Head movements and postures as pain behavior

Development, refinement, and validation of an equine musculoskeletal pain scale

Musculoskeletal disease is a common cause of chronic pain that is often overlooked and inadequately treated, impacting the quality of life of humans and horses alike. Lameness due to musculoskeletal pain is prevalent in horses, but the perception of pain by owners is low compared with veterinary diagnosis. Therefore, this study aims to establish and validate a pain scale for chronic equine orthopaedic pain that is user-friendly for horse owners and veterinarians to facilitate the identification and monitoring of pain in horses. The newly developed musculoskeletal pain scale (MPS) was applied to 154 horses (mean age 20 ± 6.4 years SD) housed at an equine sanctuary, of which 128 (83%) suffered from chronic orthopaedic disease. To complete the MPS, the horses were observed and videotaped from a distance while at rest in their box or enclosure. In addition, they received a complete clinical and orthopaedic exam. The need for veterinary intervention to address pain (assessed and executed by the sanctuary independent from this study) was used as a longitudinal health outcome to determine the MPS's predictive validity. To determine the interrater agreement, the MPS was scored for a randomly selected subset of 30 horses by six additional blinded raters, three equine veterinary practitioners, and three experienced equestrians. An iterative process was used to refine the tool based on improvements in the MPS's correlation with lameness evaluated at the walk and trot, predictive validity for longitudinal health outcomes, and interrater agreement. The intraclass correlation improved from 0.77 of the original MPS to 0.88 of the refined version (95% confidence interval: 0.8-0.94). The refined MPS correlated significantly with lameness at the walk (r = 0.44, p = 0.001) and trot (r = 0.5, p < 0.0001). The refined MPS significantly differed between horses that needed veterinary intervention (mean MPS = 8.6) and those that did not (mean MPS = 5.0, p = 0.0007). In summary, the MPS showed good interrater repeatability between expert and lay scorers, significant correlation with lameness at the walk and trot, and good predictive validity for longitudinal health outcomes, confirming its ability to identify horses with orthopaedic health problems.

Automated Electrodermal Activity and Facial Expression Analysis for Continuous Pain Intensity Monitoring on the X-ITE Pain Database

This study focuses on improving healthcare quality by introducing an automated system that continuously monitors patient pain intensity. The system analyzes the Electrodermal Activity (EDA) sensor modality modality, compares the results obtained from both EDA and facial expressions modalities, and late fuses EDA and facial expressions modalities. This work extends our previous studies of pain intensity monitoring via an expanded analysis of the two informative methods. The EDA sensor modality and facial expression analysis play a prominent role in pain recognition; the extracted features reflect the patient's responses to different pain levels. Three different approaches were applied: Random Forest (RF) baseline methods, Long-Short Term Memory Network (LSTM), and LSTM with the sample-weighting method (LSTM-SW). Evaluation metrics included Micro average F1-score for classification and Mean Squared Error (MSE) and intraclass correlation coefficient (ICC [3, 1]) for both classification and regression. The results highlight the effectiveness of late fusion for EDA and facial expressions, particularly in almost balanced datasets (Micro average F1-score around 61%, ICC about 0.35). EDA regression models, particularly LSTM and LSTM-SW, showed superiority in imbalanced datasets and outperformed guessing (where the majority of votes indicate no pain) and baseline methods (RF indicates Random Forest classifier (RFc) and Random Forest regression (RFr)). In conclusion, by integrating both modalities or utilizing EDA, they can provide medical centers with reliable and valuable insights into patients' pain experiences and responses.

Artificial Intelligence for Automatic Pain Assessment: Research Methods and Perspectives

Although proper pain evaluation is mandatory for establishing the appropriate therapy, self-reported pain level assessment has several limitations. Data-driven artificial intelligence (AI) methods can be employed for research on automatic pain assessment (APA). The goal is the development of objective, standardized, and generalizable instruments useful for pain assessment in different clinical contexts. The purpose of this article is to discuss the state of the art of research and perspectives on APA applications in both research and clinical scenarios. Principles of AI functioning will be addressed. For narrative purposes, AI-based methods are grouped into behavioral-based approaches and neurophysiology-based pain detection methods. Since pain is generally accompanied by spontaneous facial behaviors, several approaches for APA are based on image classification and feature extraction. Language features through natural language strategies, body postures, and respiratory-derived elements are other investigated behavioral-based approaches. Neurophysiology-based pain detection is obtained through electroencephalography, electromyography, electrodermal activity, and other biosignals. Recent approaches involve multimode strategies by combining behaviors with neurophysiological findings. Concerning methods, early studies were conducted by machine learning algorithms such as support vector machine, decision tree, and random forest classifiers. More recently, artificial neural networks such as convolutional and recurrent neural network algorithms are implemented, even in combination. Collaboration programs involving clinicians and computer scientists must be aimed at structuring and processing robust datasets that can be used in various settings, from acute to different chronic pain conditions. Finally, it is crucial to apply the concepts of explainability and ethics when examining AI applications for pain research and management.

Potentials and Challenges of Pervasive Sensing in the Intensive Care Unit

Patients in critical care settings often require continuous and multifaceted monitoring. However, current clinical monitoring practices fail to capture important functional and behavioral indices such as mobility or agitation. Recent advances in non-invasive sensing technology, high throughput computing, and deep learning techniques are expected to transform the existing patient monitoring paradigm by enabling and streamlining granular and continuous monitoring of these crucial critical care measures. In this review, we highlight current approaches to pervasive sensing in critical care and identify limitations, future challenges, and opportunities in this emerging field.

Rasch development and validation of a new faces scale for measuring pain, and its comparison with a gold standard: the Balparda-Herrera Pain Scale

Aim: Faces pain scales are widely used to measure pain. So far, no faces pain scale has ever been constructed by Rasch modeling. Hence the authors aimed to construct a new scale by this method. Methods: Rasch modeling was used to provide an initial calibration and development of the 'Balparda-Herrera Pain Scale' (BHPS) and this scale was compared with the existing Faces Pain Scale - Revised. The scale was later refined. Results: Both the existing scale and the initial version of the BHPS required category collapsing. Statistical tests demonstrated an excellent concordance between both scales. The final version of the BHPS was found to behave excellently and to be capable of adequately measuring pain. Conclusion: The BHPS provides an excellent instrument for measuring pain in the adult population.

How to objectively assess and observe maladaptive pain behaviors in clinical rehabilitation: a systematic search and review

Background

Cognitive-affective factors influence the perception of pain and disability. These factors can lead to pain behaviors (PB) that can persist and become maladaptive. These maladaptive PB will further increase the risk of chronicity or persistence of symptoms and disability. Thus, clinicians must be prepared to recognize maladaptive PB in a clinical context. To date, in the context of assessment in a rehabilitation setting, PB in clinical settings are poorly documented. The main objective of this study was to identify direct observation methods and critically appraise them in order to propose recommendations for practice. As a secondary objective, we explored and extracted the different observable PB that patients could exhibit and that clinicians could observe.

Methods

We conducted a comprehensive review on four databases with a generic search strategy in order to obtain the largest range of PB. For the first objective, a two-step critical appraisal used clinical criteria (from qualitative studies on barriers to implement routine measures) and psychometric criteria (from Brink and Louw critical appraisal tool) to determine which observation methods could be recommended for clinical practice. For the second objective, we extracted PB found in the literature to list potential PB that patients could exhibit, and clinicians could observe.

Results

From the 3362 retrieved studies, 47 met the inclusion criteria for the first objective. The clinical criteria allowed us to select three observation methods. After the psychometric step, two observation methods were retained and recommended for clinical practice: the Behavioral Avoidance Test-Back Pain (BAT-Back) and the Pain Behaviour Scale (PaBS). For the second objective, 107 studies met the inclusion criteria. The extraction of the PB allowed us to list a large range of PB and classify the data in 7 categories of PB.

Conclusion

Our results allowed us to recommend two observation methods for clinical practice. However, these methods have limitations and are validated only in chronic low back pain populations. With the extraction of PB presented in the literature, we contribute to better prepare clinicians to recognize PB in all patients who are experiencing pain.

Automatic vs. Human Recognition of Pain Intensity from Facial Expression on the X-ITE Pain Database

Prior work on automated methods demonstrated that it is possible to recognize pain intensity from frontal faces in videos, while there is an assumption that humans are very adept at this task compared to machines. In this paper, we investigate whether such an assumption is correct by comparing the results achieved by two human observers with the results achieved by a Random Forest classifier (RFc) baseline model (called RFc-BL) and by three proposed automated models. The first proposed model is a Random Forest classifying descriptors of Action Unit (AU) time series; the second is a modified MobileNetV2 CNN classifying face images that combine three points in time; and the third is a custom deep network combining two CNN branches using the same input as for MobileNetV2 plus knowledge of the RFc. We conduct experiments with X-ITE phasic pain database, which comprises videotaped responses to heat and electrical pain stimuli, each of three intensities. Distinguishing these six stimulation types plus no stimulation was the main 7-class classification task for the human observers and automated approaches. Further, we conducted reduced 5-class and 3-class classification experiments, applied Multi-task learning, and a newly suggested sample weighting method. Experimental results show that the pain assessments of the human observers are significantly better than guessing and perform better than the automatic baseline approach (RFc-BL) by about 1%; however, the human performance is quite poor due to the challenge that pain that is ethically allowed to be induced in experimental studies often does not show up in facial reaction. We discovered that downweighting those samples during training improves the performance for all samples. The proposed RFc and two-CNNs models (using the proposed sample weighting) significantly outperformed the human observer by about 6% and 7%, respectively.

Facial features and head movements obtained with a webcam correlate with performance deterioration during prolonged wakefulness

We have performed a direct comparison between facial features obtained from a webcam and vigilance-task performance during prolonged wakefulness. Prolonged wakefulness deteriorates working performance due to changes in cognition, emotion, and by delayed response. Facial features can be potentially collected everywhere using webcams located in the workplace. If this type of device can obtain relevant information to predict performance deterioration, this technology can potentially reduce serious accidents and fatality. We extracted 34 facial indices, including head movements, facial expressions, and perceived facial emotions from 20 participants undergoing the psychomotor vigilance task (PVT) over 25 hours. We studied the correlation between facial indices and the performance indices derived from PVT, and evaluated the feasibility of facial indices as detectors of diminished reaction time during the PVT. Furthermore, we tested the feasibility of classifying performance as normal or impaired using several machine learning algorithms with correlated facial indices. Twenty-one indices were found significantly correlated with PVT indices. Pitch, from the head movement indices, and four perceived facial emotions-anger, surprise, sadness, and disgust-exhibited significant correlations with indices of performance. The eye-related facial expression indices showed especially strong correlation and higher feasibility of facial indices as classifiers. Significantly correlated indices were shown to explain more variance than the other indices for most of the classifiers. The facial indices obtained from a webcam strongly correlate with working performance during 25 hours of prolonged wakefulness.

[From external assessment of pain to automated multimodal measurement of pain intensity : Narrative review of state of research and clinical perspectives]

BACKGROUND

In patients with limited communication skills, the use of conventional scales or external assessment is only possible to a limited extent or not at all. Multimodal pain recognition based on artificial intelligence (AI) algorithms could be a solution.

OBJECTIVE

Overview of the methods of automated multimodal pain measurement and their recognition rates that were calculated with AI algorithms.

METHODS

In April 2018, 101 studies on automated pain recognition were found in the Web of Science database to illustrate the current state of research. A selective literature review with special consideration of recognition rates of automated multimodal pain measurement yielded 14 studies, which are the focus of this review.

RESULTS

The variance in recognition rates was 52.9-55.0% (pain threshold) and 66.8-85.7%; in nine studies the recognition rate was ≥80% (pain tolerance), while one study reported recognition rates of 79.3% (pain threshold) and 90.9% (pain tolerance).

CONCLUSION

Pain is generally recorded multimodally, based on external observation scales. With regard to automated pain recognition and on the basis of the 14 selected studies, there is to date no conclusive evidence that multimodal automated pain recognition is superior to unimodal pain recognition. In the clinical context, multimodal pain recognition could be advantageous, because this approach is more flexible. In the case of one modality not being available, e.g., electrodermal activity in hand burns, the algorithm could use other modalities (video) and thus compensate for missing information.