Using sensor-fusion and machine-learning algorithms to assess acute pain in non-verbal infants: a study protocol

New postoperative pain instrument for toddlers-Secondary analysis of prospectively collected assessments after tonsil surgery

BACKGROUND

The Children's and Infant's Postoperative Pain Scale (CHIPPS) and the German version of the Parent's Postoperative Pain Measure (PPPM-D) are used to assess postoperative pain intensity in preschool children. However, they have shown low concordance in previous prospective studies on quality improvement.

AIMS

Our secondary analysis aimed to estimate the association strength between the pain score items and indication for rescue medication defined as CHIPPS ≥4 and/or PPPD-D ≥ 6. Thus, we intended to create a further developed pain instrument with fewer variables for easier routine use.

METHODS

We analyzed 1067 pain intensity assessments of hospitalized children for the development of our novel tool in two steps using modern statistical and machine-learning methods: (1) Boruta variable selection to analyze the association strength between CHIPPS score, PPPM-D items, age, weight, and elapsed time after surgery, including their interactions and pattern stability, and the binary outcome (analgesics required yes/no). (2) Symbolic regression to generate a short formula with the least number of variables and highest accuracy for rescue medication indication.

RESULTS

Additional analgesics were required in 19.96% of pain intensity assessments, whereby the PPPM-D showed higher variance than CHIPPS. Boruta identified PPPM-D score, CHIPPS score, 9 of the 15 PPPM-D variables, and time of assessment as associated with the indication for RM. Symbolic regression revealed that additional analgesics are required if CHIPPS is ≥4 OR PPPM-D item "less energy than usual" AND one of the items "more easily cry" or "more groan/moan" are answered with "yes." These PPPM-D items were not redundant and showed nonlinear course over time. The cross-validated accuracy for this assessment tool was 94.94%.

CONCLUSIONS

The new instrument is easy to use and may improve postoperative pain intensity assessment in children. However, it requires prospective validation in a new cohort.

Multimodal automatic assessment of acute pain through facial videos and heart rate signals utilizing transformer-based architectures

Accurate and objective pain evaluation is crucial in developing effective pain management protocols, aiming to alleviate distress and prevent patients from experiencing decreased functionality. A multimodal automatic assessment framework for acute pain utilizing video and heart rate signals is introduced in this study. The proposed framework comprises four pivotal modules: the Spatial Module, responsible for extracting embeddings from videos; the Heart Rate Encoder, tasked with mapping heart rate signals into a higher dimensional space; the AugmNet, designed to create learning-based augmentations in the latent space; and the Temporal Module, which utilizes the extracted video and heart rate embeddings for the final assessment. The Spatial-Module undergoes pre-training on a two-stage strategy: first, with a face recognition objective learning universal facial features, and second, with an emotion recognition objective in a multitask learning approach, enabling the extraction of high-quality embeddings for the automatic pain assessment. Experiments with the facial videos and heart rate extracted from electrocardiograms of the BioVid database, along with a direct comparison to 29 studies, demonstrate state-of-the-art performances in unimodal and multimodal settings, maintaining high efficiency. Within the multimodal context, 82.74% and 39.77% accuracy were achieved for the binary and multi-level pain classification task, respectively, utilizing 9.62 million parameters for the entire framework.

Pain assessment tools in adults with communication disorders: systematic review and meta-analysis

BACKGROUND

Verbal communication is the "gold standard" for assessing pain. Consequently, individuals with communication disorders are particularly vulnerable to incomplete pain management. This review aims at identifying the current pain assessment instruments for adult patients with communication disorders.

METHODS

A systematic review with meta-analysis was conducted on PubMed, PEDRO, EBSCOhost, VHL and Cochrane databases from 2011 to 2023 using MeSH terms "pain assessment, "nonverbal communication" and "communication disorders" in conjunction with additional inclusion criteria: studies limited to humans, interventions involving adult patients, and empirical investigations.

RESULTS

Fifty articles were included in the review. Seven studies report sufficient data to perform the meta-analysis. Observational scales are the most common instruments to evaluate pain in individuals with communication disorders followed by physiological measures and facial recognition systems. While most pain assessments rely on observational scales, current evidence does not strongly endorse one scale over others for clinical practice. However, specific observational scales appear to be particularly suitable for identifying pain during certain potentially painful procedures, such as suctioning and mobilization, in these populations. Additionally, specific observational scales appear to be well-suited for certain conditions, such as mechanically ventilated patients.

CONCLUSIONS

While observational scales dominate pain assessment, no universal tool exists for adults with communication disorders. Specific scales exhibit promise for distinct populations, yet the diverse landscape of tools hampers a one-size-fits-all solution. Crucially, further high-quality research, offering quantitative data like reliability findings, is needed to identify optimal tools for various contexts. Clinicians should be informed to select tools judiciously, recognizing the nuanced appropriateness of each in diverse clinical situations.

TRIAL REGISTRATION

This systematic review is registered in PROSPERO (International prospective register of systematic reviews) with the ID: CRD42022323655 .

Automated facial recognition system using deep learning for pain assessment in adults with cerebral palsy

Objective

Assessing pain in individuals with neurological conditions like cerebral palsy is challenging due to limited self-reporting and expression abilities. Current methods lack sensitivity and specificity, underlining the need for a reliable evaluation protocol. An automated facial recognition system could revolutionize pain assessment for such patients.The research focuses on two primary goals: developing a dataset of facial pain expressions for individuals with cerebral palsy and creating a deep learning-based automated system for pain assessment tailored to this group.

Methods

The study trained ten neural networks using three pain image databases and a newly curated CP-PAIN Dataset of 109 images from cerebral palsy patients, classified by experts using the Facial Action Coding System.

Results

The InceptionV3 model demonstrated promising results, achieving 62.67% accuracy and a 61.12% F1 score on the CP-PAIN dataset. Explainable AI techniques confirmed the consistency of crucial features for pain identification across models.

Conclusion

The study underscores the potential of deep learning in developing reliable pain detection systems using facial recognition for individuals with communication impairments due to neurological conditions. A more extensive and diverse dataset could further enhance the models' sensitivity to subtle pain expressions in cerebral palsy patients and possibly extend to other complex neurological disorders. This research marks a significant step toward more empathetic and accurate pain management for vulnerable populations.

The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review

Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units.

Face-based automatic pain assessment: challenges and perspectives in neonatal intensive care units

OBJECTIVE

To describe the challenges and perspectives of the automation of pain assessment in the Neonatal Intensive Care Unit.

DATA SOURCES

A search for scientific articles published in the last 10 years on automated neonatal pain assessment was conducted in the main Databases of the Health Area and Engineering Journal Portals, using the descriptors: Pain Measurement, Newborn, Artificial Intelligence, Computer Systems, Software, Automated Facial Recognition.

SUMMARY OF FINDINGS

Fifteen articles were selected and allowed a broad reflection on first, the literature search did not return the various automatic methods that exist to date, and those that exist are not effective enough to replace the human eye; second, computational methods are not yet able to automatically detect pain on partially covered faces and need to be tested during the natural movement of the neonate and with different light intensities; third, for research to advance in this area, databases are needed with more neonatal facial images available for the study of computational methods.

CONCLUSION

There is still a gap between computational methods developed for automated neonatal pain assessment and a practical application that can be used at the bedside in real-time, that is sensitive, specific, and with good accuracy. The studies reviewed described limitations that could be minimized with the development of a tool that identifies pain by analyzing only free facial regions, and the creation and feasibility of a synthetic database of neonatal facial images that is freely available to researchers.

Multimodal physiological sensing for the assessment of acute pain

Pain assessment is a challenging task encountered by clinicians. In clinical settings, patients' self-report is considered the gold standard in pain assessment. However, patients who are unable to self-report pain are at a higher risk of undiagnosed pain. In the present study, we explore the use of multiple sensing technologies to monitor physiological changes that can be used as a proxy for objective measurement of acute pain. Electrodermal activity (EDA), photoplethysmography (PPG), and respiration (RESP) signals were collected from 22 participants under two pain intensities (low and high) and on two different anatomical locations (forearm and hand). Three machine learning models were implemented, including support vector machines (SVM), decision trees (DT), and linear discriminant analysis (LDA) for the identification of pain. Various pain scenarios were investigated, identification of pain (no pain, pain), multiclass (no pain, low pain, high pain), and identification of pain location (forearm, hand). Reference classification results from individual sensors and from all sensors together were obtained. After feature selection, results showed that EDA was the most informative sensor in the three pain conditions, 93.2±8% in identification of pain, 68.9±10% in the multiclass problem, and 56.0±8% for the identification of pain location. These results identify EDA as the superior sensor in our experimental conditions. Future work is required to validate the obtained features to improve its feasibility in more realistic scenarios. Finally, this study proposes EDA as a candidate to design a tool that can assist clinicians in the assessment of acute pain of nonverbal patients.

Triage through telemedicine in paediatric emergency care—Results of a concordance study

Background

In the German health care system, parents with an acutely ill child can visit an emergency room (ER) 24 hours a day, seven days a week. At the ER, the patient receives a medical consultation. Many parents use these facilities as they do not know how urgently their child requires medical attention. In recent years, paediatric departments in smaller hospitals have been closed, particularly in rural regions. As a result of this, the distances that patients must travel to paediatric care facilities in these regions are increasing, causing more children to visit an ER for adults. However, paediatric expertise is often required in order to assess how quickly the patient requires treatment and select an adequate treatment. This decision is made by a doctor in German ERs. We have examined whether remote paediatricians can perform a standardised urgency assessment (triage) using a video conferencing system.

Methods

Only acutely ill patients who were brought to a paediatric emergency room (paedER) by their parents or carers, without prior medical consultation, have been included in this study. First, an on-site paediatrician assessed the urgency of each case using a standardised triage. In order to do this, the Paediatric Canadian Triage and Acuity Scale (PaedCTAS) was translated into German and adapted for use in a standardised IT-based data collection tool. After the initial on-site triage, a telemedicine paediatrician, based in a different hospital, repeated the triage using a video conferencing system. Both paediatricians used the same triage procedure. The primary outcome was the degree of concordance and interobserver agreement, measured using Cohen’s kappa, between the two paediatricians. We have also included patient and assessor demographics.

Results

A total of 266 patients were included in the study. Of these, 227 cases were eligible for the concordance analysis. In n = 154 cases (68%), there was concordance between the on-site paediatrician’s and telemedicine paediatrician’s urgency assessments. In n = 50 cases (22%), the telemedicine paediatrician rated the urgency of the patient’s condition higher (overtriage); in 23 cases (10%), the assessment indicated a lower urgency (undertriage). Nineteen medical doctors were included in the study, mostly trained paediatric specialists. Some of them acted as an on-site doctor and telemedicine doctor. Cohen’s weighted kappa was 0.64 (95% CI: 0.49–0.79), indicating a substantial agreement between the specialists.

Conclusions

Telemedical triage can assist in providing acute paediatric care in regions with a low density of paediatric care facilities. The next steps are further developing the triage tool and implementing telemedicine urgency assessment in a larger network of hospitals in order to improve the integration of telemedicine into hospitals’ organisational processes. The processes should include intensive training for the doctors involved in telemedical triage.

Trial registration

DRKS00013207.

Assessment and Management of Pain in Preterm Infants: A Practice Update

Infants born preterm are at a high risk for repeated pain exposure in early life. Despite valid tools to assess pain in non-verbal infants and effective interventions to reduce pain associated with medical procedures required as part of their care, many infants receive little to no pain-relieving interventions. Moreover, parents remain significantly underutilized in provision of pain-relieving interventions, despite the known benefit of their involvement. This narrative review provides an overview of the consequences of early exposure to untreated pain in preterm infants, recommendations for a standardized approach to pain assessment in preterm infants, effectiveness of non-pharmacologic and pharmacologic pain-relieving interventions, and suggestions for greater active engagement of parents in the pain care for their preterm infant.