A5031994646- Time Series Analysis and Forecasting
- Data Stream Mining Techniques
- Hearing, Cochlea, Tinnitus, Genetics
- Hearing Loss and Rehabilitation
- Anomaly Detection Techniques and Applications
- Vestibular and auditory disorders
- Digital Mental Health Interventions
- Olfactory and Sensory Function Studies
- Mental Health Research Topics
- Mobile Health and mHealth Applications
- Noise Effects and Management
- Neural Networks and Applications
- Multisensory perception and integration
- Machine Learning and Data Classification
- Phonocardiography and Auscultation Techniques
- Machine Learning in Healthcare
- Music and Audio Processing
- Mobile Crowdsensing and Crowdsourcing
- Network Security and Intrusion Detection
- Handwritten Text Recognition Techniques
- Neural Networks and Reservoir Computing
- Advanced Chemical Sensor Technologies
- Image and Video Quality Assessment
- Clinical Reasoning and Diagnostic Skills
- Allergic Rhinitis and Sensitization
Otto-von-Guericke University Magdeburg
2018-2024
The recording of Ecological Momentary Assessments (EMA) can assist people with chronic diseases in monitoring their health state. However, many users quickly lose interest respective EMA platforms. Therefore, we studied the adherence mHealth app TRACKYOURTINNITUS (TYT). is used to record tinnitus. 1292 users, who interacted between April 2014 and February 2017, were analyzed this work. We defined "adherence" based on dimensions interaction duration continuity. propose methods that are able...
Abstract Tinnitus is associated with a variety of aetiologies, phenotypes, and underlying pathophysiological mechanisms, available treatments have limited efficacy. A combination treatments, addressing various aspects tinnitus, might provide viable superior treatment strategy. In this international multicentre, parallel-arm, superiority, randomised controlled trial, patients chronic subjective tinnitus were recruited from five clinical sites across the EU as part interdisciplinary...
Tinnitus is a heterogeneous condition. The aim of this study as to compare the online and hospital responses Spanish version European School for Interdisciplinary Research screening-questionnaire (ESIT-SQ) in tinnitus individuals by an unsupervised age clustering.A cross-sectional was performed including 434 white patients with chronic assess demographic clinical profile through ESIT-SQ, 204 outpatients 230 from survey; K-means clustering algorithm used classify both according age.Online...
Abstract Background Tinnitus is a leading cause of disease burden globally. Several therapeutic strategies are recommended in guidelines for the reduction tinnitus distress; however, little known about potentially increased effectiveness combination treatments and personalized each patient. Methods Within Unification Treatments Interventions Patients project, multicenter, randomized clinical trial conducted with aim to compare single combined on distress (UNITI-RCT). Five different centers...
Chronic tinnitus, the perception of a phantom sound in absence corresponding stimulus, is condition known to affect patients’ quality life. Recent advances mHealth have enabled patients maintain ‘disease journal’ ecologically-valid momentary assessments, improving own awareness their disease while also providing clinicians valuable data for research. In this study, we investigate effect non-personalised tips on and continued use application. The collected from study involved three groups...
Chronic tinnitus is a clinically multidimensional phenomenon that entails audiological, psychological and somatosensory components. Previous research has demonstrated age female gender as potential risk factors, although studies to this regard are heterogeneous. Moreover, whilst recent begun identify clinical "phenotypes," little known about differences in patient population profiles at geographically separated specialized treatment centers. Identifying such might prevent biases joint...
Abstract Traditional active learning tries to identify instances for which the acquisition of label increases model performance under budget constraints. Less research has been devoted task actively acquiring feature values, whereupon both instance and must be selected intelligently even less a scenario where arrive in stream with drift. We propose an strategy data streams drift, as well evaluation framework. also implement baseline that chooses features randomly compare random approach...
Electronic health records (EHR) often include multiple perspectives on a patient's current state of well-being (e.g. vital signs and subjective indicators measured by questionnaires). In this study, we use these to build phenotypes chronic tinnitus patients investigate how are associated with response treatment. Therefore, model as nodes in network, where those interpreted layers multi-layer network. To identify the implement community detection algorithm. Some communities can be considered...
Stream classification algorithms traditionally treat arriving observations as independent. However, in many applications the examples may depend on "entity" that generated them, e.g. product reviewing or interactions of users with an application server. In this study, we investigate potential dependency by partitioning original stream into entity-centric substreams and incorporating entity-specific information learning model. We propose a k Nearest Neighbour inspired approach (kNN), which...
Current clinical decision support systems (DSS) are trained and validated on observational data from the target clinic. This is problematic for treatments in a randomized trial (RCT), but not yet introduced any In this work, we report method training validating DSS using RCT data. The key challenges address of missingness -- missing rationale treatment assignment (the at random), verification evidence, since effectiveness patient can only be verified (ground truth) what were actually...
Opinion stream mining algorithms learn and adapt a polarity model as new opinionated texts arrive. Text understanding is computationally expensive though, sensitive to the emergence of words. In this work, we study prediction for opinions on given entities investigate how quality affected when ignore text past but exploit entity-opinion link scores it. We each entity trajectory propose learning that these trajectories prediction. performance our approach Tools & Home Improvement products...
Abstract Background Tinnitus is a leading cause of disease burden globally. Several therapeutic strategies are recommended in guidelines for the reduction tinnitus distress; however, little known about potential increased effectiveness combination treatments and personalized each patient. Methods Within Unification Treatments Interventions Patients project, multicenter, randomized clinical trial conducted with aim to compare single combined on distress (UNITI-RCT). Five different centers...
Smartphones and other mobile devices offer a valu-able opportunity to gather patient-specific health data during everyday life. However, the increasing popularity of apps demands specialized analysis methods that can handle unique, patient-based, time-dependent, often multivariate collected by these apps. This work explores patient-based mHealth develop personalized prediction models. The models incor-porate not only from individual patient, but also similar patients using neighborhoods. Our...
Mobile Health (mhealth) applications are increasing in popularity, and the collection of disease-specific time series data using Ecological Momentary Assessment (EMA) questionnaires has been shown to help creation personalised predictors for next-step forecasting, which can be crucial giving preemptive interventions. In this work, we propose a framework that aims mitigate common issue EMA - some users contribute bulk while most too little. Our proposed discover `useful' neighbourhood `long'...
Ecological momentary assessment (EMA) has been used in many mHealth apps. EMA captures valuable insights into diseases. Identifying the Granger causal relationships across variables may contribute to interpretation of a disease and improve treatment decisions. In our study, we perform circadian conditional causality analysis on multivariate time series. The was done data 270 users an app tinnitus their registration data, using latter explain relationships. We discovered that some items cause...
Opinion stream classification algorithms adapt the model to arriving review texts and, depending on forgetting scheme, reduce contribution old reviews have upon model. Reviews are assumed independent, and information entity which a refers, i.e. opinion target, is thereby ignored. This implies that prediction of review's label based more referring other, popular or simply recently inspected entities, while same might be ignored as too old. In this study, we enforce each taken into account for...
A stream of users' interactions with an mHealth app can be seen as the result a stochastic process that captured by algorithm learns over whole stream. But is it only one process? We investigate to what extend learning for each user separately delivers better predictions than model Our application scenario prediction Ecological Momentary Assessments (EMA) (TinnitusTipps) on tinnitus. The data were recorded part pilot study, in which group users received non-personalized suggestions (tips)...