Digital Biomarkers最新文献

Background: Alterations in multiple domains of cognition have been observed in individuals who have experienced a traumatic stressor. These domains may provide important insights in identifying underlying neurobiological dysfunction driving an individual's clinical response to trauma. However, such assessments are burdensome, costly, and time-consuming. To overcome barriers, efforts have emerged to measure multiple domains of cognitive functioning through the application of machine learning (ML) models to passive data sources.

Methods: We utilized automated computer vision and voice analysis methods to extract facial, movement, and speech characteristics from semi-structured clinical interviews in 81 trauma survivors who additionally completed a cognitive assessment battery. A ML-based regression framework was used to identify variance in visual and auditory measures that relate to multiple cognitive domains.

Results: Models derived from visual and auditory measures collectively accounted for a large variance in multiple domains of cognitive functioning, including motor coordination (R² = 0.52), processing speed (R² = 0.42), emotional bias (R² = 0.52), sustained attention (R² = 0.51), controlled attention (R² = 0.44), cognitive flexibility (R² = 0.43), cognitive inhibition (R² = 0.64), and executive functioning (R² = 0.63), consistent with the high test-retest reliability of traditional cognitive assessments. Face, voice, speech content, and movement have all significantly contributed to explaining the variance in predicting functioning in all cognitive domains.

Conclusions: The results demonstrate the feasibility of automated measurement of reliable proxies of cognitive functioning through low-burden passive patient evaluations. This makes it easier to monitor cognitive functions and to intervene earlier and at a lower threshold without requiring a time-consuming neurocognitive assessment by, for instance, a licensed psychologist with specialized training in neuropsychology.

Objective: Differentiating benign from dangerous causes of dizziness or vertigo presents a major diagnostic challenge for many clinicians. Bedside presentations of peripheral vestibular disorders and posterior fossa strokes are often indistinguishable other than by a few subtle vestibular eye movements. The most challenging of these to interpret is the head impulse test (HIT) of vestibulo-ocular reflex (VOR) function. There have been major advances in portable video-oculography (VOG) quantification of the video HIT (vHIT), but these specialized devices are not routinely available in most clinical settings. As a first step towards smartphone-based diagnosis of strokes in patients presenting vestibular symptoms, we sought proof of concept that we could use a smartphone application ("app") to accurately record the vHIT.

Methods: This was a cross-sectional agreement study comparing a novel index test (smartphone-based vHIT app) to an accepted reference standard test (VOG-based vHIT) for measuring VOR function. We recorded passive (examiner-performed) vHIT sequentially with both methods in a convenience sample of patients visiting an otoneurology clinic. We quantitatively correlated VOR gains (ratio of eye to head movements during the HIT) from each side/ear and experts qualitatively assessed the physiologic traces by the two methods.

Results: We recruited 11 patients; 1 patient's vHIT could not be reliably quantified with either device. The novel and reference test VOR gain measurements for each ear (n = 20) were highly correlated (Pearson's r = 0.9, p = 0.0000001) and, qualitatively, clinically equivalent.

Conclusions: This preliminary study provides proof of concept that an "eyePhone" app could be used to measure vHIT and eventually developed to diagnose vestibular strokes by smartphone.

Introduction: Digital biomarkers may act as a tool for early detection of changes in cognition. It is important to understand public perception of technologies focused on monitoring cognition to better guide the design of these tools and inform patients appropriately about the associated risks and benefits. Health care systems may also play a role in the clinical, legal, and financial implications of such technologies.

Objective: To evaluate public opinion on the use of passive technology for monitoring cognition.

Methods: This was a one-time, Internet-based survey conducted in English and Spanish.

Results: Within the English survey distributed in the USA (n = 173), 58.1% of respondents would be highly likely to agree to passive monitoring of cognition via a smartphone application. Thirty-eight percent of those with a higher degree of experience with technology were likely to agree to monitoring versus 20% of those with less experience with technology (p = 0.003). Sixty-two percent of non-health-care professionals were likely to agree to monitoring versus 45% of health-care workers (p = 0.012). There were significant concerns regarding privacy (p < 0.01). We compared the surveys answered in Spanish in Costa Rica via logistic regression (n = 43, total n = 216), adjusting for age, education level, health-care profession, owning a smartphone, experience with technology, and perception of cognitive decline. Costa Rican/Spanish-speaking respondents were 7 times more likely to select a high probability of agreeing to such a technology (p < 0.01). English-speaking respondents from the USA were 5 times more likely to be concerned about the impact on health insurance (p = 0.001) and life insurance (p = 0.01).

Conclusions: Understanding public perception and ethical implications should guide the design of digital biomarkers for cognition. Privacy and the health-care system in which the participants take part are 2 major factors to be considered. It is the responsibility of researchers to convey the ethical and legal implications of cognition monitoring.

Introduction: Changes in speech have the potential to provide important information on the diagnosis and progression of various neurological diseases. Many researchers have relied on open-source speech features to develop algorithms for measuring speech changes in clinical populations as they are convenient and easy to use. However, the repeatability of open-source features in the context of neurological diseases has not been studied.

Methods: We used a longitudinal sample of healthy controls, individuals with amyotrophic lateral sclerosis, and individuals with suspected frontotemporal dementia, and we evaluated the repeatability of acoustic and language features separately on these 3 data sets.

Results: Repeatability was evaluated using intraclass correlation (ICC) and the within-subjects coefficient of variation (WSCV). In 3 sets of tasks, the median ICC were between 0.02 and 0.55, and the median WSCV were between 29 and 79%.

Conclusion: Our results demonstrate that the repeatability of speech features extracted using open-source tool kits is low. Researchers should exercise caution when developing digital health models with open-source speech features. We provide a detailed summary of feature-by-feature repeatability results (ICC, WSCV, SE of measurement, limits of agreement for WSCV, and minimal detectable change) in the online supplementary material so that researchers may incorporate repeatability information into the models they develop.

Analyzing human gait with inertial sensors provides valuable insights into a wide range of health impairments, including many musculoskeletal and neurological diseases. A representative and reliable assessment of gait requires continuous monitoring over long periods and ideally takes place in the subjects' habitual environment (real-world). An inconsistent sensor wearing position can affect gait characterization and influence clinical study results, thus clinical study protocols are typically highly proscriptive, instructing all participants to wear the sensor in a uniform manner. This restrictive approach improves data quality but reduces overall adherence. In this work, we analyze the impact of altering the sensor wearing position around the waist on sensor signal and step detection. We demonstrate that an asymmetrically worn sensor leads to additional odd-harmonic frequency components in the frequency spectrum. We propose a robust solution for step detection based on autocorrelation to overcome sensor position variation (sensitivity = 0.99, precision = 0.99). The proposed solution reduces the impact of inconsistent sensor positioning on gait characterization in clinical studies, thus providing more flexibility to protocol implementation and more freedom to participants to wear the sensor in the position most comfortable to them. This work is a first step towards truly position-agnostic gait assessment in clinical settings.

Introduction: A major challenge in the monitoring of rehabilitation is the lack of long-term individual baseline data which would enable accurate and objective assessment of functional recovery. Consumer-grade wearable devices enable the tracking of individual everyday functioning prior to illness or other medical events which necessitate the monitoring of recovery trajectories.

Methods: For 1,324 individuals who underwent surgery on a lower limb, we collected their Fitbit device data of steps, heart rate, and sleep from 26 weeks before to 26 weeks after the self-reported surgery date. We identified subgroups of individuals who self-reported surgeries for bone fracture repair (n = 355), tendon or ligament repair/reconstruction (n = 773), and knee or hip joint replacement (n = 196). We used linear mixed models to estimate the average effect of time relative to surgery on daily activity measurements while adjusting for gender, age, and the participant-specific activity baseline. We used a sub-cohort of 127 individuals with dense wearable data who underwent tendon/ligament surgery and employed XGBoost to predict the self-reported recovery time.

Results: The 1,324 study individuals were all US residents, predominantly female (84%), white or Caucasian (85%), and young to middle-aged (mean age 36.2 years). We showed that 12 weeks pre- and 26 weeks post-surgery trajectories of daily behavioral measurements (steps sum, heart rate, sleep efficiency score) can capture activity changes relative to an individual's baseline. We demonstrated that the trajectories differ across surgery types, recapitulate the documented effect of age on functional recovery, and highlight differences in relative activity change across self-reported recovery time groups. Finally, using a sub-cohort of 127 individuals, we showed that long-term recovery can be accurately predicted, on an individual level, only 1 month after surgery (AUROC 0.734, AUPRC 0.8). Furthermore, we showed that predictions are most accurate when long-term, individual baseline data are available.

Discussion: Leveraging long-term, passively collected wearable data promises to enable relative assessment of individual recovery and is a first step towards data-driven intervention for individuals.

Background: Data derived from wearable activity trackers may provide important clinical insights into disease progression and response to intervention, but only if clinicians can interpret it in a meaningful manner. Longitudinal activity data can be visually presented in multiple ways, but research has failed to explore how clinicians interact with and interpret these visualisations. In response, this study developed a variety of visualisations to understand whether alternative data presentation strategies can provide clinicians with meaningful insights into patient's physical activity patterns.

Objective: To explore clinicians' opinions on different visualisations of actigraphy data.

Methods: Four visualisations (stacked bar chart, clustered bar chart, linear heatmap and radial heatmap) were created using Matplotlib and Seaborn Python libraries. A focus group was conducted with 14 clinicians across 2 hospitals. Focus groups were audio-recorded, transcribed and analysed using inductive thematic analysis.

Results: Three major themes were identified: (1) the importance of context, (2) interpreting the visualisations and (3) applying visualisations to clinical practice. Although clinicians saw the potential value in the visualisations, they expressed a need for further contextual information to gain clinical benefits from them. Allied health professionals preferred more granular, temporal information compared to doctors. Specifically, physiotherapists favoured heatmaps, whereas the remaining members of the team favoured stacked bar charts. Overall, heatmaps were considered more difficult to interpret.

Conclusion: The current lack of contextual data provided by wearables hampers their use in clinical practice. Clinicians favour data presented in a familiar format and yet desire multi-faceted filtering. Future research should implement user-centred design processes to identify ways in which all clinical needs can be met, potentially using an interactive system that caters for multiple levels of granularity. Irrespective of how data is displayed, unless clinicians can apply it in a manner that best supports their role, the potential of this data cannot be fully realised.