Self-Supervised Machine Learning to Characterise Step Counts from Wrist-Worn Accelerometers in the UK Biobank.

PURPOSE: Step count is an intuitive measure of physical activity frequently quantified in health-related studies; however, accurate step counting is difficult in the free-living environment, with error routinely above 20% in wrist-worn devices against camera-annotated ground truth. This study aims to describe the development and validation of step count derived from a wrist-worn accelerometer and assess its association with cardiovascular and all-cause mortality in a large prospective cohort. METHODS: We developed and externally validated a self-supervised machine learning step detection model, trained on an open-source and step-annotated free-living dataset. 39 individuals will free-living ground-truth annotated step counts were used for model development. An open-source dataset with 30 individuals was used for external validation. Epidemiological analysis was performed using 75,263 UK Biobank participants without prevalent cardiovascular disease (CVD) or cancer. Cox regression was used to test the association of daily step count with fatal CVD and all-cause mortality after adjustment for potential confounders. RESULTS: The algorithm substantially outperformed reference models (free-living mean absolute percent error of 12.5%, versus 65-231%). Our data indicate an inverse dose-response association, where taking 6,430-8,277 daily steps was associated with 37% [25-48%] and 28% [20-35%] lower risk of fatal CVD and all-cause mortality up to seven years later, compared to those taking fewer steps each day. CONCLUSIONS: We have developed an open and transparent method that markedly improves the measurement of steps in large-scale wrist-worn accelerometer datasets. The application of this method demonstrated expected associations with CVD and all-cause mortality, indicating excellent face validity. This reinforces public health messaging for increasing physical activity and can help lay the groundwork for the inclusion of target step counts in future public health guidelines.

Digital health technologies and machine learning augment patient reported outcomes to remotely characterise rheumatoid arthritis.

Digital measures of health status captured during daily life could greatly augment current in-clinic assessments for rheumatoid arthritis (RA), to enable better assessment of disease progression and impact. This work presents results from weaRAble-PRO, a 14-day observational study, which aimed to investigate how digital health technologies (DHT), such as smartphones and wearables, could augment patient reported outcomes (PRO) to determine RA status and severity in a study of 30 moderate-to-severe RA patients, compared to 30 matched healthy controls (HC). Sensor-based measures of health status, mobility, dexterity, fatigue, and other RA specific symptoms were extracted from daily iPhone guided tests (GT), as well as actigraphy and heart rate sensor data, which was passively recorded from patients' Apple smartwatch continuously over the study duration. We subsequently developed a machine learning (ML) framework to distinguish RA status and to estimate RA severity. It was found that daily wearable sensor-outcomes robustly distinguished RA from HC participants (F1, 0.807). Furthermore, by day 7 of the study (half-way), a sufficient volume of data had been collected to reliably capture the characteristics of RA participants. In addition, we observed that the detection of RA severity levels could be improved by augmenting standard patient reported outcomes with sensor-based features (F1, 0.833) in comparison to using PRO assessments alone (F1, 0.759), and that the combination of modalities could reliability measure continuous RA severity, as determined by the clinician-assessed RAPID-3 score at baseline (r2, 0.692; RMSE, 1.33). The ability to measure the impact of the disease during daily life-through objective and remote digital outcomes-paves the way forward to enable the development of more patient-centric and personalised measurements for use in RA clinical trials.

Development and Validation of a Machine Learning Wrist-worn Step Detection Algorithm with Deployment in the UK Biobank.

Background: Step count is an intuitive measure of physical activity frequently quantified in a range of health-related studies; however, accurate quantification of step count can be difficult in the free-living environment, with step counting error routinely above 20% in both consumer and research-grade wrist-worn devices. This study aims to describe the development and validation of step count derived from a wrist-worn accelerometer and to assess its association with cardiovascular and all-cause mortality in a large prospective cohort study. Methods: We developed and externally validated a hybrid step detection model that involves self-supervised machine learning, trained on a new ground truth annotated, free-living step count dataset (OxWalk, n=39, aged 19-81) and tested against other open-source step counting algorithms. This model was applied to ascertain daily step counts from raw wrist-worn accelerometer data of 75,493 UK Biobank participants without a prior history of cardiovascular disease (CVD) or cancer. Cox regression was used to obtain hazard ratios and 95% confidence intervals for the association of daily step count with fatal CVD and all-cause mortality after adjustment for potential confounders. Findings: The novel step algorithm demonstrated a mean absolute percent error of 12.5% in free-living validation, detecting 98.7% of true steps and substantially outperforming other recent wrist-worn, open-source algorithms. Our data are indicative of an inverse dose-response association, where, for example, taking 6,596 to 8,474 steps per day was associated with a 39% [24-52%] and 27% [16-36%] lower risk of fatal CVD and all-cause mortality, respectively, compared to those taking fewer steps each day. Interpretation: An accurate measure of step count was ascertained using a machine learning pipeline that demonstrates state-of-the-art accuracy in internal and external validation. The expected associations with CVD and all-cause mortality indicate excellent face validity. This algorithm can be used widely for other studies that have utilised wrist-worn accelerometers and an open-source pipeline is provided to facilitate implementation.

Patient Clustering for Vital Organ Failure Using ICD Code With Graph Attention.

OBJECTIVE: Heart failure, respiratory failure and kidney failure are three severe organ failures (OF) that have high mortalities and are most prevalent in intensive care units. The objective of this work is to offer insights into OF clustering from the aspects of graph neural networks and diagnosis history. METHODS: This paper proposes a neural network-based pipeline to cluster three types of organ failure patients by incorporating embedding pre-train using an ontology graph of the International Classification of Diseases (ICD) codes. We employ an autoencoder-based deep clustering architecture jointly trained with a K-means loss, and a non-linear dimension reduction is performed to obtain patient clusters on the MIMIC-III dataset. RESULTS: The clustering pipeline shows superior performance on a public-domain image dataset. On the MIMIC-III dataset, it discovers two distinct clusters that exhibit different comorbidity spectra which can be related to the severity of diseases. The proposed pipeline is compared with several other clustering models and shows superiority. CONCLUSION: Our proposed pipeline gives stable clusters, however, they do not correspond to the type of OF which indicates these OF share significant hidden characteristics in diagnosis. These clusters can be used to signal possible complications and severity of illness and aid personalised treatment. SIGNIFICANCE: We are the first to apply an unsupervised approach to offer insights from a biomedical engineering perspective on these three types of organ failure, and publish the pre-trained embeddings for future transfer learning.

A Deep Learning Approach for the Assessment of Signal Quality of Non-Invasive Foetal Electrocardiography.

Non-invasive foetal electrocardiography (NI-FECG) has become an important prenatal monitoring method in the hospital. However, due to its susceptibility to non-stationary noise sources and lack of robust extraction methods, the capture of high-quality NI-FECG remains a challenge. Recording waveforms of sufficient quality for clinical use typically requires human visual inspection of each recording. A Signal Quality Index (SQI) can help to automate this task but, contrary to adult ECG, work on SQIs for NI-FECG is sparse. In this paper, a multi-channel signal quality classifier for NI-FECG waveforms is presented. The model can be used during the capture of NI-FECG to assist technicians to record high-quality waveforms, which is currently a labour-intensive task. A Convolutional Neural Network (CNN) is trained to distinguish between NI-FECG segments of high and low quality. NI-FECG recordings with one maternal channel and three abdominal channels were collected from 100 subjects during a routine hospital screening (102.6 min of data). The model achieves an average 10-fold cross-validated AUC of 0.95 ± 0.02. The results show that the model can reliably assess the FECG signal quality on our dataset. The proposed model can improve the automated capture and analysis of NI-FECG as well as reduce technician labour time.

Measuring and Localizing Individual Bites Using a Sensor Augmented Plate During Unrestricted Eating for the Aging Population.

Food intake monitoring can play an important role in the prevention of malnutrition in the aging population, but traditional tools may not be adequate for use in this target group. These tools typically involve the use of questionnaires or food diaries that require manual data entry. Due to their time-consuming nature, they are often incomplete, contain mistakes, or not used at all. An alternative to self-reporting tools, in the form of a plate system that automatically measures the consumed food during the meal, is presented in this paper. Furthermore, the system can estimate the location where each bite was taken on the plate. The system is compatible with an off-the-shelf plate that is mounted on top of a base station. Weight sensors are integrated in the base, allowing for easy removal and cleaning of the plate. Localization of bites is done by looking at the movement of the center of mass during eating. When used with a compartmentalized plate, the amount of consumed food per compartment can be measured. With prior knowledge of the type of food in each compartment, this can give an indication of calories and nutritional intake. We present a bite detection algorithm using a random forest decision tree classifier. Data from 24 aging adults (ages 52-95) eating a single meal with chopsticks was used to train and evaluate the model. Out of a total of 836 true annotated bites, the algorithm detected 602 with a precision and recall of 0.78 and 0.76, respectively. By summing the weights of detected bites from each compartment, the algorithm was able to estimate the amount of food taken per compartment with an average error of (8 ±8)% of the portion size.

Measuring weight and location of individual bites using a sensor augmented smart plate.

In this work, a novel plate system that can detect weight and location of individual bites during meals is presented. The system consists of a base station with sensors and a detachable off-the-shelf polymer plate with three com- partments. By combining data from multiple weight sensors, the weight of individual bites can be accurately measured and localized on the plate to determine the compartment from which they were taken. With prior knowledge of the weight of the food in each compartment at the start of the meal, the system can estimate the nutritional value of the consumed food. In a test conducted in a controlled home environment, the system was able to measure the weight of consumed food in each compartment with a maximum relative error of 1.4%. The goal of the system is to replace traditional monitoring tools and to automatically monitor the amount of consumption.

Improving the accuracy of existing camera based fall detection algorithms through late fusion.

Fall incidents remain an important health hazard for older adults. Fall detection systems can reduce the consequences of a fall incident by insuring that timely aid is given. Currently fall detection algorithms however suffer a reduction in accuracy when introduced in real-life situations. In this paper a late fusion technique is proposed that will improve the accuracy of existing fall detection systems. It combines the confidence levels of different single camera fall detection systems. Four different aggregation methods are compared to each other based on the Area Under the Curve (AUC) of precision-recall curves. Calculating the median of the confidence levels of five cameras an increase of 218% in the AUC of the precision-recall-curves is achieved compared to the AUC of the single camera fall detector. These results show that significant improvements can be made to the accuracy of single camera fall detectors in a relatively easy way.

Detection of chewing motion in the elderly using a glasses mounted accelerometer in a real-life environment.

This paper describes a method of detecting an elderly person's chewing motion using a glasses mounted accelerometer. A real-life dataset was collected from 13 elderly adults, aged 65 or older, during meal times in a care facility. A supervised classifier is used to automatically distinguish between epochs of chewing and non-chewing activity. Results are compared to a lab dataset of 5 young to middle-aged adults captured in previous work. K-Nearest Neighbor, Random Forest and Support Vector Machine classifiers are evaluated. All are able to achieve similar performance, with the Support Vector Machine performing the best with an F1-score of 0.73.

Bridging the gap between real-life data and simulated data by providing a highly realistic fall dataset for evaluating camera-based fall detection algorithms.

Fall incidents are an important health hazard for older adults. Automatic fall detection systems can reduce the consequences of a fall incident by assuring that timely aid is given. The development of these systems is therefore getting a lot of research attention. Real-life data which can help evaluate the results of this research is however sparse. Moreover, research groups that have this type of data are not at liberty to share it. Most research groups thus use simulated datasets. These simulation datasets, however, often do not incorporate the challenges the fall detection system will face when implemented in real-life. In this Letter, a more realistic simulation dataset is presented to fill this gap between real-life data and currently available datasets. It was recorded while re-enacting real-life falls recorded during previous studies. It incorporates the challenges faced by fall detection algorithms in real life. A fall detection algorithm from Debard et al. was evaluated on this dataset. This evaluation showed that the dataset possesses extra challenges compared with other publicly available datasets. In this Letter, the dataset is discussed as well as the results of this preliminary evaluation of the fall detection algorithm. The dataset can be downloaded from www.kuleuven.be/advise/datasets.