Assessment of Immune Status Using Inexpensive Cytokines: A Literature Review and Learning Approaches.

The emergence of point-of-care (POC) testing has lately been promoted to deliver rapid, reliable medical tests in critical life-threatening situations, especially in resource-limited settings. Recently, POC tests have witnessed further advances due to the technological revolution in smartphones. Smartphones are integrated as reliable readers to the POC results to improve their quantitative detection. This has enabled the use of more complex medical tests by the patient him/herself at home without the need for professional staff and sophisticated equipment. Cytokines, the important immune system biomarkers, are still measured today using the time-consuming Enzyme-Linked Immunosorbent Assay (ELISA), which can only be performed in specially equipped laboratories. Therefore, in this study, we investigate the current development of POC technologies suitable for the home testing of cytokines by conducting a PRISMA literature review. Then, we classify the collected technologies as inexpensive and expensive depending on whether the cytokines can be measured easily at home or not. Additionally, we propose a machine learning-based solution to even increase the efficiency of the cytokine measurement by leveraging the cytokines that can be inexpensively measured to predict the values of the expensive ones. In total, we identify 12 POCs for cytokine quantification. We find that Interleukin 1ß (IL-1ß), Interleukin 3 (IL-3), Interleukin 6 (IL-6), Interleukin 8 (IL-8) and Tumor necrosis factor (TNF) can be measured with inexpensive POC technology, namely at home. We build machine-learning models to predict the values of other expensive cytokines such as Interferon-gamma (IFN-γ), IL-10, IL-2, IL-17A, IL-17F, IL-4 and IL-5 by relying on the identified inexpensive ones in addition to the age of the individual. We evaluate to what extent the built machine learning models can use the inexpensive cytokines to predict the expensive ones on 351 healthy subjects from the public dataset 10k Immunomes. The models for IFN-γ show high results for the coefficient of determination: R2 = 0.743. The results for IL-5 and IL-4 are also promising, whereas the predictive model of IL-10 achieves only R2 = 0.126. Lastly, the results demonstrate the vital role of TNF and IL-6 in the immune system due to its high importance in the predictions of all the other expensive cytokines.

Potential of Point-of-Care and At-Home Assessment of Immune Status via Rapid Cytokine Detection and Questionnaire-Based Anamnesis.

Monitoring the immune system's status has emerged as an urgent demand in critical health conditions. The circulating cytokine levels in the blood reflect a thorough insight into the immune system status. Indeed, measuring one cytokine may deliver more information equivalent to detecting multiple diseases at a time. However, if the reported cytokine levels are interpreted with considering lifestyle and any comorbid health conditions for the individual, this will promote a more precise assessment of the immune status. Therefore, this study addresses the most recent advanced assays that deliver rapid, accurate measuring of the cytokine levels in human blood, focusing on add-on potentials for point-of-care (PoC) or personal at-home usage, and investigates existing health questionnaires as supportive assessment tools that collect all necessary information for the concrete analysis of the measured cytokine levels. We introduced a ten-dimensional featuring of cytokine measurement assays. We found 15 rapid cytokine assays with assay time less than 1 h; some could operate on unprocessed blood samples, while others are mature commercial products available in the market. In addition, we retrieved several health questionnaires that addressed various health conditions such as chronic diseases and psychological issues. Then, we present a machine learning-based solution to determine what makes the immune system fit. To this end, we discuss how to employ topic modeling for deriving the definition of immune fitness automatically from literature. Finally, we propose a prototype model to assess the fitness of the immune system through leveraging the derived definition of the immune fitness, the cytokine measurements delivered by a rapid PoC immunoassay, and the complementary information collected by the health questionnaire about other health factors. In conclusion, we discovered various advanced rapid cytokine detection technologies that are promising candidates for point-of-care or at-home usage; if paired with a health status questionnaire, the assessment of the immune system status becomes solid and we demonstrated potentials for promoting the assessment tool with data mining techniques.

Interactive System for Similarity-Based Inspection and Assessment of the Well-Being of mHealth Users.

Recent digitization technologies empower mHealth users to conveniently record their Ecological Momentary Assessments (EMA) through web applications, smartphones, and wearable devices. These recordings can help clinicians understand how the users' condition changes, but appropriate learning and visualization mechanisms are required for this purpose. We propose a web-based visual analytics tool, which processes clinical data as well as EMAs that were recorded through a mHealth application. The goals we pursue are (1) to predict the condition of the user in the near and the far future, while also identifying the clinical data that mostly contribute to EMA predictions, (2) to identify users with outlier EMA, and (3) to show to what extent the EMAs of a user are in line with or diverge from those users similar to him/her. We report our findings based on a pilot study on patient empowerment, involving tinnitus patients who recorded EMAs with the mHealth app TinnitusTips. To validate our method, we also derived synthetic data from the same pilot study. Based on this setting, results for different use cases are reported.

Applying Machine Learning to Daily-Life Data From the TrackYourTinnitus Mobile Health Crowdsensing Platform to Predict the Mobile Operating System Used With High Accuracy: Longitudinal Observational Study.

BACKGROUND: Tinnitus is often described as the phantom perception of a sound and is experienced by 5.1% to 42.7% of the population worldwide, at least once during their lifetime. The symptoms often reduce the patient's quality of life. The TrackYourTinnitus (TYT) mobile health (mHealth) crowdsensing platform was developed for two operating systems (OS)-Android and iOS-to help patients demystify the daily moment-to-moment variations of their tinnitus symptoms. In all platforms developed for more than one OS, it is important to investigate whether the crowdsensed data predicts the OS that was used in order to understand the degree to which the OS is a confounder that is necessary to consider. OBJECTIVE: In this study, we explored whether the mobile OS-Android and iOS-used during user assessments can be predicted by the dynamic daily-life TYT data. METHODS: TYT mainly applies the paradigms ecological momentary assessment (EMA) and mobile crowdsensing to collect dynamic EMA (EMA-D) daily-life data. The dynamic daily-life TYT data that were analyzed included eight questions as part of the EMA-D questionnaire. In this study, 518 TYT users were analyzed, who each completed at least 11 EMA-D questionnaires. Out of these, 221 were iOS users and 297 were Android users. The iOS users completed, in total, 14,708 EMA-D questionnaires; the number of EMA-D questionnaires completed by the Android users was randomly reduced to the same number to properly address the research question of the study. Machine learning methods-a feedforward neural network, a decision tree, a random forest classifier, and a support vector machine-were applied to address the research question. RESULTS: Machine learning was able to predict the mobile OS used with an accuracy up to 78.94% based on the provided EMA-D questionnaires on the assessment level. In this context, the daily measurements regarding how users concentrate on the actual activity were particularly suitable for the prediction of the mobile OS used. CONCLUSIONS: In the work at hand, two particular aspects have been revealed. First, machine learning can contribute to EMA-D data in the medical context. Second, based on the EMA-D data of TYT, we found that the accuracy in predicting the mobile OS used has several implications. Particularly, in clinical studies using mobile devices, the OS should be assessed as a covariate, as it might be a confounder.

Efficient Processing of Geospatial mHealth Data Using a Scalable Crowdsensing Platform.

Smart sensors and smartphones are becoming increasingly prevalent. Both can be used to gather environmental data (e.g., noise). Importantly, these devices can be connected to each other as well as to the Internet to collect large amounts of sensor data, which leads to many new opportunities. In particular, mobile crowdsensing techniques can be used to capture phenomena of common interest. Especially valuable insights can be gained if the collected data are additionally related to the time and place of the measurements. However, many technical solutions still use monolithic backends that are not capable of processing crowdsensing data in a flexible, efficient, and scalable manner. In this work, an architectural design was conceived with the goal to manage geospatial data in challenging crowdsensing healthcare scenarios. It will be shown how the proposed approach can be used to provide users with an interactive map of environmental noise, allowing tinnitus patients and other health-conscious people to avoid locations with harmful sound levels. Technically, the shown approach combines cloud-native applications with Big Data and stream processing concepts. In general, the presented architectural design shall serve as a foundation to implement practical and scalable crowdsensing platforms for various healthcare scenarios beyond the addressed use case.

Dimensionality Reduction and Subspace Clustering in Mixed Reality for Condition Monitoring of High-Dimensional Production Data.

Visual analytics are becoming increasingly important in the light of big data and related scenarios. Along this trend, the field of immersive analytics has been variously furthered as it is able to provide sophisticated visual data analytics on one hand, while preserving user-friendliness on the other. Furthermore, recent hardware developments such as smart glasses, as well as achievements in virtual-reality applications, have fanned immersive analytic solutions. Notably, such solutions can be very effective when they are applied to high-dimensional datasets. Taking this advantage into account, the work at hand applies immersive analytics to a high-dimensional production dataset to improve the digital support of daily work tasks. More specifically, a mixed-reality implementation is presented that will support manufacturers as well as data scientists to comprehensively analyze machine data. As a particular goal, the prototype will simplify the analysis of manufacturing data through the usage of dimensionality reduction effects. Therefore, five aspects are mainly reported in this paper. First, it is shown how dimensionality reduction effects can be represented by clusters. Second, it is presented how the resulting information loss of the reduction is addressed. Third, the graphical interface of the developed prototype is illustrated as it provides (1) a correlation coefficient graph, (2) a plot for the information loss, and (3) a 3D particle system. In addition, an implemented voice recognition feature of the prototype is shown, which was considered to be being promising to select or deselect data variables users are interested in when analyzing the data. Fourth, based on a machine learning library, it is shown how the prototype reduces computational resources using smart glasses. The main idea is based on a recommendation approach as well as the use of subspace clustering. Fifth, results from a practical setting are presented, in which the prototype was shown to domain experts. The latter reported that such a tool is actually helpful to analyze machine data daily. Moreover, it was reported that such a system can be used to educate machine operators more properly. As a general outcome of this work, the presented approach may constitute a helpful solution for the industry as well as other domains such as medicine.

Heterogeneity in response to treatment across tinnitus phenotypes.

The clinical heterogeneity of chronic tinnitus poses major challenges to patient management and prompts the identification of distinct patient subgroups (or phenotypes) that respond more predictable to a particular treatment. We model heterogeneity in treatment response among phenotypes of tinnitus patients concerning their change in self-reported health burden, psychological characteristics, and tinnitus characteristics. Before and after a 7-day multimodal treatment, 989 tinnitus patients completed 14 assessment questionnaires, from which 64 variables measured general tinnitus characteristics, quality of life, pain experiences, somatic expressions, affective symptoms, tinnitus-related distress, internal resources, and perceived stress. Our approach encompasses mechanisms for patient phenotyping, visualizations of the phenotypes and their change with treatment in a projected space, and the extraction of patient subgroups based on their change with treatment. On average, all four distinct phenotypes identified at the pre-intervention baseline showed improved values for nearly all the considered variables following the intervention. However, a considerable intra-phenotype heterogeneity was noted. Five clusters of change reflected variations in the observed improvements among individuals. These patterns of treatment effects were identified to be associated with baseline phenotypes. Our exploratory approach establishes a groundwork for future studies incorporating control groups to pinpoint patient subgroups that are more likely to benefit from specific treatments. This strategy not only has the potential to advance personalized medicine but can also be extended to a broader spectrum of patients with various chronic conditions.

Prediction meets time series with gaps: User clusters with specific usage behavior patterns.

With mHealth apps, data can be recorded in real life, which makes them useful, for example, as an accompanying tool in treatments. However, such datasets, especially those based on apps with usage on a voluntary basis, are often affected by fluctuating engagement and by high user dropout rates. This makes it difficult to exploit the data using machine learning techniques and raises the question of whether users have stopped using the app. In this extended paper, we present a method to identify phases with varying dropout rates in a dataset and predict for each. We also present an approach to predict what period of inactivity can be expected for a user in the current state. We use change point detection to identify the phases, show how to deal with uneven misaligned time series and predict the user's phase using time series classification. In addition, we examine how the evolution of adherence develops in individual clusters of individuals. We evaluated our method on the data of an mHealth app for tinnitus, and show that our approach is appropriate for the study of adherence in datasets with uneven, unaligned time series of different lengths and with missing values.

Prestimulus α/ß power in temporal-order judgments: individuals differ in direction of modulation but show consistency over auditory and visual tasks.

The processing of incoming sensory information can be differentially affected by varying levels of α-power in the electroencephalogram (EEG). A prominent hypothesis is that relatively low prestimulus α-power is associated with improved perceptual performance. However, there are studies in the literature that do not fit easily into this picture, and the reasons for this are poorly understood and rarely discussed. To evaluate the robustness of previous findings and to better understand the overall mixed results, we used a spatial TOJ task in which we presented auditory and visual stimulus pairs in random order while recording EEG. For veridical and non-veridical TOJs, we calculated the power spectral density (PSD) for 3 frequencies (5 Hz steps: 10, 15, and 20 Hz). We found on the group level: (1) Veridical auditory TOJs, relative to non-veridical, were associated with higher ß-band (20 Hz) power over central electrodes. (2) Veridical visual TOJs showed higher ß-band (10, 15 Hz) power over parieto-occipital electrodes (3) Electrode site interacted with TOJ condition in the ß-band: For auditory TOJs, PSD over central electrodes was higher for veridical than non-veridical and over parieto-occipital electrodes was lower for veridical than non-veridical trials, while the latter pattern was reversed for visual TOJs. While our group-level result showed a clear direction of prestimulus modulation, the individual-level modulation pattern was variable and included activations opposite to the group mean. Interestingly, our results at the individual-level mirror the situation in the literature, where reports of group-level prestimulus modulation were found in either direction. Because the direction of individual activation of electrodes over auditory brain regions and parieto-occipital electrodes was always negatively correlated in the respective TOJ conditions, this activation opposite to the group mean cannot be easily dismissed as noise. The consistency of the individual-level data cautions against premature generalization of group-effects and suggests different strategies that participants initially adopted and then consistently followed. We discuss our results in light of probabilistic information processing and complex system properties, and suggest that a general description of brain activity must account for variability in modulation directions at both the group and individual levels.

Stability of mental motor-imagery classification in EEG depends on the choice of classifier model and experiment design, but not on signal preprocessing.

Introduction: Modern consciousness research has developed diagnostic tests to improve the diagnostic accuracy of different states of consciousness via electroencephalography (EEG)-based mental motor imagery (MI), which is still challenging and lacks a consensus on how to best analyse MI EEG-data. An optimally designed and analyzed paradigm must detect command-following in all healthy individuals, before it can be applied in patients, e.g., for the diagnosis of disorders of consciousness (DOC). Methods: We investigated the effects of two important steps in the raw signal preprocessing on predicting participant performance (F1) and machine-learning classifier performance (area-under-curve, AUC) in eight healthy individuals, that are based solely on MI using high-density EEG (HD-EEG): artifact correction (manual correction with vs. without Independent Component Analysis [ICA]), region of interest (ROI; motor area vs. whole brain), and machine-learning algorithm (support-vector machine [SVM] vs. k-nearest neighbor [KNN]). Results: Results revealed no significant effects of artifact correction and ROI on predicting participant performance (F1) and classifier performance (AUC) scores (all ps > 0.05) in the SVM classification model. In the KNN model, ROI had a significant influence on the classifier performance [F(1,8.939) = 7.585, p = 0.023]. There was no evidence for artifact correction and ROI selection changing the prediction of participants performance and classifier performance in EEG-based mental MI if using SVM-based classification (71-100% correct classifications across different signal preprocessing methods). The variance in the prediction of participant performance was significantly higher when the experiment started with a resting-state compared to a mental MI task block [X2(1) = 5.849, p = 0.016]. Discussion: Overall, we could show that classification is stable across different modes of EEG signal preprocessing when using SVM models. Exploratory analysis gave a hint toward potential effects of the sequence of task execution on the prediction of participant performance, which should be taken into account in future studies.

The statistical analysis plan for the unification of treatments and interventions for tinnitus patients randomized clinical trial (UNITI-RCT).

BACKGROUND: Tinnitus is a leading cause of disease burden globally. Several therapeutic strategies are recommended in guidelines for the reduction of tinnitus distress; however, little is known about the potentially increased effectiveness of a combination of treatments and personalized treatments for each tinnitus patient. METHODS: Within the Unification of Treatments and Interventions for Tinnitus Patients project, a multicenter, randomized clinical trial is conducted with the aim to compare the effectiveness of single treatments and combined treatments on tinnitus distress (UNITI-RCT). Five different tinnitus centers across Europe aim to treat chronic tinnitus patients with either cognitive behavioral therapy, sound therapy, structured counseling, or hearing aids alone, or with a combination of two of these treatments, resulting in four treatment arms with single treatment and six treatment arms with combinational treatment. This statistical analysis plan describes the statistical methods to be deployed in the UNITI-RCT. DISCUSSION: The UNITI-RCT trial will provide important evidence about whether a combination of treatments is superior to a single treatment alone in the management of chronic tinnitus patients. This pre-specified statistical analysis plan details the methodology for the analysis of the UNITI trial results. TRIAL REGISTRATION: ClinicalTrials.gov NCT04663828 . The trial is ongoing. Date of registration: December 11, 2020. All patients that finished their treatment before 19 December 2022 are included in the main RCT analysis.

Predicting Ecological Momentary Assessments in an App for Tinnitus by Learning From Each User's Stream With a Contextual Multi-Armed Bandit.

Ecological Momentary Assessments (EMA) deliver insights on how patients perceive tinnitus at different times and how they are affected by it. Moving to the next level, an mHealth app can support users more directly by predicting a user's next EMA and recommending personalized services based on these predictions. In this study, we analyzed the data of 21 users who were exposed to an mHealth app with non-personalized recommendations, and we investigate ways of predicting the next vector of EMA answers. We studied the potential of entity-centric predictors that learn for each user separately and neighborhood-based predictors that learn for each user separately but take also similar users into account, and we compared them to a predictor that learns from all past EMA indiscriminately, without considering which user delivered which data, i.e., to a "global model." Since users were exposed to two versions of the non-personalized recommendations app, we employed a Contextual Multi-Armed Bandit (CMAB), which chooses the best predictor for each user at each time point, taking each user's group into account. Our analysis showed that the combination of predictors into a CMAB achieves good performance throughout, since the global model was chosen at early time points and for users with few data, while the entity-centric, i.e., user-specific, predictors were used whenever the user had delivered enough data-the CMAB chose itself when the data were "enough." This flexible setting delivered insights on how user behavior can be predicted for personalization, as well as insights on the specific mHealth data. Our main findings are that for EMA prediction the entity-centric predictors should be preferred over a user-insensitive global model and that the choice of EMA items should be further investigated because some items are answered more rarely than others. Albeit our CMAB-based prediction workflow is robust to differences in exposition and interaction intensity, experimentators that design studies with mHealth apps should be prepared to quantify and closely monitor differences in the intensity of user-app interaction, since users with many interactions may have a disproportionate influence on global models.

Standardized Clinical Profiling in Spanish Patients with Chronic Tinnitus.

BACKGROUND: Tinnitus is a heterogeneous condition. The aim of this study as to compare the online and hospital responses to the Spanish version of European School for Interdisciplinary Tinnitus Research screening-questionnaire (ESIT-SQ) in tinnitus individuals by an unsupervised age clustering. METHODS: A cross-sectional study was performed including 434 white Spanish patients with chronic tinnitus to assess the demographic and clinical profile through the ESIT-SQ, with 204 outpatients and 230 individuals from an online survey; a K-means clustering algorithm was used to classify both responses according to age. RESULTS: Online survey showed a high proportion of Meniere's disease (MD) patients compared to both the general population and the outpatient cohort. The responses showed statistically significant differences between groups regarding education level, tinnitus-related hearing disorders (MD, hyperacusis), sleep difficulties, dyslipidemia, and other tinnitus characteristics, including duration, type of onset, the report of mitigating factors and the use of treatments. However, these differences were partially confirmed after adjusting for age. CONCLUSIONS: Self-reported tinnitus surveys are a low confidence source for tinnitus phenotyping. Additional clinical evaluation is needed for tinnitus research to reach the diagnosis. Age-based cluster analysis might help to better define clinical profiles and to compare responses in ESIT-SQ among subgroups of patients with tinnitus.

Smartphone-Guided Educational Counseling and Self-Help for Chronic Tinnitus.

Tinnitus is an auditory phantom perception in the ears or head in the absence of a corresponding external stimulus. There is currently no effective treatment available that reliably reduces tinnitus. Educational counseling is a treatment approach that aims to educate patients and inform them about possible coping strategies. For this feasibility study, we implemented educational material and self-help advice in a smartphone app. Participants used the educational smartphone app unsupervised during their daily routine over a period of four months. Comparing the tinnitus outcome measures before and after smartphone-guided treatment, we measured changes in tinnitus-related distress, but not in tinnitus loudness. Improvements on the Tinnitus Severity numeric rating scale reached an effect size of 0.408, while the improvements on the Tinnitus Handicap Inventory (THI) were much smaller with an effect size of 0.168. An analysis of user behavior showed that frequent and intensive use of the app is a crucial factor for treatment success: participants that used the app more often and interacted with the app intensively reported a stronger improvement in the tinnitus. Between study allocation and final assessment, 26 of 52 participants dropped out of the study. Reasons for the dropouts and lessons for future studies are discussed in this paper.

Juxtaposing Medical Centers Using Different Questionnaires Through Score Predictors.

Background: Chronic tinnitus is a clinically multidimensional phenomenon that entails audiological, psychological and somatosensory components. Previous research has demonstrated age and female gender as potential risk factors, although studies to this regard are heterogeneous. Moreover, whilst recent research has begun to identify clinical "phenotypes," little is known about differences in patient population profiles at geographically separated and specialized treatment centers. Identifying such differences might prevent potential biases in joint randomized controlled trials (RCTs) and allow for population-specific treatment adaptations. Method: Two German tinnitus treatment centers were compared regarding pre-treatment data distributions of their patient population bases. To identify overlapping as well as center-specific factors, juxtaposition-, similarity-, and meta-data-based methods were applied. Results: Between centers, significant differences emerged. One center demonstrated some predictive power of the patients of the other center with regard to questionnaire score after treatment, indicating similarities in treatment response across center populations. Furthermore, adherence to the completion of the questionnaires was found to be an important factor in predicting post-treatment data. Discussion: Differential age and gender distributions per center should be considered as regards RCT design and individualized treatment planning.

Dimensions of Tinnitus-Related Distress.

OBJECTIVES: (1) To determine which psychosocial aspects predict tinnitus-related distress in a large self-reported dataset of patients with chronic tinnitus, and (2) to identify underlying constructs by means of factor analysis. METHODS: A cohort of 1958 patients of the Charité Tinnitus Center, Berlin completed a large questionnaire battery that comprised sociodemographic data, tinnitus-related distress, general psychological stress experience, emotional symptoms, and somatic complaints. To identify a construct of "tinnitus-related distress", significant predictive items were grouped using factor analysis. RESULTS: For the prediction of tinnitus-related distress (linear regression model with R2 = 0.7), depressive fatigue symptoms (concentration, sleep, rumination, joy decreased), the experience of emotional strain, somatization tendencies (pain experience, doctor contacts), and age appeared to play a role. The factor analysis revealed five factors: "stress", "pain experience", "fatigue", "autonomy", and low "educational level". CONCLUSIONS: Tinnitus-related distress is predicted by psychological and sociodemographic indices. Relevant factors seem to be depressive exhaustion with somatic expressions such as sleep and concentration problems, somatization, general psychological stress, and reduced activity, in addition to higher age.

Assessing the difficulty of annotating medical data in crowdworking with help of experiments.

BACKGROUND: As healthcare-related data proliferate, there is need to annotate them expertly for the purposes of personalized medicine. Crowdworking is an alternative to expensive expert labour. Annotation corresponds to diagnosis, so comparing unlabeled records to labeled ones seems more appropriate for crowdworkers without medical expertise. We modeled the comparison of a record to two other records as a triplet annotation task, and we conducted an experiment to investigate to what extend sensor-measured stress, task duration, uncertainty of the annotators and agreement among the annotators could predict annotation correctness. MATERIALS AND METHODS: We conducted an annotation experiment on health data from a population-based study. The triplet annotation task was to decide whether an individual was more similar to a healthy one or to one with a given disorder. We used hepatic steatosis as example disorder, and described the individuals with 10 pre-selected characteristics related to this disorder. We recorded task duration, electro-dermal activity as stress indicator, and uncertainty as stated by the experiment participants (n = 29 non-experts and three experts) for 30 triplets. We built an Artificial Similarity-Based Annotator (ASBA) and compared its correctness and uncertainty to that of the experiment participants. RESULTS: We found no correlation between correctness and either of stated uncertainty, stress and task duration. Annotator agreement has not been predictive either. Notably, for some tasks, annotators agreed unanimously on an incorrect annotation. When controlling for Triplet ID, we identified significant correlations, indicating that correctness, stress levels and annotation duration depend on the task itself. Average correctness among the experiment participants was slightly lower than achieved by ASBA. Triplet annotation turned to be similarly difficult for experts as for non-experts. CONCLUSION: Our lab experiment indicates that the task of triplet annotation must be prepared cautiously if delegated to crowdworkers. Neither certainty nor agreement among annotators should be assumed to imply correct annotation, because annotators may misjudge difficult tasks as easy and agree on incorrect annotations. Further research is needed to improve visualizations for complex tasks, to judiciously decide how much information to provide, Out-of-the-lab experiments in crowdworker setting are needed to identify appropriate designs of a human-annotation task, and to assess under what circumstances non-human annotation should be preferred.

Reasons for Discontinuing Active Participation on the Internet Forum Tinnitus Talk: Mixed Methods Citizen Science Study.

BACKGROUND: Tinnitus Talk is a nonprofit online self-help forum. Asking inactive users about their reasons for discontinued usage of health-related online platforms such as Tinnitus Talk is important for quality assurance. OBJECTIVE: The aim of this study was to explore reasons for discontinued use of Tinnitus Talk, and their associations to the perceptions of Tinnitus Talk and the age of users who ceased logging on to the platform. METHODS: Initially, 13,745 users that did not use Tinnitus Talk within the previous 2 months were contacted and the response rate was 20.47% (n=2814). After dataset filtering, a total of 2172 past members of Tinnitus Talk were included in the analyses. Nine predefined reasons for discontinued usage of Tinnitus Talk were included in the survey as well as one open question. Moreover, there were 14 predefined questions focusing on perception of Tinnitus Talk (usefulness, content, community, and quality of members' posts). Mixed methods analyses were performed. Frequencies and correlation coefficients were calculated for quantitative data, and grounded theory methodology was utilized for exploration of the qualitative data. RESULTS: Quantitative analysis revealed reasons for discontinued use of Tinnitus Talk as well as associations of these reasons with perceptions of Tinnitus Talk and age. Among the eight predefined reasons for discontinued use of Tinnitus Talk, the most frequently reported was not finding the information they were looking for (451/2695, 16.7%). Overall, the highest rated perception of Tinnitus Talk was content-related ease of understanding (mean 3.9, SD 0.64). A high number (nearly 40%) of participants provided additional free text explaining why they discontinued use. Qualitative analyses identified a total of 1654 specific reasons, more than 93% of which (n=1544) could be inductively coded. The coding system consisted of 33 thematically labeled codes clustered into 10 categories. The most frequent additional reason for discontinuing use was thinking that there is no cure or help for tinnitus symptoms (375/1544, 24.3%). Significant correlations (P<.001) were observed between reasons for discontinued usage and perception of Tinnitus Talk. Several reasons for discontinued usage were associated with the examined dimensions of perception of Tinnitus Talk (usefulness, content, community, as well as quality of members' posts). Moreover, significant correlations (P<.001) between age and reasons for discontinued use were found. Older age was associated with no longer using Tinnitus Talk because of not finding what they were looking for. In addition, older participants had a generally less positive perception of Tinnitus Talk than younger participants (P<.001). CONCLUSIONS: This study contributes to understanding the reasons for discontinued usage of online self-help platforms, which are typically only reported according to the dropout rates. Furthermore, specific groups of users who did not benefit from Tinnitus Talk were identified, and several practical implications for improvement of the structure, content, and goals of Tinnitus Talk were suggested.

Immune-checkpoint blockade of CTLA-4 (CD152) in antigen-specific human T-cell responses differs profoundly between neonates, children, and adults.

The monoclonal antibody against CTLA-4, Ipilimumab, is a first-in-class immune-checkpoint inhibitor approved for treatment of advanced melanoma in adults but not extensively studied in children. In light of the fact that the immune response early in life differs from that of adults, we have applied a human in vitro model stimulating CD4+ T-cells from neonates, children (1-5 years), and adults antigen-specifically with Staphylococcus aureus (S. aureus) for assessment of CTLA-4 blockade early in life. We show that T-cell proliferation as well as frequencies of antigen-specific T-cells (CD40L+CD4+) were enhanced in neonatal T-cells upon CTLA-4 blockade showing a larger variance within the group (F-test p < .0001). Using machine learning algorithm Random Forest, adult and neonatal T-cell responses can be unambiguously categorized (F1 score-0.75) on the basis of their cytokine (co-)expression. Blockade of CTLA-4 enhanced frequencies of IL-8, IFNγ, and IL-10 producers among CD40L+ T-cells. Of note, antigen-specific T-cells from neonates displayed higher cytokine coproduction at baseline, while T-cells from children caught up to neonates, and adults to baseline of children upon CTLA-4 blockade. These findings reveal that in neonatal T-cells blockade of CTLA-4 mainly unleashes the antigen-specific capacity by increasing the numbers of responding T-cells, whereas in children and adults it promotes the coexpression of cytokines by individual T-cells. Thus, CTLA-4 blockade boosts antitumor immunity through different mechanisms depending on the patients' age. These data implicate a strong impact of the developmental stage of the T-cell compartment on the effects of immune-checkpoint therapy.

Using Big Data to Develop a Clinical Decision Support System for Tinnitus Treatment.

Tinnitus is a common symptom of a phantom sound perception with a considerable socioeconomic impact. Tinnitus pathophysiology is enigmatic and its significant heterogeneity reflects a wide spectrum of clinical manifestations, severity and annoyance among tinnitus sufferers. Although several interventions have been suggested, currently there is no universally accepted treatment. Moreover, there is no well-established correlation between tinnitus features or patients' characteristics and projection of treatment response. At the clinical level, this practically means that selection of treatment is not based on expected outcomes for the particular patient.The complexity of tinnitus and lack of well-adapted prognostic factors for treatment selection highlight a potential role for a decision support system (DSS). A DSS is an informative system, based on big data that aims to facilitate decision-making based on: specific rules, retrospective data reflecting results, patient profiling and predictive models. Therefore, it can use algorithms evaluating numerous parameters and indicate the weight of their contribution to the final outcome. This means that DSS can provide additional information, exceeding the typical questions of superiority of one treatment versus another, commonly addressed in literature.The development of a DSS for tinnitus treatment selection will make use of an underlying database consisting of medical, epidemiological, audiological, electrophysiological, genetic and tinnitus subtyping data. Algorithms will be developed with the use of machine learning and data mining techniques. Based on the profile features identified as prognostic these algorithms will be able to suggest whether additional examinations are needed for a robust result as well as which treatment or combination of treatments is optimal for every patient in a personalized level.In this manuscript we carefully define the conceptual basis for a tinnitus treatment selection DSS. We describe the big data set and the knowledge base on which the DSS will be based and the algorithms that will be used for prognosis and treatment selection.