Schmerz und Anästhesiologie

Using AI to Detect Pain through Facial Expressions: A Review

Pain assessment is a complex task largely dependent on the patient's self-report. Artificial intelligence (AI) has emerged as a promising tool for automating and objectifying pain assessment through the identification of pain-related facial expressions. However, the capabilities and potential of AI in clinical settings are still largely unknown to many medical professionals. In this literature review, we present a conceptual understanding of the application of AI to detect pain through facial expressions. We provide an overview of the current state of the art as well as the technical foundations of AI/ML techniques used in pain detection. We highlight the ethical challenges and the limitations associated with the use of AI in pain detection, such as the scarcity of databases, confounding factors, and medical conditions that affect the shape and mobility of the face. The review also highlights the potential impact of AI on pain assessment in clinical practice and lays the groundwork for further study in this area.

The Effectiveness of Music Therapy for Terminally Ill Patients: A Meta-Analysis and Systematic Review

CONTEXT

The quality of death has increasingly raised concern because of the physical and psychological suffering of patients with advanced disease. Music therapy has been widely used in palliative care; however, its physical and mental effectiveness remains unclear.

OBJECTIVE

To assess the effectiveness of music therapy during palliative care in improving physiology and psychology outcomes.

METHODS

Randomized controlled trials evaluating music therapy for terminally ill patients were searched and included from inception up to April 25, 2018. The quality of the studies was assessed using the risk of bias tool recommended by the Cochrane Handbook V.5.1.0.

RESULTS

In this study, 11 randomized controlled trials (inter-rater agreement, κ = 0.86) involving 969 participants were included. The quality of the included studies ranged from moderate to high. Compared with general palliative care, music therapy can reduce pain (standardized mean difference: -0.44, 95% confidence interval: -0.60 to -0.27, P < 0.00001) and improve the quality of life (standardized mean difference: 0.61, 95% confidence interval: 0.41 to 0.82, P < 0.00001) in terminally ill patients. In addition, anxiety, depression, and emotional function are improved as well. However, no significant differences were found in the patient's physical status, fatigue, and social function.

CONCLUSION

This meta-analysis study demonstrated that music therapy served as an effective intervention to alleviate pain and psychological symptoms of terminally ill patients. However, considering the limitation of the quantity of the studies included, these results would need to be further confirmed.

Machine learning methods for automatic pain assessment using facial expression information: Protocol for a systematic review and meta-analysis

INTRODUCTION

Prediction of pain using machine learning algorithms is an emerging field in both computer science and clinical medicine. Several machine algorithms were developed and validated in recent years. However, the majority of studies in this topic was published on bioinformatics or computer science journals instead of medical journals. This tendency and preference led to a gap of knowledge and acknowledgment between computer scientists who invent the algorithm and medical researchers who may use the algorithms in practice. As a consequence, some of these prediction papers did not discuss the clinical utility aspects and were causally reported without following related professional guidelines (e.g., TRIPOD statement). The aim of this protocol is to systematically summarize the current evidences about performance and utility of different machine learning methods used for automatic pain assessments based on human facial expression. In addition, this study is aimed to demonstrate and fill the knowledge gap to promote interdisciplinary collaboration.

METHODS AND ANALYSIS

We will search all English language literature in the following electronic databases: PubMed, Web of Science and IEEE Xplore. A systematic review and meta-analysis summarizing the accuracy, interpretability, generalizability, and computational efficiency of machine learning methods will be conducted. Subgroup analyses by machine learning method types will be conducted.

TIMELINE

The formal meta-analysis will start on Jan 15, 2019 and expected to finish by April 15, 2019.

ETHICS AND DISSEMINATION

Ethical approval will be exempted or will not be required because the data collected and analyzed in this meta-analysis will not be on an individual level. The results will be disseminated in the form of an official publication in a peer-reviewed journal and/or presentation at relevant conferences.

REGISTRATION

PROSPERO CRD42018103059.

Different measures, different outcomes? Survey into the effectiveness of chronic pain clinics in a London tertiary referral center

Background

Chronic pain clinics aim to improve challenging conditions, and although numerous studies have evaluated specific aspects of therapies and outcomes in this context, data concerning service impact on outcome measures in a general pain population are sparse. In addition, current trends in commissioning increasingly warrant services to provide evidence for their effectiveness. While a plethora of outcome measures, such as pain-intensity or improvement scores, exist for this purpose, it remains surprisingly unclear which one to use. It also remains uncertain what variables predict treatment success.

Objectives

This cross-sectional study was conducted to evaluate clinic performance employing different tools (pain scores, pain categories, responder analysis, subjective improvement, satisfaction), and to determine predictors of outcome measures.

Patients and methods

Patients attending scheduled clinic follow-up appointments were approached. They were asked to complete the modified short-form Brief Pain Inventory (BPI-SF) that also included assessments for satisfaction and subjective improvement. Comparisons were made with BPI-SF responses that were completed by each patient on admission. Nonparametric tests were employed to evaluate service impact and to determine predictors for outcome.

Results

Data of 118 patients were analyzed. There was considerable variation in impact of pain clinics depending on the outcome measure employed. While median pain scores did not differ between admission and follow-up, scores improved individually in 30% of cases, such that more patients had mild pain on follow-up than on admission (relative risk 2.7). Furthermore, while only 41% reported at least moderate subjective improvement after admission to the service, the majority (83%) were satisfied with the service. Positive treatment responses were predicted by “number of painful regions” and “changes in mood”, whereas subjective improvement was predicted by “helpfulness of treatments”.

Conclusion

Depending on the outcome measure employed, pain clinics showed varying degrees of impact on patients’ pain experiences. This calls into question the current practice of using nonstandardized outcome reporting for evaluation of service performances.