:S3–S8. doi: 10.3233/JPD-181474.
- Marras C, et al. Prevalence of Parkinson’s disease across North America. NPJ Parkinson’s Dis. 2018;4:21. doi: 10.1038/s41531-018-0058-0.
- Yang W, et al. Current and projected future economic burden of Parkinson’s disease in the U.S. NPJ Parkinsons Dis. 2020;6:15. doi: 10.1038/s41531-020-0117-1.
- Armstrong MJ, Okun MS. Diagnosis and treatment of Parkinson disease: a review. JAMA. 2020;323:548–560. doi: 10.1001/jama.2019.22360.
- Delenclos M, Jones DR, McLean PJ, Uitti RJ. Biomarkers in Parkinson’s disease: advances and strategies. Parkinsonism Relat. Disord. 2016;22:S106–S110. doi: 10.1016/j.parkreldis.2015.09.048.
- Jankovic J. Parkinson’s disease: clinical features and diagnosis. J. Neurol. Neurosurg. Psychiatry. 2008;79:368–376. doi: 10.1136/jnnp.2007.131045.
- Hauser RA, et al. A home diary to assess functional status in patients with Parkinson’s disease with motor fluctuations and dyskinesia. Clin. Neuropharmacol. 2000;23:75–81. doi: 10.1097/00002826-200003000-00003.
- Goetz CG, et al. Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): scale presentation and clinimetric testing results. Mov. Disord. 2008;23:2129–2170. doi: 10.1002/mds.22340.
- Evers LJW, Krijthe JH, Meinders MJ, Bloem BR, Heskes TM. Measuring Parkinson’s disease over time: the real-world within-subject reliability of the MDS-UPDRS. Mov. Disord. 2019;34:1480–1487. doi: 10.1002/mds.27790.
- Regnault A, et al. Does the MDS-UPDRS provide the precision to assess progression in early Parkinson’s disease? Learnings from the Parkinson’s progression marker initiative cohort. J. Neurol. 2019;266:1927–1936. doi: 10.1007/s00415-019-09348-3.
- Ellis TD, et al. Identifying clinical measures that most accurately reflect the progression of disability in Parkinson disease. Parkinsonism Relat. Disord. 2016;25:65–71. doi: 10.1016/j.parkreldis.2016.02.006.
- Athauda D, Foltynie T. The ongoing pursuit of neuroprotective therapies in Parkinson disease. Nat. Rev. Neurol. 2015;11:25–40. doi: 10.1038/nrneurol.2014.226.
- Kieburtz K, Katz R, Olanow CW. New drugs for Parkinson’s disease: the regulatory and clinical development pathways in the United States. Mov. Disord. 2018;33:920–927. doi: 10.1002/mds.27220.
- Zhang J, et al. Longitudinal assessment of tau and amyloid beta in cerebrospinal fluid of Parkinson disease. Acta Neuropathol. 2013;126:671–682. doi: 10.1007/s00401-013-1121-x.
- Parnetti L, et al. Cerebrospinal fluid β-glucocerebrosidase activity is reduced in Parkinson’s disease patients. Mov. Disord. 2017;32:1423–1431. doi: 10.1002/mds.27136.
- Parnetti L, et al. CSF and blood biomarkers for Parkinson’s disease. Lancet Neurol. 2019;18:573–586. doi: 10.1016/S1474-4422(19)30024-9.
- Tang Y, et al. Identifying the presence of Parkinson’s disease using low-frequency fluctuations in BOLD signals. Neurosci. Lett. 2017;645:1–6. doi: 10.1016/j.neulet.2017.02.056.
- Parkinson J. An essay on the shaking palsy. 1817. J. Neuropsychiatry Clin. Neurosci. 2002;14:223–226, discussion 222. doi: 10.1176/jnp.14.2.223.
- Benarroch EE, Schmeichel AM, Low PA, Parisi JE. Depletion of ventromedullary NK-1 receptor-immunoreactive neurons in multiple system atrophy. Brain. 2003;126:2183–2190. doi: 10.1093/brain/awg220.
- Baille G, et al. Early occurrence of inspiratory muscle weakness in Parkinson’s disease. PLoS ONE. 2018;13:e0190400. doi: 10.1371/journal.pone.0190400.
- Wang Y, et al. Abnormal pulmonary function and respiratory muscle strength findings in Chinese patients with Parkinson’s disease and multiple system atrophy–comparison with normal elderly. PLoS ONE. 2014;9:e116123. doi: 10.1371/journal.pone.0116123.
- Torsney KM, Forsyth D. Respiratory dysfunction in Parkinson’s disease. J. R. Coll. Physicians Edinb. 2017;47:35–39. doi: 10.4997/jrcpe.2017.108.
- Pokusa M, Hajduchova D, Buday T, Kralova Trancikova A. Respiratory function and dysfunction in Parkinson-type neurodegeneration. Physiol. Res. 2020;69:S69–S79. doi: 10.33549/physiolres.934405.
- Baille G, et al. Ventilatory dysfunction in Parkinson’s disease. J. Parkinsons Dis. 2016;6:463–471. doi: 10.3233/JPD-160804.
- Seccombe LM, et al. Abnormal ventilatory control in Parkinson’s disease–further evidence for non-motor dysfunction. Respir. Physiol. Neurobiol. 2011;179:300–304. doi: 10.1016/j.resp.2011.09.012.
- Quan SF, et al. The Sleep Heart Health Study: design, rationale, and methods. Sleep. 1997;20:1077–1085.
- Adib, F., Mao, H., Kabelac, Z., Katabi, D. & Miller, R. C. Smart homes that monitor breathing and heart rate. In Proc. of the 33rd Annual ACM Conference on Human Factors in Computing Systems (eds Begole, B. et al.) 837–846 (ACM, 2015).
- Yue S, He H, Wang H, Rahul H, Katabi D. Extracting multi-person respiration from entangled RF signals. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 2018;2:86. doi: 10.1145/3214289.
- Yue S, Yang Y, Wang H, Rahul H, Katabi D. BodyCompass: monitoring sleep posture with wireless signals. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 2020;4:66. doi: 10.1145/3397311.
- Blackwell T, et al. Associations of sleep architecture and sleep-disordered breathing and cognition in older community-dwelling men: the Osteoporotic Fractures in Men Sleep Study. J. Am. Geriatr. Soc. 2011;59:2217–2225. doi: 10.1111/j.1532-5415.2011.03731.x.
- Guttman L. A basis for analyzing test-retest reliability. Psychometrika. 1945;10:255–282. doi: 10.1007/BF02288892.
- Hoehn MM, Yahr MD. Parkinsonism: onset, progression and mortality. Neurology. 1967;17:427–442. doi: 10.1212/WNL.17.5.427.
- He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. Computer Vision and Pattern Recognition (CVPR) (Eds. Bajcsy, R. et al.) 770–778 (IEEE, 2016).
- Zahed H, et al. The neurophysiology of sleep in Parkinson’s disease. Mov. Disord. 2021;36:1526–1542. doi: 10.1002/mds.28562.
- Brunner H, et al. Microstructure of the non-rapid eye movement sleep electroencephalogram in patients with newly diagnosed Parkinson’s disease: effects of dopaminergic treatment. Mov. Disord. 2002;17:928–933. doi: 10.1002/mds.10242.
- González-Naranjo, J. E. et al. Analysis of sleep macrostructure in patients diagnosed with Parkinson’s disease. Behav. Sci. (Basel). 9, 6 (2019).
- Soikkeli R, Partanen J, Soininen H, Pääkkönen A, Riekkinen P., Sr. Slowing of EEG in Parkinson’s disease. Electroencephalogr. Clin. Neurophysiol. 1991;79:159–165. doi: 10.1016/0013-4694(91)90134-P.
- Klassen BT, et al. Quantitative EEG as a predictive biomarker for Parkinson disease dementia. Neurology. 2011;77:118–124. doi: 10.1212/WNL.0b013e318224af8d.
- Railo, H. et al. Resting state EEG as a biomarker of Parkinson’s disease: influence of measurement conditions. Preprint at10.1101/2020.05.08.084343 (2020).
- Boser, B. E., Vapnik, V. N. & Guyon, I. M. Training algorithm margin for optimal classifiers. In COLT '92: Proc. Fifth Annual Workshop on Computational learning theory (ed. Haussler, D) 144–152 (Association for Computing Machinery, 1992).
- Yang, Y., Zha, K., Chen, Y-C., Wang, H. & Katabi, D. Delving into deep imbalanced regression. In Proc. 38th International Conference on Machine Learning, Vol. 139 (Eds. Meila, M. & Zhang, T.) 11842–11851 (PMLR, 2021).
- Willis AW, Schootman M, Evanoff BA, Perlmutter JS, Racette B. A. neurologist care in Parkinson disease: a utilization, outcomes, and survival study. Neurology. 2011;77:851–857. doi: 10.1212/WNL.0b013e31822c9123.
- Braak H, et al. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol. Aging. 2003;24:197–211. doi: 10.1016/S0197-4580(02)00065-9.
- Mestre TA, et al. Parkinson’s disease subtypes: critical appraisal and recommendations. J. Parkinsons. Dis. 2021;11:395–404. doi: 10.3233/JPD-202472.
- Zeng, Y. et al. MultiSense: enabling multi-person respiration sensing with commodity WiFi. In Proc. ACM Interactive, Mobile, Wearable Ubiquitous Technol. Vol. 4 (Ed. Santini, S.) (Association for Computing Machinery, New York, NY, USA, 2020).
- Lei, T., Zhang, Y., Wang, S. I., Dai, H. & Artzi, Y. Simple recurrent units for highly parallelizable recurrence. In Proc. 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP) (Eds. Riloff, E. et al.), 4470–4481 (Association for Computational Linguistics, 2018).
- Heck DH, et al. Breathing as a fundamental rhythm of brain function. Front. Neural Circuits. 2017;10:115. doi: 10.3389/fncir.2016.00115.
- Welch P. The use of fast Fourier Transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. IEEE Trans. Audio Electroacoust. 1967;15:70–73. doi: 10.1109/TAU.1967.1161901.
- Ronneberger, O., Fischer, P. & Brox, T. U-net: convolutional networks for biomedical image segmentation. In Proc.Med. Image Comput Comput Assist Interv. (MICCAI) (Ed. Santini, S.) 234–241 (Association for Computing Machinery, New York, NY, USA, 2015).
- Zhao M, et al. Learning sleep stages from radio signals: a conditional adversarial architecture. Proc. 34th Int. Conf. Mach. Learn. (ICML) 2017;70:4100–4109.
- Platt, J. in Advances in Large Margin Classifiers Vol. 10 (eds Smola, A. J. et al.) 61–74 (MIT Press, 1999).
- LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521:436–444. doi: 10.1038/nature14539.
- Noble WS. What is a support vector machine? Nat. Biotechnol. 2006;24:1565–1567. doi: 10.1038/nbt1206-1565.
- Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat. Med. 1998;17:857–872. doi: 10.1002/(SICI)1097-0258(19980430)17:8<857::AID-SIM777>;2-E.
Source: PubMed