The problem of continuous emotion recognition has been the subject of several studies. The proposed affective computing approaches employ sequential machine learning algorithms for improving the classification stage, accounting for the time ambiguity of emotional responses. Modeling and predicting the affective state over time is not a trivial problem because continuous data labeling is costly and not always feasible. This is a crucial issue in real-life applications, where data labeling is sparse and possibly captures only the most important events rather than the typical continuous subtle affective changes that occur. In this work, we introduce a framework from the machine learning literature called Multiple Instance Learning, which is able to model time intervals by capturing the presence or absence of relevant states, without the need to label the affective responses continuously (as required by standard sequential learning approaches). This choice offers a viable and natural solution for learning in a weakly supervised setting, taking into account the ambiguity of affective responses. We demonstrate the reliability of the proposed approach in a gold-standard scenario and towards real-world usage by employing an existing dataset (DEAP) and a purposely built one (Consumer). We also outline the advantages of this method with respect to standard supervised machine learning algorithms.

Automatic emotion recognition has powerful opportunities in the clinical field, but several critical aspects are still open, such as heterogeneity of methodologies or technologies tested mainly on healthy people. This systematic review aims to survey automatic emotion recognition systems applied in real clinical contexts, to deeply analyse clinical and technical aspects, how they were addressed, and relationships among them. The literature review was conducted on: IEEEXplore, ScienceDirect, Scopus, PubMed, ACM. Inclusion criteria were the presence of an automatic emotion recognition algorithm and the enrollment of at least 2 patients in the experimental protocol. The review process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Moreover, the works were analysed according to a reference model to deeply examine both clinical and technical topics. 52 scientific papers passed inclusion criteria. Most clinical scenarios involved neurodevelopmental, neurological and psychiatric disorders with the aims of diagnosing, monitoring, or treating emotional symptoms. The most adopted signals are video and audio, while supervised shallow learning is mostly used for emotion recognition. A poor study design, tiny samples, and the absence of a control group emerged as methodological weaknesses. Heterogeneity of performance metrics, datasets and algorithms challenges results comparability, robustness, reliability and reproducibility.

Detecting stress in computer users, while technically challenging, is of the utmost importance in the workplace, especially now that remote working scenarios are becoming ubiquitous. In this context, cost-effective, subject-independent systems are needed that can be embedded in consumer devices and classify users’ stress in a reliable and unobtrusive fashion. Leveraging keyboard and mouse dynamics is particularly appealing in this context as it exploits readily available sensors. However, available studies are mostly performed in laboratory conditions, and there is a lack of on-field investigations in closer-to-real-world settings. In this study, keyboard and mouse data from 62 volunteers were experimentally collected in-the-wild using a purpose-built Web application, designed to induce stress by asking each subject to perform 8 computer tasks under different stressful conditions. The application of Multiple Instance Learning (MIL) to Random Forest (RF) classification allowed the devised system to successfully distinguish 3 stress-level classes from keyboard (76% accuracy) and mouse (63% accuracy) data. Classifiers were further evaluated via confusion matrix, precision, recall, and F1-score.

Background: Wearable sensors are a promising tool for the remote, continuous monitoring of motor symptoms and physical activity, especially in individuals with neurological or chronic conditions. Despite many experimental trials, clinical adoption remains limited. A major barrier is the lack of awareness and confidence among healthcare professionals in these technologies. Methods: This systematic review analyzed the use of wearable sensors for continuous motor monitoring at home, focusing on their purpose, type, feasibility, and effectiveness in neurological, musculoskeletal, or rheumatologic conditions. This review followed PRISMA guidelines and included studies from PubMed, Scopus, and Web of Science. Results: Seventy-two studies with 7949 participants met inclusion criteria. Neurological disorders, particularly Parkinson’s disease, were the most frequently studied. Common sensors included inertial measurement units (IMUs), accelerometers, and gyroscopes, often integrated into medical devices, smartwatches, or smartphones. Monitoring periods ranged from 24 h to over two years. Feasibility studies showed high patient compliance (≥70%) and good acceptance, with strong agreement with clinical assessments. However, only half of the studies were controlled trials, and just 5.6% were randomized. Conclusions: Wearable sensors offer strong potential for real-world motor function monitoring. Yet, challenges persist, including ethical issues, data privacy, standardization, and healthcare access. Artificial intelligence integration may boost predictive accuracy and personalized care.