Optimization and Validation of a Classification Algorithm for Assessment of Physical Activity in Hospitalized Patients

Accelerometer techniques for capturing human movement validated against direct observation: a scoping review

Objective.Accelerometers are devices commonly used to measure human physical activity and sedentary time. Accelerometer capabilities and analytical techniques have evolved rapidly, making it difficult for researchers to keep track of advances and best practices for data processing and analysis. The objective of this scoping review is to determine the existing methods for analyzing accelerometer data for capturing human movement which have been validated against the criterion measure of direct observation.Approach.This scoping review searched 14 academic and 5 grey databases. Two independent raters screened by title and abstract, then full text. Data were extracted using Microsoft Excel and checked by an independent reviewer.Mainresults.The search yielded 1039 papers and the final analysis included 115 papers. A total of 71 unique accelerometer models were used across a total of 4217 participants. While all studies underwent validation from direct observation, most direct observation occurred live (55%) or using recordings (42%). Analysis techniques included machine learning (ML) approaches (22%), the use of existing cut-points (18%), receiver operating characteristic curves to determine cut-points (14%), and other strategies including regressions and non-ML algorithms (8%).Significance.ML techniques are becoming more prevalent and are often used for activity identification. Cut-point methods are still frequently used. Activity intensity is the most assessed activity outcome; however, both the analyses and outcomes assessed vary by wear location. This scoping review provides a comprehensive overview of accelerometer analysis and validation techniques using direct observation and is a useful tool for researchers using accelerometers.

Validation of the Activ8 Activity Monitor for Monitoring Postures, Motions, Transfers, and Steps of Hospitalized Patients

Sedentary behaviors and low physical activity among hospitalized patients have detrimental effects on health and recovery. Wearable activity monitors are a promising tool to promote mobilization and physical activity. However, existing devices have limitations in terms of their outcomes and validity. The Activ8 device was optimized for the hospital setting. This study assessed the concurrent validity of the modified Activ8. Hospital patients performed an activity protocol that included basic (e.g., walking) and functional activities (e.g., room activities), with video recordings serving as the criterion method. The assessed outcomes were time spent walking, standing, upright, sedentary, and newly added elements of steps and transfers. Absolute and relative time differences were calculated, and Wilcoxon and Bland-Altman analyses were conducted. Overall, the observed relative time differences were lower than 2.9% for the basic protocol and 9.6% for the functional protocol. Statistically significant differences were detected in specific categories, including basic standing (p < 0.05), upright time (p < 0.01), and sedentary time (p < 0.01), but they did not exceed the predetermined 10% acceptable threshold. The modified Activ8 device is a valid tool for assessing body postures, motions, steps, and transfer counts in hospitalized patients. This study highlights the potential of wearable activity monitors to accurately monitor and promote PA among hospital patients.

A New Index for the Quantitative Evaluation of Surgical Invasiveness Based on Perioperative Patients' Behavior Patterns: Machine Learning Approach Using Triaxial Acceleration

BACKGROUND

The minimally invasive nature of thoracoscopic surgery is well recognized; however, the absence of a reliable evaluation method remains challenging. We hypothesized that the postoperative recovery speed is closely linked to surgical invasiveness, where recovery signifies the patient's behavior transition back to their preoperative state during the perioperative period.

OBJECTIVE

This study aims to determine whether machine learning using triaxial acceleration data can effectively capture perioperative behavior changes and establish a quantitative index for quantifying variations in surgical invasiveness.

METHODS

We trained 7 distinct machine learning models using a publicly available human acceleration data set as supervised data. The 3 top-performing models were selected to predict patient actions, as determined by the Matthews correlation coefficient scores. Two patients who underwent different levels of invasive thoracoscopic surgery were selected as participants. Acceleration data were collected via chest sensors for 8 hours during the preoperative and postoperative hospitalization days. These data were categorized into 4 actions (walking, standing, sitting, and lying down) using the selected models. The actions predicted by the model with intermediate results were adopted as the actions of the participants. The daily appearance probability was calculated for each action. The 2 differences between 2 appearance probabilities (sitting vs standing and lying down vs walking) were calculated using 2 coordinates on the x- and y-axes. A 2D vector composed of coordinate values was defined as the index of behavior pattern (iBP) for the day. All daily iBPs were graphed, and the enclosed area and distance between points were calculated and compared between participants to assess the relationship between changes in the indices and invasiveness.

RESULTS

Patients 1 and 2 underwent lung lobectomy and incisional tumor biopsy, respectively. The selected predictive model was a light-gradient boosting model (mean Matthews correlation coefficient 0.98, SD 0.0027; accuracy: 0.98). The acceleration data yielded 548,466 points for patient 1 and 466,407 points for patient 2. The iBPs of patient 1 were [(0.32, 0.19), (-0.098, 0.46), (-0.15, 0.13), (-0.049, 0.22)] and those of patient 2 were [(0.55, 0.30), (0.77, 0.21), (0.60, 0.25), (0.61, 0.31)]. The enclosed areas were 0.077 and 0.0036 for patients 1 and 2, respectively. Notably, the distances for patient 1 were greater than those for patient 2 ({0.44, 0.46, 0.37, 0.26} vs {0.23, 0.0065, 0.059}; P=.03 [Mann-Whitney U test]).

CONCLUSIONS

The selected machine learning model effectively predicted the actions of the surgical patients with high accuracy. The temporal distribution of action times revealed changes in behavior patterns during the perioperative phase. The proposed index may facilitate the recognition and visualization of perioperative changes in patients and differences in surgical invasiveness.

The Effect of a Smartphone App with an Accelerometer on the Physical Activity Behavior of Hospitalized Patients: A Randomized Controlled Trial

Inactive behavior is common in hospitalized patients. This study investigated the effectiveness of using a smartphone app with an accelerometer (Hospital Fit) in addition to usual care physiotherapy on increasing patients' physical activity (PA) behavior. A randomized controlled trial was performed at Maastricht University Medical Centre. Patients receiving physiotherapy while hospitalized at the department of Pulmonology or Internal Medicine were randomized to usual care physiotherapy or using Hospital Fit additionally. Daily time spent walking, standing, and upright (standing/walking) (min) and daily number of postural transitions were measured with an accelerometer between the first and last treatment. Multiple linear regression analysis was performed to determine the association between PA behavior and Hospital Fit use, corrected for functional independence (mILAS). Seventy-eight patients were included with a median (IQR) age of 63 (56-68) years. Although no significant effects were found, a trend was seen in favor of Hospital Fit. Effects increased with length of use. Corrected for functional independence, Hospital Fit use resulted in an average increase of 27.4 min (95% CI: -2.4-57.3) standing/walking on day five and 29.2 min (95% CI: -6.4-64.7) on day six compared to usual care. Hospital Fit appears valuable in increasing PA in functionally independent patients.

Identification and Quantification of Activities Common to Intensive Care Patients; Development and Validation of a Dual-Accelerometer-Based Algorithm

The aim of this study was to develop and validate an algorithm that can identify the type, frequency, and duration of activities common to intensive care (IC) patients. Ten healthy participants wore two accelerometers on their chest and leg while performing 14 activities clustered into four protocols (i.e., natural, strict, healthcare provider, and bed cycling). A video served as the reference standard, with two raters classifying the type and duration of all activities. This classification was reliable as intraclass correlations were all above 0.76 except for walking in the healthcare provider protocol, (0.29). The data of four participants were used to develop and optimize the algorithm by adjusting body-segment angles and rest-activity-threshold values based on percentage agreement (%Agr) with the reference. The validity of the algorithm was subsequently assessed using the data from the remaining six participants. %Agr of the algorithm versus the reference standard regarding lying, sitting activities, and transitions was 95%, 74%, and 80%, respectively, for all protocols except transitions with the help of a healthcare provider, which was 14-18%. For bed cycling, %Agr was 57-76%. This study demonstrated that the developed algorithm is suitable for identifying and quantifying activities common for intensive care patients. Knowledge on the (in)activity of these patients and their impact will optimize mobilization.

ProHealth eCoach: user-centered design and development of an eCoach app to promote healthy lifestyle with personalized activity recommendations

Background

Regular physical activity (PA), healthy habits, and an appropriate diet are recommended guidelines to maintain a healthy lifestyle. A healthy lifestyle can help to avoid chronic diseases and long-term illnesses. A monitoring and automatic personalized lifestyle recommendation system (i.e., automatic electronic coach or eCoach) with considering clinical and ethical guidelines, individual health status, condition, and preferences may successfully help participants to follow recommendations to maintain a healthy lifestyle. As a prerequisite for the prototype design of such a helpful eCoach system, it is essential to involve the end-users and subject-matter experts throughout the iterative design process.

Methods

We used an iterative user-centered design (UCD) approach to understend context of use and to collect qualitative data to develop a roadmap for self-management with eCoaching. We involved researchers, non-technical and technical, health professionals, subject-matter experts, and potential end-users in design process. We designed and developed the eCoach prototype in two stages, adopting different phases of the iterative design process. In design workshop 1, we focused on identifying end-users, understanding the user’s context, specifying user requirements, designing and developing an initial low-fidelity eCoach prototype. In design workshop 2, we focused on maturing the low-fidelity solution design and development for the visualization of continuous and discrete data, artificial intelligence (AI)-based interval forecasting, personalized recommendations, and activity goals.

Results

The iterative design process helped to develop a working prototype of eCoach system that meets end-user’s requirements and expectations towards an effective recommendation visualization, considering diversity in culture, quality of life, and human values. The design provides an early version of the solution, consisting of wearable technology, a mobile app following the “Google Material Design” guidelines, and web content for self-monitoring, goal setting, and lifestyle recommendations in an engaging manner between the eCoach app and end-users.

Conclusions

The adopted iterative design process brings in a design focus on the user and their needs at each phase. Throughout the design process, users have been involved at the heart of the design to create a working research prototype to improve the fit between technology, end-user, and researchers. Furthermore, we performed a technological readiness study of ProHealth eCoach against standard levels set by European Union (EU).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-022-08441-0.

Fine Detection of Human Motion During Activities of Daily Living as a Clinical Indicator for the Detection and Early Treatment of Chronic Diseases: The E-Mob Project

Methods to measure physical activity and sedentary behaviors typically quantify the amount of time devoted to these activities. Among patients with chronic diseases, these methods can provide interesting behavioral information, but generally do not capture detailed body motion and fine movement behaviors. Fine detection of motion may provide additional information about functional decline that is of clinical interest in chronic diseases. This perspective paper highlights the need for more developed and sophisticated tools to better identify and track the decomposition, structuration, and sequencing of the daily movements of humans. The primary goal is to provide a reliable and useful clinical diagnostic and predictive indicator of the stage and evolution of chronic diseases, in order to prevent related comorbidities and complications among patients.