General movements (GMs) are spontaneous movements of infants up to five months post-term involving the whole body varying in sequence, speed, and amplitude. The assessment of GMs has shown its importance for identifying infants at risk for neuromotor deficits, especially for the detection of cerebral palsy. As the assessment is based on videos of the infant that are rated by trained professionals, the method is time-consuming and expensive. Therefore, approaches based on Artificial Intelligence have gained significantly increased attention in the last years. In this article, we systematically analyze and discuss the main design features of all existing technological approaches seeking to transfer the Prechtl’s assessment of general movements from an individual visual perception to computer-based analysis. After identifying their shared shortcomings, we explain the methodological reasons for their limited practical performance and classification rates. As a conclusion of our literature study, we conceptually propose a methodological solution to the defined problem based on the groundbreaking innovation in the area of Deep Learning.

While even the most common definition of pain is under debate, pain assessment has remained the same for decades. But the paramount importance of precise pain management for successful healthcare has encouraged initiatives to improve the way pain is assessed. Recent approaches have proposed automatic pain evaluation systems using machine learning models trained with data coming from behavioural or physiological sensors. Although yielding promising results, machine learning studies for sensor-based pain recognition remain scattered and not necessarily easy to compare to each other. In particular, the important process of extracting features is usually optimised towards specific datasets. We thus introduce a comparison of feature extraction methods for pain recognition based on physiological sensors in this paper. In addition, the PainMonit Database (PMDB), a new dataset including both objective and subjective annotations for heat-induced pain in 52 subjects, is introduced. In total, five different approaches including techniques based on feature engineering and feature learning with deep learning are evaluated on the BioVid and PMDB datasets. Our studies highlight the following insights: (1) Simple feature engineering approaches can still compete with deep learning approaches in terms of performance. (2) More complex deep learning architectures do not yield better performance compared to simpler ones. (3) Subjective self-reports by subjects can be used instead of objective temperature-based annotations to build a robust pain recognition system.

To calibrate or not to calibrate? This question is raised by almost everyone designing an experimental pain study with supra-threshold stimulation. The dilemma is whether to individualize stimulus intensity to the pain threshold / supra-threshold pain level of each participant or whether to provide the noxious stimulus at a fixed intensity so that everyone receives the identical input. Each approach has unique pros and cons which need to be considered to i) accurately design an experiment, ii) enhance statistical inference in the given data and, iii) reduce bias and the influence of confounding factors in the individual study e.g., body composition, differences in energy absorption and previous experience. Individualization requires calibration, a procedure already irritating the nociceptive system but allowing to match the pain level across individuals. It leads to a higher variability of the stimulus intensity, thereby influencing the encoding of "noxiousness" by the central nervous system. Results might be less influenced by statistical phenomena such as ceiling/floor effects and the approach does not seem to rise ethical concerns. On the other hand, applying a fixed (standardized) intensity reduces the problem of intensity encoding leading to a large between-subjects variability in pain responses. Fixed stimulation intensities do not require pre-exposure. It can be proposed that one method is not preferable over another, however the choice depends on the study aim and the desired level of external validity. This paper discusses considerations for choosing the optimal approach for experimental pain studies and provides recommendations for different study designs. PERSPECTIVE: To calibrate pain or not? This dilemma is related to almost every experimental pain research. The decision is a trade-off between statistical power and greater control of stimulus encoding. The article decomposes both approaches and presents the pros and cons of either approach supported by data and simulation experiment.

Artificial intelligence and especially deep learning methods have achieved outstanding results for various applications in the past few years. Pain recognition is one of them, as various models have been proposed to replace the previous gold standard with an automated and objective assessment. While the accuracy of such models could be increased incrementally, the understandability and transparency of these systems have not been the main focus of the research community thus far. Thus, in this work, several outcomes and insights of explainable artificial intelligence applied to the electrodermal activity sensor data of the PainMonit and BioVid Heat Pain Database are presented. For this purpose, the importance of hand-crafted features is evaluated using recursive feature elimination based on impurity scores in Random Forest (RF) models. Additionally, Gradient-weighted class activation mapping is applied to highlight the most impactful features learned by deep learning models. Our studies highlight the following insights: (1) Very simple hand-crafted features can yield comparative performances to deep learning models for pain recognition, especially when properly selected with recursive feature elimination. Thus, the use of complex neural networks should be questioned in pain recognition, especially considering their computational costs; and (2) both traditional feature engineering and deep feature learning approaches rely on simple characteristics of the input time-series data to make their decision in the context of automated pain recognition.

The perception of hunger and satiety is of great importance to maintaining a healthy body weight and avoiding chronic diseases such as obesity, underweight, or deficiency syndromes due to malnutrition. There are a number of disease patterns, characterized by a chronic loss of this perception. To our best knowledge, hunger and satiety cannot be classified using non-invasive measurements. Aiming to develop an objective classification system, this paper presents a multimodal sensory system using associated signal processing and pattern recognition methods for hunger and satiety detection based on non-invasive monitoring. We used an Empatica E4 smartwatch, a RespiBan wearable device, and JINS MEME smart glasses to capture physiological signals from five healthy normal weight subjects inactively sitting on a chair in a state of hunger and satiety. After pre-processing the signals, we compared different feature extraction approaches, either based on manual feature engineering or deep feature learning. Comparative experiments were carried out to determine the most appropriate sensor channel, device, and classifier to reliably discriminate between hunger and satiety states. Our experiments showed that the most discriminative features come from three specific sensor modalities: Electrodermal Activity (EDA), infrared Thermopile (Tmp), and Blood Volume Pulse (BVP).

Access to large amounts of data is essential for successful machine learning research. However, there is insufficient data for many applications, as data collection is often challenging and time-consuming. The same applies to automated pain recognition, where algorithms aim to learn associations between a level of pain and behavioural or physiological responses. Although machine learning models have shown promise in improving the current gold standard of pain monitoring (self-reports) only a handful of datasets are freely accessible to researchers. This paper presents the PainMonit Dataset for automated pain detection using physiological data. The dataset consists of two parts, as pain can be perceived differently depending on its underlying cause. (1) Pain was triggered by heat stimuli in an experimental study during which nine physiological sensor modalities (BVP, 2×EDA, skin temperature, ECG, EMG, IBI, HR, respiration) were recorded from 55 healthy subjects. (2) Eight modalities (2×BVP, 2×EDA, EMG, skin temperature, respiration, grip) were recorded from 49 participants to assess their pain during a physiotherapy session.