A Machine Learning Classification Model for Monitoring the Daily Physical Behaviour of Lower-Limb Amputees

There are currently limited data on how prosthetic devices are used to support lower-limb prosthesis users in their free-living environment. Possessing the ability to monitor a patient’s physical behaviour while using these devices would enhance our understanding of the impact of different prosthetic products. The current approaches for monitoring human physical behaviour use a single thigh or wrist-worn accelerometer, but in a lower-limb amputee population, we have the unique opportunity to embed a device within the prosthesis, eliminating compliance issues. This study aimed to develop a model capable of accurately classifying postures (sitting, standing, stepping, and lying) by using data from a single shank-worn accelerometer. Free-living posture data were collected from 14 anatomically intact participants and one amputee over three days. A thigh worn activity monitor collected labelled posture data, while a shank worn accelerometer collected 3-axis acceleration data. Postures and the corresponding shank accelerations were extracted in window lengths of 5–180 s and used to train several machine learning classifiers which were assessed by using stratified cross-validation. A random forest classifier with a 15 s window length provided the highest classification accuracy of 93% weighted average F-score and between 88 and 98% classification accuracy across all four posture classes, which is the best performance achieved to date with a shank-worn device. The results of this study show that data from a single shank-worn accelerometer with a machine learning classification model can be used to accurately identify postures that make up an individual’s daily physical behaviour. This opens up the possibility of embedding an accelerometer-based activity monitor into the shank component of a prosthesis to capture physical behaviour information in both above and below-knee amputees. The models and software used in this study have been made open source in order to overcome the current restrictions of applying activity monitoring methods to lower-limb prosthesis users.

Using motivational interviewing combined with digital shoe-fitting to improve adherence to wearing orthopedic shoes in people with diabetes at risk of foot ulceration: study protocol for a cluster-randomized controlled trial

Background Diabetic foot ulcers have a high impact on mobility and daily functioning and lead to high treatment costs, for example, by hospitalization and amputation. To prevent (re)ulcerations, custom-made orthopedic shoes are considered essential. However, adherence to wearing the orthopedic shoes is low, and improving adherence was not successful in the past. We propose a novel care approach that combines motivational interviewing (MI) with a digital shoe-fitting procedure to improve adherence to orthopedic shoes. The aim of this trial is to assess the (cost-)effectiveness of this novel care approach compared to usual care (no MI and casting-based shoe-fitting) in promoting footwear adherence and ulcer prevention. Methods The trial will include people with diabetes, with IWGDF Risk categories 1–3, who have been prescribed orthopedic shoes. Participants will be randomized at the level of the podiatrist to the novel care approach or usual care. The primary outcome is the proportion of participants who adhere to the use of their orthopedic shoes, that is, who take at least 80% of their total daily steps with orthopedic shoes. A temperature microsensor will be built into the participants’ orthopedic shoes to measure wearing time continuously over 12 months. In addition, daily activity will be measured periodically using log data with an activity monitor. Data from the temperature microsensor and activity monitor will be combined to calculate adherence. (Re-)experienced complications after receiving orthopedic shoes will be registered. Questionnaires and interviews will measure the experiences of participants regarding orthopedic shoes, experiences of podiatrists regarding motivational interviewing, care consumption, and quality of life. Differences in costs and quality of life will be determined in a cost-effectiveness analysis. Discussion This trial will generate novel insights into the socio-economic and well-being impact and the clinical effectiveness of the novel care approach on adherence to wearing orthopedic shoes. Trial registration Netherlands Trial Register NL7710. Registered on 6 May 2019

Integration of Consumer-Based Activity Monitors into Clinical Practice for Children with Type 1 Diabetes: A Feasibility Study

Current technology commonly utilized in diabetes care includes continuous glucose monitors (CGMs) and insulin pumps. One often overlooked critical component to the human glucose response is daily physical activity habits. Consumer-based activity monitors may be a valid way for clinics to collect physical activity data, but whether or not children with type 1 diabetes (T1D) would wear them or use the associated mobile application is unknown. Therefore, the purpose of this study was to test the feasibility of implementing a consumer-based accelerometer directly into ongoing care for adolescents managing T1D. Methods: Adolescents with T1D were invited to participate in this study and instructed to wear a mobile physical activity monitor while also completing a diet log for a minimum of 3 days. Clinical compliance was defined as the number of participants who were compliant with all measures while also having adequate glucose recordings using either a CGM, insulin pump, or on the diet log. Feasibility was defined as >50% of the total sample reaching clinical compliance. Results: A total of 57 children and teenagers between the ages of 7 and 19 agreed to participate in this study and were included in the final analysis. Chi-square results indicated significant compliance for activity tracking (p < 0.001), diet logs (p = 0.04), and overall clinical compliance (p = 0.04). Conclusion: More than half the children in this study were compliant for both activity monitoring and diet logs. This indicates that it is feasible for children with T1D to wear a consumer-based activity monitor while also recording their diet for a minimum of three days.

Effectiveness of a customised mobile phone text messaging intervention supported by data from activity monitors for improving lifestyle factors related to the risk of type 2 diabetes among women after gestational diabetes: protocol for a multicentre randomised controlled trial (SMART MUMS with smart phones 2)

IntroductionGestational diabetes (GDM) contributes substantially to the population burden of type 2 diabetes (T2DM), with a high long-term risk of developing T2DM. This study will assess whether a structured lifestyle modification programme for women immediately after a GDM pregnancy, delivered via customised text messages and further individualised using data from activity monitors, improves T2DM risk factors, namely weight, physical activity (PA) and diet.Methods and analysisThis multicentre randomised controlled trial will recruit 180 women with GDM attending Westmead, Campbelltown or Blacktown hospital services in Western Sydney. They will be randomised (1:1) on delivery to usual care with activity monitor (active control) or usual care plus activity monitor and customised education, motivation and support delivered via text messaging (intervention). The intervention will be customised based on breastfeeding status, and messages including their step count achievements to encourage PA. Messages on PA and healthy eating will encourage good lifestyle habits. The primary outcome of the study is healthy lifestyle composed of weight, dietary and PA outcomes, to be evaluated at 6 months. The secondary objectives include the primary objective components, body mass index, breastfeeding duration and frequency, postnatal depression, utilisation of the activity monitor, adherence to obtaining an oral glucose tolerance test post partum and the incidence of dysglycaemia at 12 months. Relative risks and their 95% CIs will be presented for the primary objective and the appropriate regression analysis, adjusting for the baseline outcome results, will be done for each outcome.Ethics and disseminationEthics approval has been received from the Western Sydney Local Health District Human Research Ethics Committee (2019/ETH13240). All patients will provide written informed consent. Study results will be disseminated via the usual channels including peer-reviewed publications and presentations at national and international conferences.Trial registration numberACTRN12620000615987; Pre-results.

An inexpensive setup for robust activity tracking in small animals: Portable Locomotion Activity Monitor (pLAM)

Advances in computer vision and deep learning have automated animal behaviour studies that previously required tedious manual input. However, tracking activity of small and fast flying animals remains a hurdle, especially in a field setting with variable light conditions. Commercial locomotor activity monitors (LAMs) can be expensive, closed source, and generally limited to laboratory settings.Here, we present a portable locomotion activity monitor (pLAM), a mobile activity detector to quantify small animal circadian activity. Our setup uses inexpensive components, is based on open-source motion tracking software, and is easy to assemble and use in the field. It runs off-grid, supports low-light tracking with infrared lights, and can implement arbitrary light cycle colours and brightnesses with programmable LEDs. We provide a user-friendly guide to assembling pLAM hardware and accessing its pre-configured software and guidelines for using it in other systems.We benchmarked pLAM for insects under various lab and field conditions, then compared results to a commercial activity detector. They offer broadly similar activity measures, but our setup captures flight and bouts of motion that are often missed by beam-breaking activity detection.pLAM will enable high-throughput quantification of small animal location and activity in a low-cost and accessible manner, crucial to studying behaviour that can help inform conservation and management decisions.