Representation bias in health data can lead to unfair decisions and compromise the generalisability of research findings. As a consequence, underrepresented subpopulations, such as those from specific ethnic backgrounds or genders, do not benefit equally from clinical discoveries. Several approaches have been developed to mitigate representation bias, ranging from simple resampling methods, such as SMOTE, to recent approaches based on generative adversarial networks (GAN). However, generating high-dimensional time-series synthetic health data remains a significant challenge. In response, we devised a novel architecture (CA-GAN) that synthesises authentic, high-dimensional time series data. CA-GAN outperforms state-of-the-art methods in a qualitative and a quantitative evaluation while avoiding mode collapse, a serious GAN failure. We perform evaluation using 7535 patients with hypotension and sepsis from two diverse, real-world clinical datasets. We show that synthetic data generated by our CA-GAN improves model fairness in Black patients as well as female patients when evaluated separately for each subpopulation. Furthermore, CA-GAN generates authentic data of the minority class while faithfully maintaining the original distribution of data, resulting in improved performance in a downstream predictive task.

Deep Learning (DL) models have received increasing attention in the clinical setting, particularly in intensive care units (ICU). In this context, the interpretability of the outcomes estimated by the DL models is an essential step towards increasing adoption of DL models in clinical practice. To address this challenge, we propose an ante-hoc, interpretable neural network model. Our proposed model, named double self-attention architecture (DSA), uses two attention-based mechanisms, including self-attention and effective attention. It can capture the importance of input variables in general, as well as changes in importance along the time dimension for the outcome of interest. We evaluated our model using two real-world clinical datasets covering 22840 patients in predicting onset of delirium 12 h and 48 h in advance. Additionally, we compare the descriptive performance of our model with three post-hoc interpretable algorithms as well as with the opinion of clinicians based on the published literature and clinical experience. We find that our model covers the majority of the top-10 variables ranked by the other three post-hoc interpretable algorithms as well as the clinical opinion, with the advantage of taking into account both, the dependencies among variables as well as dependencies between varying time-steps. Finally, our results show that our model can improve descriptive performance without sacrificing predictive performance.

Optimal performance is desired for decision-making in any field with binary classifiers and diagnostic tests, however common performance measures lack depth in information. The area under the receiver operating characteristic curve (AUC) and the area under the precision recall curve are too general because they evaluate all decision thresholds including unrealistic ones. Conversely, accuracy, sensitivity, specificity, positive predictive value and the F1 score are too specific-they are measured at a single threshold that is optimal for some instances, but not others, which is not equitable. In between both approaches, we propose deep ROC analysis to measure performance in multiple groups of predicted risk (like calibration), or groups of true positive rate or false positive rate. In each group, we measure the group AUC (properly), normalized group AUC, and averages of: sensitivity, specificity, positive and negative predictive value, and likelihood ratio positive and negative. The measurements can be compared between groups, to whole measures, to point measures and between models. We also provide a new interpretation of AUC in whole or part, as balanced average accuracy, relevant to individuals instead of pairs. We evaluate models in three case studies using our method and Python toolkit and confirm its utility.

Occupational stress has a significant adverse effect on workers’ well-being, productivity, and performance and is becoming a major concern for both individual companies and the overall economy. To reduce negative consequences, early detection of stress is a key factor. In response several stress prediction methods have been proposed, whose primary aim is to analyse physiological and behavioural data. However, evidence suggests that solutions based on physiological and behavioural data alone might be challenging when implemented in real-world settings. These solutions are sensitive to data problems arising from losses in signal quality or alterations in body responses, which are common in everyday activities. The contagious nature of stress and its sensitivity to the surroundings can be used to improve these methods. In this study, we sought to investigate automatic stress prediction using both surrounding stress data, which we define as close colleagues’ stress levels and the stress level history of the individuals. We introduce a real-life, unconstrained study conducted with 30 workers monitored over 8 weeks. Furthermore, we propose a method to investigate the effect of stress levels of close colleagues on the prediction of an individual’s stress levels. Our method is also validated on an external, independent dataset. Our results show that surrounding stress can be used to improve stress prediction in the workplace, where we achieve 80% of F-score in predicting individuals’ stress levels from the surrounding stress data in a multiclass stress classification.

Stress assessment is a complex issue and numerous studies have examined factors that influence stress in working environments. Research studies have shown that monitoring individuals' behaviour parameters during daily life can also help assess stress levels. In this study, we examine assessment of work-related stress using features derived from sensors in smartphones. In particular, we use information from physical activity levels, location, social-interactions, social-activity, and application usage during working days. Our study included 30 employees chosen from two different private companies, monitored over a period of 8 weeks in real work environments. The findings suggest that information from phone sensors shows important correlation with employees perceived stress level. Second, we used machine learning methods to classify perceived stress levels based on the analysis of information provided by smartphones. We used decision trees obtaining 67.57 percent accuracy and 71.73 percent after applying a semi-supervised method. Our results show that stress levels can be monitored in unobtrusive manner, through analysis of smartphone data.

Increase in workload across many organizations and consequent increase in occupational stress are negatively affecting the health of the workforce. Measuring stress and other human psychological dynamics is difficult due to subjective nature of selfreporting and variability between and within individuals. With the advent of smartphones, it is now possible to monitor diverse aspects of human behavior, including objectively measured behavior related to psychological state and consequently stress. We have used data from the smartphone's built-in accelerometer to detect behavior that correlates with subjects stress levels. Accelerometer sensor was chosen because it raises fewer privacy concerns (e.g., in comparison to location, video, or audio recording), and because its low-power consumption makes it suitable to be embedded in smaller wearable devices, such as fitness trackers. About 30 subjects from two different organizations were provided with smartphones. The study lasted for eight weeks and was conducted in real working environments, with no constraints whatsoever placed upon smartphone usage. The subjects reported their perceived stress levels three times during their working hours. Using combination of statistical models to classify selfreported stress levels, we achieved a maximum overall accuracy of 71% for user-specific models and an accuracy of 60% for the use of similar-users models, relying solely on data from a single accelerometer.

Today's health care is difficult to imagine without the possibility to objectively measure various physiological parameters related to patients' symptoms (from temperature through blood pressure to complex tomographic procedures). Psychiatric care remains a notable exception that heavily relies on patient interviews and self-assessment. This is due to the fact that mental illnesses manifest themselves mainly in the way patients behave throughout their daily life and, until recently there were no "behavior measurement devices." This is now changing with the progress in wearable activity recognition and sensor enabled smartphones. In this paper, we introduce a system, which, based on smartphone-sensing is able to recognize depressive and manic states and detect state changes of patients suffering from bipolar disorder. Drawing upon a real-life dataset of ten patients, recorded over a time period of 12 weeks (in total over 800 days of data tracing 17 state changes) by four different sensing modalities, we could extract features corresponding to all disease-relevant aspects in behavior. Using these features, we gain recognition accuracies of 76% by fusing all sensor modalities and state change detection precision and recall of over 97%. This paper furthermore outlines the applicability of this system in the physician-patient relations in order to facilitate the life and treatment of bipolar patients.

Mobile computing is changing the landscape of clinical monitoring and self-monitoring. One of the major impacts will be in healthcare, where increase in number of sensing modalities is providing more and more information on the state of overall wellbeing, behaviour and health. There are numerous applications of mobile computing that range from wellbeing applications, such as physical fitness, stress or burnout up to applications that target mental disorders including bipolar disorder. Use of information provided by mobile computing devices can track the state of the subjects and also allow for experience sampling in order to gather subjective information. This paper reports on the results obtained from a medical trial with monitoring of bipolar disorder patients and how the episodes of the diseases correlate to the analysis of the data sampled from mobile phone acting as a monitoring device.

Localization plays an essential role in many ubiquitous computing applications. While the outdoor location-aware services based on GPS are becoming increasingly popular, their proliferation to indoor environments is limited due to the lack of widely available indoor localization systems. The de-facto standard for indoor positioning is based on Wi-Fi and while other localization alternatives exist, they either require expensive hardware or provide a low accuracy. This paper presents an investigation into localization system that leverages signals of broadcasting FM radio stations. The FM stations provide a worldwide coverage, while FM tuners are readily available in many mobile devices. The experimental results show that FM radio can be used for indoor localization, while providing longer battery life than Wi-Fi, making FM an alternative to consider for positioning.

Position of mobile users has become highly important information in pervasive computing environments. Indoor localization systems based on Wi-Fi signal strength fingerprinting techniques are widely used in office buildings with existing Wi-Fi infrastructure. Our previous work has proposed a solution based on exploitation of FM signal to deal with environments not covered with Wi-Fi signal or environments with only single Wi-Fi access point. However, a general problem of indoor wireless positioning systems pertains to signal degradation due to the environmental factors affecting signal propagation. Therefore, in order to maintain a desirable level of localization accuracy, it becomes necessary to perform periodic calibration of the system, which is either time consuming or requires dedicated equipment and expert knowledge. In this paper, we present a comparison of FM versus Wi-Fi positioning systems and a combination of both systems, exploiting their strengths for indoors positioning. Finally, we address the problem of recalibration by introducing a novel concept of spontaneous recalibration and demonstrate it using the FM localization system.

In this paper we describe an activity recognition system capable of monitoring user activities in home environments. Activities are monitored by processing information from various sensors embedded in the environment that provide information pertinent to user's actions. We utilise the concept of self-organising object networks to gather and hierarchically process information related to user actions in a distributed manner. This information is then fed to the decision module which matches this information in the user's activity map in order to deduce user's activity. The decision module comprises a Bayesian network coupled with a rule-based engine which is used to provide accurate activity inference process.