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

Abstract

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.

© 2018, American College of Rheumatology.

References

Cookies must be enabled

Enable cookies for pubmed.ncbi.nlm.nih.gov and reload this page to continue.

const challengeId = "uyEPy93Y4_cZHS8s46hd9g"; const challengeDomain = "pubmed.ncbi.nlm.nih.gov"; const bits = 8; const maxMs = Math.max(30000, 50 * 100); const diagnosticMode = false; function normalizedHostname(value) { return String(value || "").trim().replace(/^\.+/, "").toLowerCase(); } function cookieDomainForCurrentHost(domain) { const host = normalizedHostname(window.location.hostname); const target = normalizedHostname(domain); if (!target || !host) return null; return host === target || host.endsWith(`.${target}`) ? target : null; } function cookieString(name, value, maxAge) { const domain = cookieDomainForCurrentHost(challengeDomain); if (domain === null) return ""; const parts = [ `${encodeURIComponent(name)}=${encodeURIComponent(value)}`, `Max-Age=${maxAge}`, "Path=/", "SameSite=Lax" ]; parts.push(`Domain=${domain}`); if (window.location.protocol === "https:") parts.push("Secure"); return parts.join("; "); } function cookieHasValue(name, value) { const expected = `${encodeURIComponent(name)}=${encodeURIComponent(value)}`; return document.cookie.split(";").some(part => part.trim() === expected); } function browserAcceptsCookies() { if (navigator.cookieEnabled === false) return false; const suffix = String(challengeId).replace(/[^A-Za-z0-9]/g, "").slice(0, 32) || "probe"; const name = `__identity_pow_cookie_check_${suffix}`; const value = String(Date.now()); const setCookie = cookieString(name, value, 60); const clearCookie = cookieString(name, "", 0); if (!setCookie || !clearCookie) return false; try { document.cookie = setCookie; const accepted = cookieHasValue(name, value); document.cookie = clearCookie; return accepted; } catch (_error) { return false; } } function showCookiesRequired() { const message = document.getElementById("cookie-required"); if (message) message.hidden = false; document.body.classList.add("cookies-required"); } function leadingZeroBits(bytes) { let total = 0; for (const byte of bytes) { if (byte === 0) { total += 8; continue; } for (let bit = 7; bit >= 0; bit--) { if ((byte & (1 << bit)) === 0) total += 1; else return total; } } return total; } async function sha256Bytes(text) { return new Uint8Array(await crypto.subtle.digest("SHA-256", new TextEncoder().encode(text))); } async function submitSolve(nonce) { const redirectUrl = diagnosticMode ? `${window.location.href.split("#", 1)[0]}#pow-complete` : window.location.href; const response = await fetch("/_pow/solve", { method: "POST", cache: "no-store", credentials: "same-origin", headers: {"Content-Type": "application/json"}, body: JSON.stringify({ challenge_id: challengeId, nonce: String(nonce), redirect_url: redirectUrl }) }); const result = await response.json(); if (result.success) { if (diagnosticMode || result.diagnostic === true) { window.location.hash = "pow-complete"; } else { finishNavigation(result.redirect_url || window.location.href); } return true; } return false; } function finishNavigation(targetUrl) { try { const current = new URL(window.location.href); const target = new URL(targetUrl || current.href, current.href); if (target.origin === current.origin && target.pathname === current.pathname && target.search === current.search) { if (target.hash && target.hash !== current.hash) { window.location.hash = target.hash; } window.location.reload(); return; } window.location.replace(target.href); } catch (_error) { window.location.reload(); } } async function run() { if (!browserAcceptsCookies()) { showCookiesRequired(); return; } const started = Date.now(); for (let nonce = 0; Date.now() - started < maxMs; nonce++) { const digest = await sha256Bytes(`${challengeId}:${nonce}`); if (leadingZeroBits(digest) >= bits) { if (await submitSolve(nonce)) return; break; } if (nonce % 100 === 0) { await new Promise(resolve => setTimeout(resolve, 0)); } } if (diagnosticMode) return; window.location.reload(); } run().catch(() => { if (!diagnosticMode) window.location.reload(); });