Detection of Flares by Decrease in Physical Activity, Collected Using Wearable Activity Trackers in Rheumatoid Arthritis or Axial Spondyloarthritis: An Application of Machine Learning Analyses in Rheumatology

被引:100
作者
Gossec, Laure [1 ,2 ]
Guyard, Frederic [3 ]
Leroy, Didier [3 ]
Lafargue, Thomas [3 ]
Seiler, Michel [4 ]
Jacquemin, Charlotte [1 ,2 ]
Molto, Anna [5 ]
Sellam, Jerernie [1 ]
Foltz, Violaine [1 ,2 ]
Gandjbakhch, Frederique [1 ,2 ]
Hudry, Christophe [6 ]
Mitrovic, Stephane [1 ,2 ]
Fautrel, Bruno [1 ,2 ]
Servy, Herve [7 ]
机构
[1] Sorbonne Univ, INSERM UMRS 938, Paris, France
[2] Hop La Pitie Salpetriere, AP HP, Paris, France
[3] Orange IMT, Nice, France
[4] Orange Healthcare, Paris, France
[5] Paris Descartes Univ, Cochin Hosp, AP HP, PRES Sorbonne Paris Cite,INSERM U1153, Paris, France
[6] Cochin Hosp, AP HP, Paris, France
[7] Sanoia, E Hlth Serv, Gemenos, France
关键词
ANKYLOSING-SPONDYLITIS; VALIDATION; PATIENT; HEALTH; INDEX; FREQUENCY; BASDAI; IMPACT; ASDAS; ASAS;
D O I
10.1002/acr.23768
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Objective Flares in rheumatoid arthritis (RA) and axial spondyloarthritis (SpA) may influence physical activity. The aim of this study was to assess longitudinally the association between patient-reported flares and activity-tracker-provided steps per minute, using machine learning. Methods This prospective observational study (ActConnect) included patients with definite RA or axial SpA. For a 3-month time period, physical activity was assessed continuously by number of steps/minute, using a consumer grade activity tracker, and flares were self-assessed weekly. Machine-learning techniques were applied to the data set. After intrapatient normalization of the physical activity data, multiclass Bayesian methods were used to calculate sensitivities, specificities, and predictive values of the machine-generated models of physical activity in order to predict patient-reported flares. Results Overall, 155 patients (1,339 weekly flare assessments and 224,952 hours of physical activity assessments) were analyzed. The mean +/- SD age for patients with RA (n = 82) was 48.9 +/- 12.6 years and was 41.2 +/- 10.3 years for those with axial SpA (n = 73). The mean +/- SD disease duration was 10.5 +/- 8.8 years for patients with RA and 10.8 +/- 9.1 years for those with axial SpA. Fourteen patients with RA (17.1%) and 41 patients with axial SpA (56.2%) were male. Disease was well-controlled (Disease Activity Score in 28 joints mean +/- SD 2.2 +/- 1.2; Bath Ankylosing Spondylitis Disease Activity Index score mean +/- SD 3.1 +/- 2.0), but flares were frequent (22.7% of all weekly assessments). The model generated by machine learning performed well against patient-reported flares (mean sensitivity 96% [95% confidence interval (95% CI) 94-97%], mean specificity 97% [95% CI 96-97%], mean positive predictive value 91% [95% CI 88-96%], and negative predictive value 99% [95% CI 98-100%]). Sensitivity analyses were confirmatory. Conclusion Although these pilot findings will have to be confirmed, the correct detection of flares by machine-learning processing of activity tracker data provides a framework for future studies of remote-control monitoring of disease activity, with great precision and minimal patient burden.
引用
收藏
页码:1336 / 1343
页数:8
相关论文
共 49 条
[1]  
Agresti A., 2002, Categorical data analysis
[2]   Rheumatoid arthritis near remission: clinical rather than laboratory inflammation is associated with radiographic progression [J].
Aletaha, Daniel ;
Alasti, Farideh ;
Smolen, Josef S. .
ANNALS OF THE RHEUMATIC DISEASES, 2011, 70 (11) :1975-1980
[3]   Feasibility and Domain Validation of Rheumatoid Arthritis (RA) Flare Core Domain Set: Report of the OMERACT 2014 RA Flare Group Plenary [J].
Bartlett, Susan J. ;
Bykerk, Vivian P. ;
Cooksey, Roxanne ;
Choy, Ernest H. ;
Alten, Rieke ;
Christensen, Robin ;
Furst, Daniel E. ;
Guillemin, Francis ;
Halls, Serena ;
Hewlett, Sarah ;
Leong, Amye L. ;
Lyddiatt, Anne ;
March, Lyn ;
Montie, Pamela ;
Orbai, Ana Maria ;
Pohl, Christoph ;
Voshaar, Marieke Scholte ;
Woodworth, Thasia G. ;
Bingham, Clifton O., III .
JOURNAL OF RHEUMATOLOGY, 2015, 42 (11) :2185-2189
[4]  
Boulle M, 2008, EXTRACTION GESTION C, P229
[5]  
Boullé M, 2007, J MACH LEARN RES, V8, P1659
[6]   MODL:: A Bayes optimal discretization method for continuous attributes [J].
Boulle, Marc .
MACHINE LEARNING, 2006, 65 (01) :131-165
[7]   The effect of physical activity and motivation on function in ankylosing spondylitis: A cohort study [J].
Brophy, Sinead ;
Cooksey, Roxanne ;
Davies, Helen ;
Dennis, Michael S. ;
Zhou, Shang-Ming ;
Siebert, Stefan .
SEMINARS IN ARTHRITIS AND RHEUMATISM, 2013, 42 (06) :619-626
[8]   Identifying flares in rheumatoid arthritis: reliability and construct validation of the OMERACT RA Flare Core Domain Set [J].
Bykerk, Vivian P. ;
Bingham, Clifton O. ;
Choy, Ernest H. ;
Lin, Daming ;
Alten, Rieke ;
Christensen, Robin ;
Furst, Daniel E. ;
Hewlett, Sarah ;
Leong, Amye ;
March, Lyn ;
Woodworth, Thasia ;
Boire, Gilles ;
Haraoui, Boulos ;
Hitchon, Carol ;
Jamal, Shahin ;
Keystone, Edward C. ;
Pope, Janet ;
Tin, Diane ;
Thorne, J. Carter ;
Bartlett, Susan J. .
RMD OPEN, 2016, 2 (01)
[9]   Flares in Rheumatoid Arthritis: Frequency and Management. A Report from the BRASS Registry [J].
Bykerk, Vivian P. ;
Shadick, Nancy ;
Frits, Michelle ;
Bingham, Clifton O., III ;
Jeffery, Iain ;
Iannaccone, Christine ;
Weinblatt, Michael ;
Solomon, Daniel H. .
JOURNAL OF RHEUMATOLOGY, 2014, 41 (02) :227-234
[10]  
Claudepierre P, 1997, J RHEUMATOL, V24, P1954