Digital Biomarkers最新文献

Introduction: The menstrual cycle (MC) reflects multifaceted hormonal changes influencing women's metabolism, making it a key aspect of women's health. Changes in hormonal levels throughout the MC have been demonstrated to influence various physiological parameters, including exhaled carbon dioxide (CO₂). Lumen is a small handheld device that measures metabolic fuel usage via exhaled CO₂. This study leverages exhaled CO₂ patterns measured by the Lumen device to elucidate metabolic variations during the MC, which may hold significance for fertility management. Additionally, CO₂ changes are explored in menopausal women with and without hormonal replacement therapy (HRT).

Methods: This retrospective cohort study analyzed exhaled CO₂ data from 3,981 Lumen users, including eumenorrheal women and menopausal women with and without HRT. Linear mixed models assessed both CO₂ changes of eumenorrheal women during the MC phases and compared between menopausal women with or without HRT.

Results: Eumenorrheic women displayed cyclical CO₂ patterns during the MC, characterized by elevated levels during the menstrual, estrogenic and ovulation phases and decreased levels during post-ovulation and pre-menstrual phases. Notably, despite variations in cycle length affecting the timing of maximum and minimum CO₂ levels within a cycle, the overall pattern remained consistent. Furthermore, CO₂ levels in menopausal women without HRT differed significantly from those with HRT, which showed lower levels.

Conclusion: This study reveals distinct CO₂ patterns across MC phases, providing insights into hormonal influences on metabolic activity. Menopausal women exhibit altered CO₂ profiles in relation to the use or absence of HRT. CO₂ monitoring emerges as a potential tool for tracking the MC and understanding metabolic changes during menopause.

Background: The prevalence of neurological disorders is increasing, underscoring the importance of objective gait analysis to help clinicians identify specific deficits. Nevertheless, existing technological solutions for gait analysis often suffer from impracticality in daily clinical use, including excessive cost, time constraints, and limited processing capabilities.

Summary: This review aims to evaluate existing techniques for clustering patients with the same neurological disorder to assist clinicians in optimizing treatment options. A narrative review of thirteen relevant studies was conducted, characterizing their methods, and evaluating them against seven criteria. Additionally, the results are summarized in two comprehensive tables. Recent approaches show promise; however, our results indicate that, overall, only three approaches display medium or high process maturity, and only two show high clinical applicability.

Key messages: Our findings highlight the necessity for advancements, specifically regarding the use of markerless optical tracking systems, the optimization of experimental plans, and the external validation of results. This narrative review provides a comprehensive overview of existing clustering techniques, bridging the gap between instrumented gait analysis and its real-world clinical utility. We encourage researchers to use our findings and those from other medical fields to enhance clustering techniques for patients with neurological disorders, facilitating the identification of disparities within groups and their extent, ultimately improving patient outcomes.

Introduction: Given the traffic safety and occupational injury prevention implications associated with cannabis impairment, there is a need for objective and validated measures of recent cannabis use. Pupillary light response may offer an approach for detection.

Method: Eighty-four participants (mean age: 32, 42% female) with daily, occasional, and no-use cannabis use histories participated in pupillary light response tests before and after smoking cannabis ad libitum or relaxing for 15 min (no use). The impact of recent cannabis consumption on trajectories of the pupillary light response was modeled using functional data analysis tools. Logistic regression models for detecting recent cannabis use were compared, and average pupil trajectories across cannabis use groups and times since light test administration were estimated.

Results: Models revealed small, significant differences in pupil response to light after cannabis use comparing the occasional use group to the no-use control group, and similar statistically significant differences in pupil response patterns comparing the daily use group to the no-use comparison group. Trajectories of pupillary light response estimated using functional data analysis found that acute cannabis smoking was associated with less initial and sustained pupil constriction compared to no cannabis smoking.

Conclusion: These analyses show the promise of pairing pupillary light response and functional data analysis methods to assess recent cannabis use.

Background: Fatigue is a prominent symptom in many diseases and is strongly associated with impaired daily function. The measurement of daily function is currently almost always done with questionnaires, which are subjective and imprecise. With the recent advances of digital wearable technologies, novel approaches to evaluate daily function quantitatively and objectively in real-life conditions are increasingly possible. This also creates new possibilities to measure fatigue-related changes of daily function using such technologies.

Summary: This review examines which digitally assessable parameters in immune-mediated inflammatory and neurodegenerative diseases may have the greatest potential to reflect fatigue-related changes of daily function.

Key messages: Results of a standardized analysis of the literature reporting about perception-, capacity-, and performance-evaluating assessment tools indicate that changes of the following parameters: physical activity, independence of daily living, social participation, working life, mental status, cognitive and aerobic capacity, and supervised and unsupervised mobility performance have the highest potential to reflect fatigue-related changes of daily function. These parameters thus hold the greatest potential for quantitatively measuring fatigue in representative diseases in real-life conditions, e.g., with digital wearable technologies. Furthermore, to the best of our knowledge, this is a new approach to analysing evidence for the design of performance-based digital assessment protocols in human research, which may stimulate further systematic research in this area.

Introduction: Image-based machine learning holds great promise for facilitating clinical care; however, the datasets often used for model training differ from the interventional clinical trial-based findings frequently used to inform treatment guidelines. Here, we draw on longitudinal imaging of psoriasis patients undergoing treatment in the Ultima 2 clinical trial (NCT02684357), including 2,700 body images with psoriasis area severity index (PASI) annotations by uniformly trained dermatologists.

Methods: An image-processing workflow integrating clinical photos of multiple body regions into one model pipeline was developed, which we refer to as the "One-Step PASI" framework due to its simultaneous body detection, lesion detection, and lesion severity classification. Group-stratified cross-validation was performed with 145 deep convolutional neural network models combined in an ensemble learning architecture.

Results: The highest-performing model demonstrated a mean absolute error of 3.3, Lin's concordance correlation coefficient of 0.86, and Pearson correlation coefficient of 0.90 across a wide range of PASI scores comprising disease classifications of clear skin, mild, and moderate-to-severe disease. Within-person, time-series analysis of model performance demonstrated that PASI predictions closely tracked the trajectory of physician scores from severe to clear skin without systematically over- or underestimating PASI scores or percent changes from baseline.

Conclusion: This study demonstrates the potential of image processing and deep learning to translate otherwise inaccessible clinical trial data into accurate, extensible machine learning models to assess therapeutic efficacy.