Smartphone-Based Tapping Frequency as a Surrogate for Perceived Fatigue

Abstract

Fatigue is a common symptom in various diseases, including multiple sclerosis (MS). The current standard method to assess fatigue is through questionnaires, which has several shortcomings; questionnaires are subjective, prone to recall bias, and potentially confounded by other symptoms like stress and depression. Thus, there is an unmet medical need to develop objective and reliable methods to evaluate fatigue. Our study seeks to develop an objective and ubiquitous monitoring tool for assessing fatigue. Leveraging a smartphone-based rapid tapping task, we conducted a two-week in-the-wild study with 35 MS patients. We explore the association between tapping derived metrics and perceived fatigue assessed with two standard clinical scales: fatigue severity scale (FSS) and fatigue scale for motor and cognitive function (FSMC). Our novel smartphone-based fatigue metric, mean tapping frequency, objectively ranks perceived fatigue with a mean AUCROC = .76, CI = [.71, .81] according to the FSMC, and a mean AUCROC = .81, CI = [.76, .86] according to the FSS. These results demonstrate that our approach is feasible and valid in uncontrolled environments. In this work, we provide a promising tool for objective fatigue monitoring to be used in clinical trials and routine medical care.

Goal-directed upper limb movement patterns and hand grip forces in multiple sclerosis

“…Recently, tablet- and smartphone-based approaches that promise frequent, unsupervised, and remote assessments of hand and finger movements have been validated in pwMS. 30 – 32 In the future, it needs to be evaluated how such approaches compare to highly standardized assessments based on robotic devices that also include proximal movements and can record hand grip forces.…”

Beyond Accuracy: A Critical Review of Fairness in Machine Learning for Mobile and Wearable Computing

“…Similarly, Zhou et al [155] employed a fairness-aware client selection mechanism for federated learning to ensure equal representation for subjects with worse connectivity. 7 Post data collection, Su et al [125] performed data balancing, conditioned on the sensitive attribute, managing to narrow the impact of gender voice differences on their speech recognition model. Similarly, a strand of work explored data splitting, conditioned on the sensitive attribute (gender, age, BMI, skin tone, country, and health condition) to enable model personalization [59,77,91,125,150].…”

“…Gender biases have been reported in monitoring sleep posture with wireless signals [148], opioid usage tracking [46], diaphragmatic breathing monitor based on acoustic signals [43], and speech recognition via accelerometer sensors [125]. Age biases have been reported in medication adherence monitoring through gait assessment [150], fatigue estimation via smartphone tapping frequency [7], mobility purpose and route choice inference [100], and neural activation prediction [55]. Biases based on physiological measurements have been reported by Li et al [71] in fine-grained activity sensing (e.g., eye blinking, finger tracking) using acoustic signals against people of small stature, by Wang et al [136] in vital sign monitoring through acoustic sensing against obese or overweight people, and by Griffiths et al [44] in image processing with binocular thermal cameras against people of non-average height.…”

Modeling multiple sclerosis using mobile and wearable sensor data

“…MS is a chronic disease, with symptoms often worsening over time. The most occurring and troubling symptom of MS is fatigue , which refers to the “ subjective sensations of weariness, increasing sense of effort, mismatch between effort expended and actual performance, or exhaustion ” 3 – 5 . Recurring fatigue leads to low productivity, sick leave, and work disability 6 , 7 .…”

“…Such devices have already been used to monitor the fatigability 5 , 18 , fatigue 4 , EDSS level 19 – 22 and other outcomes of PwMS 23 . For instance, Motl et al use a two-minute walk test 19 and the timed 25-foot walk test 20 , which reflect the walking disability level, to approximate the EDSS level.…”