More Like this

1. Equine Pain Behavior Classification via Self-Supervised Disentangled Pose Representation

   https://doi.org/10.1109/wacv51458.2022.00023  

   Rashid, Maheen; Broome, Sofia; Ask, Katrina; Hernlund, Elin; Andersen, Pia Haubro; Kjellstrom, Hedvig; Lee, Yong Jae (January 2022, Winter Conference on Applications of Computer Vision (WACV))

   Timely detection of horse pain is important for equine welfare. Horses express pain through their facial and body behavior, but may hide signs of pain from unfamiliar human observers. In addition, collecting visual data with detailed annotation of horse behavior and pain state is both cumbersome and not scalable. Consequently, a pragmatic equine pain classification system would use video of the unobserved horse and weak labels. This paper proposes such a method for equine pain classification by using multi-view surveillance video footage of unobserved horses with induced orthopaedic pain, with temporally sparse video level pain labels. To ensure that pain is learned from horse body language alone, we first train a self-supervised generative model to disentangle horse pose from its appearance and background before using the disentangled horse pose latent representation for pain classification. To make best use of the pain labels, we develop a novel loss that formulates pain classification as a multi-instance learning problem. Our method achieves pain classification accuracy better than human expert performance with 60% accuracy. The learned latent horse pose representation is shown to be viewpoint covariant, and disentangled from horse appearance. Qualitative analysis of pain classified segments shows correspondence between the pain symptoms identified by our model, and equine pain scales used in veterinary practice. 

   more » « less
2. COMPASS App: A Patient-centered Physiological based Pain Assessment System

   https://doi.org/10.1177/21695067231192200  

   Zhu, Wenchao; Liu, Chang; Yu, Haoxiang; Guo, Yikang; Xiao, Yan; Lin, Yingzi (September 2023, Proceedings of the Human Factors and Ergonomics Society Annual Meeting)

   Chronic pain patients lack at-home pain assessment and management tools. The existing chronic-pain mobile applications are either solely relying on self-report pain levels or restricted to formal clinical settings. Our app, abbreviated from an NSF-funded project entitled Novel Computational Methods for Continuous Objective Multimodal Pain Assessment Sensing System (COMPASS), is a multi-dimensional pain app that collects physiological signals to predict objective pain levels and trace daily at-home activities by incorporating a daily check-in section. We conducted a usability test with 33 healthy participants under pain conditions. The results provided initial support for the validity of the signals in predicting internalizing pain levels among the participants. With further development and testing, we believe the COMPASS app system has the potential to be used by both patients and clinicians as an additional tool to better assess and manage pain, especially for mobile healthcare applications. 

   more » « less
3. The association between the problem list and chronic pain management

   https://doi.org/10.1007/s44250-022-00016-4  

   Rylee, Tina L; Copenhaver, David; Drake, Christiana; Joseph, Jill (December 2023, Discover Health Systems)

   Abstract BackgroundRoughly a quarter of the US population suffers from moderate to severe chronic pain for at least six months in any given year. The complexity of managing chronic pain has encouraged providers to use innovative methods to address it. Research has shown that problem lists are potential tools that support the care of patients with diabetes and chronic kidney disease. ObjectivesTo examine the extent to which the inclusion of chronic pain in a problem list is associated with follow-up specialty pain care. MethodsThe association between chronic pain documentation on the problem list and specialty pain care was investigated in this retrospective cohort study using 4531 patient records. ResultsChronic pain documentation in the problem list was associated with higher odds of receiving specialty pain care. The most common diagnosis was chronic pain (69.7%). A migraine diagnosis was associated with decreased odds of receiving specialty care, and chronic pain syndrome was associated with increased odds of receiving specialty care compared with the other chronic pain groups. ConclusionDocumenting chronic pain on the problem list was associated with a higher likelihood of patients receiving specialty pain care. 

   more » « less
4. Lifetime ostracism experiences and mechanisms of pain

   https://doi.org/10.3389/fpain.2022.1037472  

   Walsh, Kaitlyn T.; Boring, Brandon L.; Nanavaty, Namrata; Carter-Sowell, Adrienne R.; Mathur, Vani A. (December 2022, Frontiers in Pain Research)

   One social mechanism by which marginalization is enacted is via ostracism. Recent research has demonstrated ostracism's impact on physical health, but little is known about the relationship between accumulated lifetime experiences of ostracism and pain. Despite recent calls for added attention to social modulation of pain and social indicators of pain disparities, the impact of specific social factors on pain—including those of ostracism—are not well understood. Results of laboratory studies on the effects of acute ostracism experiences on pain sensitivity have been mixed. However, these studies have not considered lived and repeated experiences of ostracism, and primarily included single static measures of pain sensitivity. Additionally, inclusion and representation of the relationship between ostracism experiences and pain among people with minoritized identities are lacking in the current literature. In this study, we explored accumulated lifetime experiences of ostracism as a potential contributing factor to enhanced pain and one social mechanism by which societal inequity may create and maintain inequity in pain. We extracted measures of lifetime experiences of ostracism from six studies focused on social factors and (non-chronic) pain conducted between 2016 and 2020 ( n  = 505 adults). To retain and examine diversity within the sample, we used moderation and within-group analyses. Results indicate that greater experiences of lifetime ostracism are associated with lower cold pain tolerance, but not other pain measures, in the whole sample. Moderation and within-group analyses reveal opposing patterns of results between populations included in the extant literature (White participants, convenience samples) and those under-represented in the scientific literature (racialized groups, community samples). This study provides an example of a diversity science approach to examining social indicators of pain, illustrates the limited generalizability of previous studies on ostracism and pain, and highlights the need for increased representation and inclusion to understand mechanisms of pain and inequity. 

   more » « less
5. Performance Evaluation of a Supervised Machine Learning Pain Classification Model Developed by Neonatal Nurses

   https://doi.org/10.1097/ANC.0000000000001145  

   Manworren, Renee CB; Horner, Susan; Joseph, Ralph; Dadar, Priyansh; Kaduwela, Naomi (May 2024, Advances in Neonatal Care)

   Background: Early-life pain is associated with adverse neurodevelopmental consequences; and current pain assessment practices are discontinuous, inconsistent, and highly dependent on nurses’ availability. Furthermore, facial expressions in commonly used pain assessment tools are not associated with brain-based evidence of pain. Purpose: To develop and validate a machine learning (ML) model to classify pain. Methods: In this retrospective validation study, using a human-centered design for Embedded Machine Learning Solutions approach and the Neonatal Facial Coding System (NFCS), 6 experienced neonatal intensive care unit (NICU) nurses labeled data from randomly assigned iCOPEvid (infant Classification Of Pain Expression video) sequences of 49 neonates undergoing heel lance. NFCS is the only observational pain assessment tool associated with brain-based evidence of pain. A standard 70% training and 30% testing split of the data was used to train and test several ML models. NICU nurses’ interrater reliability was evaluated, and NICU nurses’ area under the receiver operating characteristic curve (AUC) was compared with the ML models’ AUC. Results: Nurses weighted mean interrater reliability was 68% (63%-79%) for NFCS tasks, 77.7% (74%-83%) for pain intensity, and 48.6% (15%-59%) for frame and 78.4% (64%-100%) for video pain classification, with AUC of 0.68. The best performing ML model had 97.7% precision, 98% accuracy, 98.5% recall, and AUC of 0.98. Implications for Practice and Research: The pain classification ML model AUC far exceeded that of NICU nurses for identifying neonatal pain. These findings will inform the development of a continuous, unbiased, brain-based, nurse-in-the-loop Pain Recognition Automated Monitoring System (PRAMS) for neonates and infants. 

   more » « less