Daryl J Wile1, Ranjit Ranawaya2, Zelma H T Kiss3. 1. Department of Clinical Neurosciences, Hotchkiss Brain Institute, University of Calgary, Canada. Electronic address: djwile@ucalgary.ca. 2. Department of Clinical Neurosciences, Hotchkiss Brain Institute, University of Calgary, Canada. Electronic address: ranjit.ranawaya@albertahealthservices.ca. 3. Department of Clinical Neurosciences, Hotchkiss Brain Institute, University of Calgary, Canada. Electronic address: zkiss@ucalgary.ca.
Abstract
BACKGROUND: Distinguishing the postural re-emergent tremor of Parkinson disease from essential tremor can be difficult clinically. Use of accelerometry to aid diagnosis is limited to laboratory settings. We sought to record and differentiate these tremors using a smart watch device in an outpatient clinic. NEW METHOD: 41 patients were enrolled. Recordings were made with a smart watch device on the predominantly affected hand (all patients), and simultaneously with an analog accelerometer (10 patients) with hands at rest and outstretched. Tremor peak frequency, peak power, and power of the first four harmonics was calculated and compared between the two devices. Mean power at the first four harmonics was calculated and used to classify tremor as parkinsonian or essential. Test characteristics were calculated to compare the device and clinical diagnoses. RESULTS: Mean harmonic peak power was both highly sensitive and specific for distinction of Parkinson disease postural tremor from essential tremor with an optimal threshold for our sample (sensitivity 90.9%, 95% CI 58.7-99.8%; specificity 100%, 95% CI 76.8-100%; Cohen's kappa=0.91, SE=0.08). COMPARISON WITH EXISTING METHODS: The smart watch and analog devices had nearly perfect concordance of peak frequency and proportional harmonic power. The smart watch recordings in clinic took 3-6 min. CONCLUSIONS: A smart watch device can provide accurate and diagnostically relevant information about postural tremor. Its portability and ease of use could help translate such techniques into routine clinic use or to the community.
BACKGROUND: Distinguishing the postural re-emergent tremor of Parkinson disease from essential tremor can be difficult clinically. Use of accelerometry to aid diagnosis is limited to laboratory settings. We sought to record and differentiate these tremors using a smart watch device in an outpatient clinic. NEW METHOD: 41 patients were enrolled. Recordings were made with a smart watch device on the predominantly affected hand (all patients), and simultaneously with an analog accelerometer (10 patients) with hands at rest and outstretched. Tremor peak frequency, peak power, and power of the first four harmonics was calculated and compared between the two devices. Mean power at the first four harmonics was calculated and used to classify tremor as parkinsonian or essential. Test characteristics were calculated to compare the device and clinical diagnoses. RESULTS: Mean harmonic peak power was both highly sensitive and specific for distinction of Parkinson disease postural tremor from essential tremor with an optimal threshold for our sample (sensitivity 90.9%, 95% CI 58.7-99.8%; specificity 100%, 95% CI 76.8-100%; Cohen's kappa=0.91, SE=0.08). COMPARISON WITH EXISTING METHODS: The smart watch and analog devices had nearly perfect concordance of peak frequency and proportional harmonic power. The smart watch recordings in clinic took 3-6 min. CONCLUSIONS: A smart watch device can provide accurate and diagnostically relevant information about postural tremor. Its portability and ease of use could help translate such techniques into routine clinic use or to the community.
Authors: Hernán A González Rojas; Pedro Chaná Cuevas; Enrique E Zayas Figueras; Salvador Cardona Foix; Antonio J Sánchez Egea Journal: Med Biol Eng Comput Date: 2017-10-23 Impact factor: 2.602
Authors: Christiana Ossig; Angelo Antonini; Carsten Buhmann; Joseph Classen; Ilona Csoti; Björn Falkenburger; Michael Schwarz; Jürgen Winkler; Alexander Storch Journal: J Neural Transm (Vienna) Date: 2015-08-08 Impact factor: 3.575
Authors: Xiaodong Yang; Syed Aziz Shah; Aifeng Ren; Dou Fan; Nan Zhao; Dongjian Cao; Fangming Hu; Masood Ur Rehman; Weigang Wang; Karen M Von Deneen; Jie Tian Journal: IEEE J Transl Eng Health Med Date: 2018-01-24 Impact factor: 3.316
Authors: Patrick Mcgurrin; James Mcnames; Tianxia Wu; Mark Hallett; Dietrich Haubenberger Journal: IEEE J Transl Eng Health Med Date: 2020-10-21 Impact factor: 3.316
Authors: Sergio Márquez-Sánchez; Israel Campero-Jurado; Daniel Robles-Camarillo; Sara Rodríguez; Juan M Corchado-Rodríguez Journal: Sensors (Basel) Date: 2021-05-12 Impact factor: 3.576