Data Engineering

Type 1 diabetes (T1D) management can be significantly enhanced through the use of predictive machine learning (ML) algorithms, which can mitigate the risk of adverse events like hypoglycemia. Hypoglycemia, characterized by blood glucose levels below 70 mg/dL, is a life-threatening condition typically caused by excessive insulin administration, missed meals, or physical activity. Its asymptomatic nature impedes timely intervention, making ML models crucial for early detection. This study integrates short- (up to 2h) and long-term (up to 24h) prediction horizons (PHs) within a single classification model to enhance decision support. The predicted times are 5-15 min, 15-30 min, 30 min-1h, 1-2h, 2-4h, 4-8h, 8-12h, and 12-24h before hypoglycemia. In addition, a simplified model classifying up to 4h before hypoglycemia is compared. We trained ResNet and LSTM models on glucose levels, insulin doses, and acceleration data. The results demonstrate the superiority of the LSTM models when classifying nine classes. In particular, subject-specific models yielded better performance but achieved high recall only for classes 0, 1, and 2 with 98%, 72%, and 50%, respectively. A population-based six-class model improved the results with at least 60% of events detected. In contrast, longer PHs remain challenging with the current approach and may be considered with different models.

Specific phobias, such as spider phobia, are a widespread condition, that can negatively impact the quality of life of affected people. Phobias are typically characterised by avoidance behaviour, which can be measured using a behavioural avoidance test (BAT). While behavioural avoidance tests are traditionally performed using a real stimulus (e.g., a spider), virtual reality has also gained popularity in the field of psychology. This offers a more accessible and affordable alternative to traditional diagnostic methods. Our work focuses on a comparison of BATs in vivo and in virtuo, and considering two different approach modalities, by analysing the physiological responses of participants. This study aims to investigate whether these responses during in virtuo BATs are comparable to those observed during in vivo BATs, and whether the modality influences the outcome. In this work, we present our study involving 25 participants and an initial look at the data collected.

The attention we pay to food can provide information about our lifestyle and our relationship to eating. The visual search task is a psychological method commonly used to study searching behaviour and attention, and to infer how these processes relate to the objects being searched for. Efforts have already been made to adapt this method to virtual world, but most implementations remain in 2D, typically on computer monitors. This is noteworthy because, by integrating head position tracking and enabling novel interaction methods that capitalise on this capability, virtual reality offers significant potential to enhance the study of search behaviour. This paper describes our planned preliminary study comparing our custom VR application for advanced search behaviour analysis utilizing the Unity Gaze Interactor, to a classical visual search implementation on a computer. The Unity Gaze Interactor enables interactions with virtual objects and precise tracking of users' gaze direction, providing new opportunities for analysing search behaviour in immersive environments. Our research investigates the behaviours of people who primarily tend towards low-calorie or high-calorie food, comparing their responses across these both approaches. The study started in December 2024.

Psychological trials, such as behavioural avoidance tests (BAT), are a fundamental part of the phobia therapeutic process. In order to link physiological reactions with specific points in time during psychological trials, it is necessary to integrate observation data with data collected automatically by sensors, such as wearable devices. To this end, this paper introduces FRAME - a framework for combining real-world events occurring during psychological trials with physiological data collected by wearables. FRAME consists of three parts, an Observation App, a data integration module and a Virtual Reality (VR) App. The Observation App captures events and their exact time of occurrence. The integration module links the observations with the respective physiological data, allowing an in-depth analysis of physiological reactions. The VR App provides a virtual scenario based on the BAT in vivo, thus enabling a BAT in virtuo. The practical applicability of FRAME is tested within a study comparing behavioural avoidance tests in vivo and in virtuo, assessing 25 patients with arachnophobia wearing an Empatica E4.

Patients with type 1 diabetes (T1D) have a higher risk of experiencing hypoglycemia, which is a severe condition of decreased blood glucose levels under 70 mg/dL and can result in coma, or death in the worst case. Prediction algorithms could improve diabetes care by enabling preventive actions, but research is limited by available multivariate datasets. Thus, this work investigates the feasibility of transfer learning between two different datasets of people with T1D and type 2 pre-diabetes using a 1 Dimensional Convolutional Neural Network (1DCNN) model. Moreover, different thresholds for defining hypoglycemia are compared for the pre-diabetes group. The results show that transfer learning could be feasible if the model is trained on T1D with a threshold of 70 mg/dL, while a threshold of 80–85 mg/dL achieved the best training performance for the pre-diabetes data.

Immersion and presence are important aspects of virtual reality (VR). Efforts are therefore being made to enhance these effects in order to increase the impact of VR and its transferability to reality. This is particularly important in research areas such as medical education and psychology. The development of haptic gloves has therefore increased significantly in recent years. These are immersive input devices designed to make virtual objects tangible. However, the research field is still very new and there is insufficient research on their handling and effect. For this reason, an application was developed to compare different haptic data gloves with different haptic technologies. A glove with vibrotactile feedback (Manus Prime X Haptic VR) and a glove with additional force feedback (SenseGlove) were used and compared in multiple tasks. These include finger tracking and gesture recognition, as well as hand tracking and the behaviour of gloves when grasping objects with and without haptics. The aim is to determine how well the haptics could be integrated and which is more accepted by the test subjects. For this purpose, a questionnaire has been designed to evaluate the effect of the haptic gloves on users. It aims to get a first impression of the tactile gloves available on the market and to advance research in the area.

Cognitive performance is an important aspect of life. Not only does thinking enable us to act, there is also an interaction between cognitive performance and mental health. It is therefore increasingly important to train and practice one’s cognitive abilities and benefit from brain plasticity. According to Vester’s model of learning types visual, auditiv and haptic impression are imporant for learning. An existing memory training application, which transfers the concept of the game Pairs with personal memories into virtual reality (VR), has now been supplemented with a haptic impression. The original application already establishes the visual and auditory perception channels. With new haptic hardware constantly appearing on the market, the Tactsuit X40 haptic vest from bhaptics has now made it possible to include a third perceptual channel with a wide variation of haptic sensations on various parts of the upper body. Each playing card is associated with an individual image, a piece of music and a vibration pattern. All of which are designed to help the user remember the position of the card. Each of these aspects can be freely enabled and disabled. The result of this work is an application which aims to find if and how strongly haptic impressions influence memory performance. By taking more types of learning into account, we hope to achieve a better memory training. A study has been designed to evaluate the specific aspects of the application and is currently being carried out. In addition to the self-generated haptic patterns, it is also possible to generate haptic impulses based on audio sources. These could possibly provide further support in this application.

In this paper, we introduce the MLM (Multiple Languages and Modalities) dataset - a new resource to train and evaluate multitask systems on samples in multiple modalities and three languages. The generation process and inclusion of semantic data provide a resource that further tests the ability for multitask systems to learn relationships between entities. The dataset is designed for researchers and developers who build applications that perform multiple tasks on data encountered on the web and in digital archives. A second version of MLM provides a geo-representative subset of the data with weighted samples for countries of the European Union. We demonstrate the value of the resource in developing novel applications in the digital humanities with a motivating use case and specify a benchmark set of tasks to retrieve modalities and locate entities in the dataset. Evaluation of baseline multitask and single task systems on the full and geo-representative versions of MLM demonstrate the challenges of generalising on diverse data. In addition to the digital humanities, we expect the resource to contribute to research in multimodal representation learning, location estimation, and scene understanding.