Performance Evaluation of a Supervised Machine Learning Pain Classification Model Developed by Neonatal Nurses.

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.

Application of a human-centered design for embedded machine learning model to develop data labeling software with nurses: Human-to-Artificial Intelligence (H2AI).

BACKGROUND: Nurses are essential for assessing and managing acute pain in hospitalized patients, especially those who are unable to self-report pain. Given their role and subject matter expertise (SME), nurses are also essential for the design and development of a supervised machine learning (ML) model for pain detection and clinical decision support software (CDSS) in a pain recognition automated monitoring system (PRAMS). Our first step for developing PRAMS with nurses was to create SME-friendly data labeling software. PURPOSE: To develop an intuitive and efficient data labeling software solution, Human-to-Artificial Intelligence (H2AI). METHOD: The Human-centered Design for Embedded Machine Learning Solutions (HCDe-MLS) model was used to engage nurses. In this paper, HCDe-MLS will be explained using H2AI and PRAMS as illustrative cases. FINDINGS: Using HCDe-MLS, H2AI was developed and facilitated labeling of 139 videos (mean = 29.83 min) with 3189 images labeled (mean = 75 s) by 6 nurses. OpenCV was used for video-to-image pre-processing; and MobileFaceNet was used for default landmark placement on images. H2AI randomly assigned videos to nurses for data labeling, tracked labelers' inter-rater reliability, and stored labeled data to train ML models. CONCLUSIONS: Nurses' engagement in CDSS development was critical for ensuring the end-product addressed nurses' priorities, reflected nurses' cognitive and decision-making processes, and garnered nurses' trust for technology adoption.

Pain Quality After Pediatric Pectus Excavatum Repair.

BACKGROUND: Neuropathic pain medications are included in multimodal postoperative analgesic strategies, but quality of perioperative pain is rarely assessed. AIMS: The purpose of this study was to describe adolescents' pain experiences after thoracoscopic pectus excavatum repair (Nuss procedure) using the Adolescent Pediatric Pain Tool. DESIGN: This prospective descriptive longitudinal study was designed to test the hypothesis that pain quality descriptors reported are consistent with neuropathic pain. METHODS: A convenience sample of 23 adolescents aged 12 to 17 years from a single urban, university affiliated, nonprofit children's hospital consented to self-report pain using the Adolescent Pediatric Pain Tool before and during hospitalization, and up to 14 months after Nuss procedure. Visual analytic techniques were used to analyze reported pain intensity, location, and affective, evaluative, sensory, and temporal qualities. RESULTS: Postoperative pain quality, intensity, number of sites, and surface area decreased over time. Word clouds illustrated that neuropathic sensory and temporal pain quality descriptors increased in frequency 2 to 6 weeks after surgery and were the predominant descriptors 2 to 4 months after surgery. Dot matrix charts illustrated an inconsistent relationship of pain quality and intensity with pain surface area. CONCLUSIONS: Pain quality should be assessed with valid, reliable, and developmentally appropriate tools. Visual analytics help illustrate pain quality at single points in time and longitudinally and may be helpful in guiding postoperative pain treatment.

Children's Opioid Use at Home After Laparoscopic Appendectomy.

BACKGROUND: To combat the opioid epidemic, prescribers need accurate information about pediatric home opioid requirements to manage acute pain after surgery. Current opioid use estimates come from retrospective surveys; this study used medication adherence technology (eCAP) to track home opioid use. PURPOSE: To describe children's pain treatment at home after laparoscopic appendectomy, and to compare self-reported opioid analgesic use to eCAP data and counts of returned pills. DESIGN: Prospective exploratory and descriptive study METHODS: A convenience sample of 96 patients, 10-17 years of age, from a single urban nonprofit children's hospital consented to self-report pain treatment in 14-day diaries and use eCAP to monitor prescribed opioid use at home after laparoscopic appendectomy. RESULTS: Patients were prescribed 5-45 opioid-containing pills (mean ± standard deviation 15 ± 7.2). Of 749 opioid-containing pills prescribed to 49 patients who returned data, 689 pills were dispensed, 167.5 were used for the reason prescribed, 488 were returned to families for disposal, and 53.5 were missing. The majority of the 49 patients were opioid naïve (72%), Caucasian (64%), and male (56%), with a mean age of 14 years. Patients used 6.6 ± 6.3 opioid-containing pills by pill count and 5.6 ± 5.1 by self-report, a significant difference (p = .004). Unreported eCAP-enabled pill bottle openings typically occurred on weekends. CONCLUSION: Medication adherence technology (eCAP) is a more rigorous method than self-report to estimate opioid needs and detect early opioid misuse. Additional rigorously designed studies of postoperative opioid use are needed to guide opioid prescribing.

Children's Pain at Home After Laparoscopic Appendectomy.

BACKGROUND: Hospitalized children experience moderate-to-severe pain after laparoscopic appendectomy, but knowledge of children's pain experiences after discharge home is limited. Accurate pain assessments are needed to guide appropriate pain treatment. AIMS: To describe children's pain at home after laparoscopic appendectomy. DESIGN: Prospective exploratory and descriptive METHODS: A convenience sample of 100 patients, aged 10-17 years, who spoke or wrote in English or Spanish, volunteered to complete 14-day pain diaries at home after laparoscopic appendectomy. Visual analytic techniques were used to analyze patterns of pain experiences. RESULTS: Diaries were returned by 45 patients/parents, the majority of whom were White (64%), male (56%), adolescents (mean age 14 years) with no previous surgical history (70%), and whose appendix was inflamed (87%) but not perforated. More than 50% reported severe pain (4 or 5 on a 0-5 scale) on the first full day home after laparoscopic appendectomy. On day 7, 40% reported pain and on day 14, 16% were still reporting pain. Only rarely were pain scores not clinically significantly lower 1 hour after pain treatment, regardless of treatment type (e.g., nondrug, nonopioid, opioid). Reported pain intensity steadily decreased over time as did frequency of recorded pain scores. CONCLUSION: Adolescents experience severe pain at home after laparoscopic appendectomy and some experience pain for 7 to 14 days after hospital discharge. Visual analytics better represent the dynamics of pain experiences than measures of central tendency.