Redefining and Validating Digital Biomarkers as Fluid, Dynamic Multi-Dimensional Digital Signal Patterns

“Digital biomarker” is a term broadly and indiscriminately applied and often limited in its conceptualization to mimic well-established biomarkers as defined and approved by regulatory agencies such as the United States Food and Drug Administration (FDA). There is a practical urgency to revisit the definition of a digital biomarker and expand it beyond current methods of identification and validation. Restricting the promise of digital technologies within the realm of currently defined biomarkers creates a missed opportunity. A whole new field of prognostic and early diagnostic digital biomarkers driven by data science and artificial intelligence can break the current cycle of high healthcare costs and low health quality that is being driven by today's chronic disease detection and treatment approaches. This new class of digital biomarkers will be dynamic and require developing new FDA approval pathways and next-generation gold standards.

Speech as a Biomarker: Opportunities, Interpretability, and Challenges

Purpose: Over the past decade, the signal processing and machine learning literature has demonstrated notable advancements in automated speech processing with the use of artificial intelligence for medical assessment and monitoring (e.g., depression, dementia, and Parkinson's disease, among others). Meanwhile, the clinical speech literature has identified several interpretable, theoretically motivated measures that are sensitive to abnormalities in the cognitive, linguistic, affective, motoric, and anatomical domains. Both fields have, thus, independently demonstrated the potential for speech to serve as an informative biomarker for detecting different psychiatric and physiological conditions. However, despite these parallel advancements, automated speech biomarkers have not been integrated into routine clinical practice to date. Conclusions: In this article, we present opportunities and challenges for adoption of speech as a biomarker in clinical practice and research. Toward clinical acceptance and adoption of speech-based digital biomarkers, we argue for the importance of several factors such as robustness, specificity, diversity, and physiological interpretability of speech analytics in clinical applications.

Can detection and prediction models for Alzheimer’s Disease be applied to Prodromal Parkinson’s Disease using explainable artificial intelligence? A brief report on Digital Neuro Signatures.

Parkinson's disease (PD) is the fastest growing neurodegeneration and has a prediagnostic phase with a lot of challenges to identify clinical and laboratory biomarkers for those in the earliest stages or those 'at risk'. Despite the current research effort, further progress in this field hinges on the more effective application of digital biomarker and artificial intelligence applications at the prediagnostic stages of PD. It is of the highest importance to stratify such prediagnostic subjects that seem to have the most neuroprotective benefit from drugs. However, current initiatives to identify individuals at risk or in the earliest stages that might be candidates for future clinical trials are still challenging due to the limited accuracy and explainability of existing prediagnostic detection and progression prediction solutions. In this brief paper, we report on a novel digital neuro signature (DNS) for prodromal-PD based on selected digital biomarkers previously discovered on preclinical Alzheimer's disease. (AD). Our preliminary results demonstrated a standard DNS signature for both preclinical AD and prodromal PD, containing a ranked selection of features. This novel DNS signature was rapidly repurposed out of 793 digital biomarker features and selected the top 20 digital biomarkers that are predictive and could detect both the biological signature of preclinical AD and the biological mechanism of a-synucleinopathy in prodromal PD. The resulting model can provide physicians with a pool of patients potentially eligible for therapy and comes along with information about the importance of the digital biomarkers that are predictive, based on SHapley Additive exPlanations (SHAP). Similar initiatives could clarify the stage before and around diagnosis, enabling the field to push into unchartered territory at the earliest stages of the disease.

Statistical Validation of Cardiovascular Digital Biomarkers Towards Monitoring the Cardiac Risk in COPD: A Lyfas Case Study

Background: Mobile health (mHealth) is gaining popularity due to its pervasiveness. Lyfas is a smartphone-based optical biomarker instrument catering to mHealth. It captures the Pulse Rate Variability (PRV) and its associated digital biomarkers from the index finger capillary circulation using the principle of arterial photoplethysmography. PRV surrogates for the Cardiovascular Autonomic Modulation (CvAM) and provides a snapshot of psychophysiological homeostasis of the body. Objective: The paper investigates the roles of (a) physiological factors, e.g., Age, Duration of illness, Heart Rate (HR), Respiration Rate (RR), SpO2 level, and (b) popular digital biomarkers, such as SDNN, LF/HF, RMSSD, pNN50, SD1/SD2 to evaluate the cardiac risk. The paper hypothesizes that low FEV1, which is another physiological factor, plays a critical role in defining such risk. Method: A total of 50 males and females each, suffering from Chronic Obstructive Pulmonary Disease (COPD) took the Lyfas test after appropriate ethical measures. Data, thus collected by Lyfas had been statistically analyzed using histogram plots and Kolmogorov-Smirnov test for normality check, Pearson's Correlations (PC) to measure the strength of associations, and linear regressions to test the goodness of fit of the model. Results: Positive PCs are noted between (a) RMSSD and SDNN ('very high'-females: 0.86 and males: 0.91), (b) pNN50 and RMSSD (PC: moderate 0.46), (c) pNN50 and SDNN (PC: moderate 0.44), (d) Duration of illness and Age ('high'-females: 0.71 and males: 0.77), and (e) Age and RR ('high'-females: 0.67, males: 0.53). Negative PC is noted between (a) LF/HF and FEV1 ('moderately high'-males 0.42) and (b) LF/HF and SpO2 ('moderately high'-males 0.30). Although the R2 values are not so encouraging (most are < 0.5), yet, the models are statistically significant (p-values 0.0336; CI 95%). Conclusion: The paper concludes that Lyfas may be used to predict the cardiac risk in COPD patients based on the LF/HF values correlated to SpO2 and FEV1 levels.

An Aging Focused Unobtrusive and Privacy-Preserving Digital Behaviorome

Digital measures are increasingly used as objective health measures in remote-monitoring settings. In addition to their use in purely clinical research, such as in clinical trials, one promising application area for sensor-derived digital measures is in technology-assisted ageing and ageing-related research. In this context, digital measures may be used to measure the risk of certain adverse events such as falls, and also to provide novel research insights into ageing and ageing-related conditions, like cognitive impairment. While major emphasis has been placed on deriving one or more digital measures from wearable devices, a more holistic approach inspired by systems biology that leverages large, non-exhaustive sets of digital measures may prove highly beneficial. Such an approach would be useful if combined with modern big data approaches like machine learning. As such, extensive sets of digital measures, which may be referred to as digital behavioromes, could help characterise new phenotypes in deep phenotyping efforts. These measures could also assist in the discovery of novel digital biomarkers or in the creation of digital clinical outcome assessments. While clinical research into digital measures focuses primarily on measures derived from wearable devices, proven technology used for long-term remote monitoring of older adults is generally contactless, unobtrusive, and privacy-preserving. In this context, we introduce and describe a digital behaviorome: a large, non-exhaustive set of digital measures based entirely on contactless, unobtrusive, and privacy-preserving sensor technologies. We also demonstrate how such a behaviorome can be used to build digital clinical outcome assessments that are relevant to ageing and derived from machine learning. These outcomes included fall risk, frailty, mild cognitive impairment, and late-life depression. With the exception of late-life depression, all digital outcome assessments demonstrated a promising ability (ROC AUC ≥ 0.7) to discriminate between positive and negative health outcomes, often in the range of comparable work with wearable devices. Finally, we highlight the possibility of using these digital behaviorome-based outcome assessments to discover novel potential digital biomarkers for each outcome. Here, we found reasonable contributors but also some potentially interesting new candidates regarding fall risk and mild cognitive impairment.

Scoping review of digital biomarker-based studies (Preprint)

BACKGROUND Sensors and digital devices have revolutionized the process of measuring, collecting, and storing health data. Digital biomarkers are defined as objective, quantifiable, physiological, and behavioral measures contained in digital devices that are portable, wearable, implantable, or digestible. The clinical utility of digital biomarkers is being supported by an increasing body of research. OBJECTIVE The present study intends to investigate the scope of digital biomarker-based systematic reviews. METHODS The current scoping review was organized using PRISMA-ScR. Limiting the search to English full-text systematic reviews of digital biomarkers that included at least one randomized controlled trial involving a human population and reporting changes in participants' health status. PubMed and the Cochrane library were searched. Separately, two reviewers screened and selected records. In addition, the qualified papers' reference lists were examined for additional reviews. The World Health Organization's (WHO) classification systems for diseases (ICD-11), health interventions (ICHI), and bodily functions (ICF) were used to classify populations, interventions, and outcomes. RESULTS 66 reviews met the inclusion criteria, mostly were published by authors from the United States of America (18, 27.28%). The most prevalent disease areas were Circulatory System (n=12, 18.18%) and External Causes (n=12, 18.18%). 27 and 23 interventions were connected to health-related behaviors and the circulatory system, respectively. Looking after one's health (physical activity) (n=22) and demographic changes (mortality) (n=19) were the most commonly reported outcomes. A substantial number of digital devices, mostly in the form of wearables (n=39/66, 59.09 %) were employed as interventions (n=43/66, 65.15 %). Position sensors (n=33/66) and heart /pulse rate sensors (n=32/66) were identified as the most prevalent types of sensors utilized to capture digital biomarkers. CONCLUSIONS Digital biomarker clinical research encompasses a wide range of technologies, populations, interventions, and clinical outcomes, with cardiovascular and physical activity sensors being the most explored. This necessitates a more thorough examination of the strength and quality of evidence regarding the health consequences of digital biomarker-based therapy.

Remote Assessment of Depression Using Digital Biomarkers From Cognitive Tasks

We describe the design and evaluation of a sub-clinical digital assessment tool that integrates digital biomarkers of depression. Based on three standard cognitive tasks (D2 Test of Attention, Delayed Matching to Sample Task, Spatial Working Memory Task) on which people with depression have been known to perform differently than a control group, we iteratively designed a digital assessment tool that could be deployed outside of laboratory contexts, in uncontrolled home environments on computer systems with widely varying system characteristics (e.g., displays resolution, input devices). We conducted two online studies, in which participants used the assessment tool in their own homes, and completed subjective questionnaires including the Patient Health Questionnaire (PHQ-9)—a standard self-report tool for assessing depression in clinical contexts. In a first study (n = 269), we demonstrate that each task can be used in isolation to significantly predict PHQ-9 scores. In a second study (n = 90), we replicate these results and further demonstrate that when used in combination, behavioral metrics from the three tasks significantly predicted PHQ-9 scores, even when taking into account demographic factors known to influence depression such as age and gender. A multiple regression model explained 34.4% of variance in PHQ-9 scores with behavioral metrics from each task providing unique and significant contributions to the prediction.

Clinical validation of digital biomarkers for pediatric patients with asthma and cystic fibrosis - Potential for clinical trials and clinical care

BackgroundDigital biomarkers are a promising novel method to capture clinical data in a home-setting. However, clinical validation prior to implementation is of vital importance. The aim of this study was to clinically validate physical activity, heart rate, sleep and FEV1 as digital biomarkers measured by a smartwatch and portable spirometer in children with asthma and cystic fibrosis (CF).MethodsThis was a prospective cohort study including 60 children with asthma and 30 children with CF (age 6–16). Participants wore a smartwatch, performed daily spirometry at home and completed a daily symptom questionnaire for 28-days. Physical activity, heart rate, sleep and FEV1 were considered candidate digital endpoints. Data from 128 healthy children was used for comparison. Reported outcomes were compliance, difference between patients and controls, correlation with disease-activity and potential to detect clinical events. Analysis was performed with linear mixed effect models.ResultsMedian compliance was 88%. On average, patients exhibited lower physical activity and FEV1 compared to healthy children, whereas the heart rate of children with asthma was higher compared to healthy children. Days with a higher symptom score were associated with lower physical activity for children with uncontrolled asthma and CF. Furthermore, FEV1 was lower and (nocturnal) heart rate was higher for both patient groups on days with more symptoms. Candidate biomarkers and showed a distinct pattern before- and after a pulmonary exacerbation.ConclusionPortable spirometer- and smartwatch-derived digital biomarkers show promise as candidate endpoints for use in clinical trials or clinical care in pediatric lung disease.