Prediction of 30-Day Readmission for COPD Patients Using Accelerometer-Based Activity Monitoring

Rationale and Design of Healthy at Home for COPD: an Integrated Remote Patient Monitoring and Virtual Pulmonary Rehabilitation Pilot Study

Chronic Obstructive Pulmonary Disease (COPD) is a common, costly, and morbid condition. Pulmonary rehabilitation, close monitoring, and early intervention during acute exacerbations of symptoms represent a comprehensive approach to improve outcomes, but the optimal means of delivering these services is uncertain. Logistical, financial, and social barriers to providing healthcare through face-to-face encounters, paired with recent developments in technology, have stimulated interest in exploring alternative models of care. The Healthy at Home study seeks to determine the feasibility of a multimodal, digitally enhanced intervention provided to participants with COPD longitudinally over six months. This paper details the recruitment, methods, and analysis plan for the study, which is recruiting 100 participants in its pilot phase. Participants were provided with several integrated services including a smartwatch to track physiological data, a study app to track symptoms and study instruments, access to a mobile integrated health program for acute clinical needs, and a virtual comprehensive pulmonary support service. Participants shared physiologic, demographic, and symptom reports, electronic health records, and claims data with the study team, facilitating a better understanding of their symptoms and potential care needs longitudinally. The Healthy at Home study seeks to develop a comprehensive digital phenotype of COPD by tracking and responding to multiple indices of disease behavior and facilitating early and nuanced responses to changes in participants' health status. This study is registered at Clinicaltrials.gov (NCT06000696).

Machine Learning-Based 30-Day Hospital Readmission Predictions for COPD Patients Using Physical Activity Data of Daily Living with Accelerometer-Based Device

Chronic obstructive pulmonary disease (COPD) is a significantly concerning disease, and is ranked highest in terms of 30-day hospital readmission. Generally, physical activity (PA) of daily living reflects the health status and is proposed as a strong indicator of 30-day hospital readmission for patients with COPD. This study attempted to predict 30-day hospital readmission by analyzing continuous PA data using machine learning (ML) methods. Data were collected from 16 patients with COPD over 3877 days, and clinical information extracted from the patients’ hospital records. Activity-based parameters were conceptualized and evaluated, and ML models were trained and validated to retrospectively analyze the PA data, identify the nonlinear classification characteristics of different risk factors, and predict hospital readmissions. Overall, this study predicted 30-day hospital readmission and prediction performance is summarized as two distinct approaches: prediction-based performance and event-based performance. In a prediction-based performance analysis, readmissions predicted with 70.35% accuracy; and in an event-based performance analysis, the total 30-day readmissions were predicted with a precision of 72.73%. PA data reflect the health status; thus, PA data can be used to predict hospital readmissions. Predicting readmissions will improve patient care, reduce the burden of medical costs burden, and can assist in staging suitable interventions, such as promoting PA, alternate treatment plans, or changes in lifestyle to prevent readmissions.

Effect of Self-Management Program on Health Status and Dyspnea in Patients with Chronic Obstructive Pulmonary Disease

Background

Self-management is becoming increasingly important in patients with chronic obstructive pulmonary disease (COPD) as it is associated with improved clinical outcomes. The aim of this study was to determine the effect of a self-management program on health status and dyspnea severity in patients with COPD.

Materials and Methods

This semi-experimental study was done on patients with COPD who were hospitalized at Ayatollah Kashani and Hajar hospitals from July 2017 to November 2019 in Shahrekord, Iran. Sixty patients with COPD were selected and completed three questionnaires, including a demographic questionnaire, COPD Assessment Test (CAT), and Dyspnea Scale (mMRC). The researcher educated the self-management skills to the intervention group, and the control group received routine care only. Three months after the intervention, the questionnaires were completed again by the control and intervention groups. Data were analyzed using descriptive and analytical statistics by SPSS version 24.

Results

The results of this study showed that there was no significant difference in health status between the two groups before the intervention (p=0.827). Three months after the intervention, the health status score in the control and intervention groups were 20.93±5.00 and 18.00±5.18, respectively, which this difference was statistically significant (p=0.030). Also, at the beginning of the study, the difference in dyspnea score between the two groups was not statistically significant (p=0.593). However, three months after the intervention, the score of dyspnea in the control and intervention groups was 2.1±0.995 and 1.53±1.074, respectively, which this difference was significant (p= 0.038).

Conclusion

This study showed that self-management intervention could improve the health status and reduce the dyspnea of patients with COPD.

A Low-Cost Calibration Method for Low-Cost MEMS Accelerometers Based on 3D Printing

A ubiquitous sensor in embedded systems is the accelerometer, as it enables a range of applications. However, accelerometers experience nonlinearities in their outputs caused by error terms and axes misalignment. These errors are a major concern because, in applications such as navigations systems, they accumulate over time, degrading the position accuracy. Through a calibration procedure, the errors can be modeled and compensated. Many methods have been proposed; however, they require sophisticated equipment available only in laboratories, which makes them complex and expensive. In this article, a simple, practical, and low-cost calibration method is proposed. It uses a 3D printed polyhedron, benefiting from the popularisation and low-cost of 3D printing in the present day. Additionally, each polyhedron could hold as much as 14 sensors, which can be calibrated simultaneously. The method was performed with a low-cost sensor and it significantly reduced the root-mean-square error (RMSE) of the sensor output. The RMSE was compared with the reported in similar proposals, and our method resulted in higher performance. The proposal enables accelerometer calibration at low-cost, and anywhere and anytime, not only by experts in laboratories. Compensating the sensor’s inherent errors thus increases the accuracy of its output.

Toward a Regulatory Qualification of Real-World Mobility Performance Biomarkers in Parkinson's Patients Using Digital Mobility Outcomes

Wearable inertial sensors can be used to monitor mobility in real-world settings over extended periods. Although these technologies are widely used in human movement research, they have not yet been qualified by drug regulatory agencies for their use in regulatory drug trials. This is because the first generation of these sensors was unreliable when used on slow-walking subjects. However, intense research in this area is now offering a new generation of algorithms to quantify Digital Mobility Outcomes so accurate they may be considered as biomarkers in regulatory drug trials. This perspective paper summarises the work in the Mobilise-D consortium around the regulatory qualification of the use of wearable sensors to quantify real-world mobility performance in patients affected by Parkinson's Disease. The paper describes the qualification strategy and both the technical and clinical validation plans, which have recently received highly supportive qualification advice from the European Medicines Agency. The scope is to provide detailed guidance for the preparation of similar qualification submissions to broaden the use of real-world mobility assessment in regulatory drug trials.

Spanish Costaleros' Physical Activity and Their Quality of Life

(1) Physical activity is one of the most influencing factors in people’ quality of life. Likewise, the costaleros of the Holy Week of Andalusia (Spain) carry out an important effort with high intensity during an extended time without any preparation. This study was the aim of knowing the intensity of the physical activity practiced by the costaleros in relation to their quality of life. (2) A transversal study was carried out with 1057 costaleros in Andalusia (Spain), where 930 were male and 127 female, between the ages of 18–61 years old (31.26 ± 7.60). For this purpose, descriptive, inferential, and correlative analyses were developed. Accelerometers (ActiGraph) were used during the procession to know the intensity of physical activity and the SF-36 test to know the self-perceived state of health and quality of life. (3) The intensity of physical activity practiced by costaleros is moderate, and it is related with their quality of life. In addition, positive associations are found between general health and physical activity. (4) Participants’ quality of life is associated with physical activity and freedom from injury. In addition, the measurement by accelerometry provides real data on the intensity of the effort made.

A role of color vision in emmetropization in C57BL/6J mice

Spectral composition affects emmetropization in both humans and animal models. Because color vision interacts the effects of chromatic defocus, we developed a method to bypass the effects of longitudinal chromatic aberration by placing a spectral filter behind the optics of the eye, using genetic tools. Newborn C57BL/6J (B6) mice were reared in quasi-monochromatic red (585–660 nm) or blue (410–510 nm) light beginning before eye-opening. Refractive states and ocular dimensions were compared at 4, 6, 8, and 10 weeks with mice reared in normal white light. Cre recombinase-dependent Ai9 reporter mice were crossed with Chx10-Cre to obtain Chx10-Cre;Ai9 mice, expressing red fluorescent protein in retinal Cre-positive cells. Ai9 offsprings, with and without Cre, were reared under a normal visual environment. Refraction and axial components were measured as described above. Expression levels of M and S opsin were quantified by western blotting at 10 weeks. Compared with those reared in white light, B6 mice reared in red light developed relative hyperopia, principally characterized by flattening of corneal curvature. Emmetropization was not affected by blue light, possibly because the reduction in vitreous chamber depth compensated for the increase in corneal curvature. Compared with Cre-negative littermates, the refraction and axial dimensions of Chx10-Cre;Ai9 mice were not significantly different at the follow-up timepoints. M opsin levels were higher in Chx10-Cre;Ai9 mice at 10 weeks while S opsin levels were not different. Red light induced a hyperopic shift in mouse refractive development. Emmetropization was not impacted in mice with perturbed color vision caused by intrinsic red-fluorescent protein, suggesting that color vision may not be necessary in mouse emmetropization when other mechanisms are present.