The purpose of our study was to identify predictors of a high risk of involuntary psychiatric in-patient treatment. We carried out a detailed analysis of the 1773 mental health records of all the persons treated as in-patients under the PsychKG NRW (Mental Health Act for the state of North Rhine-Westphalia, Germany) in a metropolitan region of Germany (the City of Cologne) in 2011. 3991 mental health records of voluntary in-patients from the same hospitals served as a control group. We extracted medical, sociodemographic and socioeconomic data from these records. Apart from descriptive statistics, we used a prediction model employing chi-squared automatic interaction detection (CHAID). Among involuntary patients, organic mental disorders (ICD10: F0) and schizophrenia and other psychotic disorders (ICD10: F2) were overrepresented. Patients treated as in-patients against their will were on average older, they were more often retired and had a migratory background. The Exhaustive CHAID analysis confirmed the main diagnosis to be the strongest predictor of involuntary in-patient psychiatric treatment. Other predictors were the absence of outpatient treatment prior to admission, admission outside of regular service hours and migratory background. The highest risk of involuntary treatment was associated with patients with organic mental disorders (ICD 10: F0) who were married or widowed and patients with non-organic psychotic disorders (ICD10: F2) or mental retardation (ICD10: F7) in combination with a migratory background. Also, referrals from general hospitals were frequently encountered. We identified modifiable risk factors for involuntary psychiatric in-patient treatment. This implies that preventive measures may be feasible and should be implemented to reduce the rate of involuntary psychiatric in-patient treatment. This may include efforts to establish crisis resolution teams to improve out-patient treatment, train general hospital staff in deescalation techniques, and develop special programs for patients with a migratory background.

The aim of this study was to develop classification models for mastitis and lameness treatments in Holstein dairy cows as the target variables based on continuous data from herd management software with modern machine learning methods. Data was collected over a period of 40 months from a total of 167 different cows with daily individual sensor information containing milking parameters, pedometer activity, feed and water intake, and body weight (in the form of differently aggregated data) as well as the entered treatment data. To identify the most important predictors for mastitis and lameness treatments, respectively, Random Forest feature importance, Pearson’s correlation and sequential forward feature selection were applied. With the selected predictors, various machine learning models such as Logistic Regression (LR), Support Vector Machine (SVM), K-nearest neighbors (KNN), Gaussian Naïve Bayes (GNB), Extra Trees Classifier (ET) and different ensemble methods such as Random Forest (RF) were trained. Their performance was compared using the receiver operator characteristic (ROC) area-under-curve (AUC), as well as sensitivity, block sensitivity and specificity. In addition, sampling methods were compared: Over- and undersampling as compensation for the expected unbalanced training data had a high impact on the ratio of sensitivity and specificity in the classification of the test data, but with regard to AUC, random oversampling and SMOTE (Synthetic Minority Over-sampling) even showed significantly lower values than with non-sampled data. The best model, ET, obtained a mean AUC of 0.79 for mastitis and 0.71 for lameness, respectively, based on testing data from practical conditions and is recommended by us for this type of data, but GNB, LR and RF were only marginally worse, and random oversampling and SMOTE even showed significantly lower values than without sampling. We recommend the use of these models as a benchmark for similar self-learning classification tasks. The classification models presented here retain their interpretability with the ability to present feature importances to the farmer in contrast to the “black box” models of Deep Learning methods.

The prediction of health disorders is the goal of many sensor systems in dairy farming. Although mastitis and lameness are the most common health disorders in dairy cows, these diseases or treatments are a rare event related to a single day and cow. A number of studies already developed and evaluated models for classifying cows in need of treatment for mastitis and lameness with machine learning methods, but few have illustrated the effects of the positive predictive value (PPV) on practical application. The objective of this study was to investigate the importance of low-frequency treatments of mastitis or lameness for the applicability of these classification models in practice. Data from three German dairy farms contained animal individual sensor data (milkings, activity, feed intake) and were classified using machine learning models developed in a previous study. Subsequently, different risk criteria (previous treatments, information from milk recording, early lactation) were designed to isolate high-risk groups. Restricting selection to cows with previous mastitis or hoof treatment achieved the highest increase in PPV from 0.07 to 0.20 and 0.15, respectively. However, the known low daily risk of a treatment per cow remains the critical factor that prevents the reduction of daily false-positive alarms to a satisfactory level. Sensor systems should be seen as additional decision-support aid to the farmers’ expert knowledge.

Emotions influence processes of learning and thinking in all people. However, there is a lack of studies in the field of emotion research including people with intellectual disabilities (ID) addressing the existing diversity. The present study investigates the emotional competence of people with ID (N = 32). The first aim was to assess the emotional development using the Scale of Emotional Development (SEED). Based on these insights, the second objective was to replicate existing findings, validating the emotional reaction of people with ID to pictures of the International Affective Picture System (IAPS) based on self-reports. In an additional pilot-like analysis, the third aim was to investigate if these self-reported emotional reactions match the emotions expressed in their faces using the automated and video-based facial expression analysis software ‘Affectiva (Affdex SDK)’. In the present study, the self-reported emotional reactions of participants with ID were in line with previous research. In addition, the present study shows the general potential of this innovative approach of using and applying commercially available automated emotion recognition software for the field of special needs and social science.