To Calibrate or not to Calibrate? A Methodological Dilemma in Experimental Pain Research

Adamczyk, Wacław M.; Szikszay, Tibor M.; Nahman‐Averbuch, Hadas; Skalski, Jacek; Nastaj, Jakub; Gouverneur, Philip; Luedtke, Kerstin

doi:10.1016/j.jpain.2022.07.007

Cited by 13 publications

(6 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A multitude of psychological, physiological, anatomical and genetic factors (which are beyond the scope of this discussion) could contribute to the overall pain experience, particularly when all participants received the same dose (1 mL) of hypertonic saline. High levels of interindividual variability from the same nociceptive input can complicate the interpretation of outcomes of interest (Adamczyk et al, 2022), particularly when a number of ‘low‐responders’ are present. For example, in our cohort, the lowest peak pain response observed was 22 and the greatest was 77 (Figure 2a).…”

Section: Discussionmentioning

confidence: 99%

Intra‐ and interindividual reliability of muscle pain induced by an intramuscular injection of hypertonic saline injection into the quadriceps

Smith

Norbury

Hunt

et al. 2023

European Journal of Pain

View full text Add to dashboard Cite

BackgroundIntramuscular injections of hypertonic saline are commonly used to induce experimental muscle pain, but reliability data on this technique are lacking. This study investigated the intra‐ and interindividual reliability of pain measures from a hypertonic saline injection into the vastus lateralis.MethodsFourteen healthy participants (6 female) attended three laboratory visits where they received an intramuscular injection of 1 mL hypertonic saline into the vastus lateralis. Changes in pain intensity were recorded on an electronic visual analogue scale, and pain quality was assessed after pain had resolved. Reliability was assessed with the coefficient of variation (CV), minimum detectable change (MDC) and intraclass correlation coefficient (ICC) with 95% CIs.ResultsMean pain intensity displayed high levels of intraindividual variability (CV = 16.3 [10.5–22.0]%) and ‘poor’ to ‘very good’ relative reliability (ICC = 0.71 [0.45–0.88]) but had a MDC of 11 [8–16] au (out of 100). Peak pain intensity exhibited high levels of intraindividual variability (CV = 14.8 [8.8–20.8]%) with ‘moderate’ to ‘excellent’ levels of relative reliability (ICC = 0.81 [0.62–0.92]), whereas the MDC was 18 [14–26] au. Measures of pain quality exhibited good reliability. Interindividual variability in pain measures was high (CV > 37%).ConclusionsIntramuscular injections of 1 mL of hypertonic saline into the vastus lateralis display substantial levels of interindividual variability, but MDC is below the clinically important changes in pain. This model of experimental pain is suitable for studies involving repeated exposures.SignificanceMany pain research studies have performed intramuscular injections of hypertonic saline to investigate responses to muscle pain. However, the reliability of this technique is not well established. We examined the pain response over three repeated sessions of a hypertonic saline injection. The pain induced by hypertonic saline has considerable interindividual variability but has largely acceptable intraindividual reliability. Therefore, the injections of hypertonic saline to induce muscle pain are a reliable model of experimental muscle pain.

show abstract

Section: Discussionmentioning

confidence: 99%

Intra‐ and interindividual reliability of muscle pain induced by an intramuscular injection of hypertonic saline injection into the quadriceps

Smith

Norbury

Hunt

et al. 2023

European Journal of Pain

View full text Add to dashboard Cite

show abstract

“…Finally, there is a debate on whether it is appropriate to calibrate the applied stimulus intensity to the perception of the individual (e.g., to a subjective 5 of 10). A recent publication discusses the advantages and disadvantages of this approach ( 48 ). For the present publication, a calibration would have benefitted the analysis of within-subject variability as all participants would have centered their ratings around the same value (e.g., a subjective 5 of 10).…”

Section: Discussionmentioning

confidence: 99%

Interindividual variability and individual stability of pain- and touch-related neuronal gamma oscillations

Valentini

Shindy

Witkovský

et al. 2023

Journal of Neurophysiology

View full text Add to dashboard Cite

Background. Brief painful laser and innocuous tactile stimuli have been associated with an increase of neuronal oscillations in the gamma range. However, it is indicated that event-related gamma oscillations may be highly variable across individuals, but no study has systematically studied interindividual variability and individual stability of induced gamma synchronisation. Here, we addressed this question using two datasets. Methods. The present study includes two EEG datasets. The first dataset contains two repeated sessions of tactile and painful stimulation from 22 participants. The second dataset contains a single session of painful stimulation from 48 participants. Results. In the first dataset, we observed gamma responses in the majority of the included participants. We found a broad interindividual variety of gamma magnitudes, time-frequency response patterns, and scalp topographies. Some participants showed a gamma response with individually unique time-frequency patterns, others did not exhibit any gamma response. This was reproducible and therefore stable; subjects with a large gamma magnitude in the first session showed a large gamma magnitude and a similar response pattern in the follow-up session. The second dataset confirmed the large between-subject variability, but only a fraction of the included participants exhibited laser-induced gamma synchronisation. Conclusions. Our results indicate that current EEG measures do not reflect the complex reality of the diverse individual response patterns to brief pain and touch experiences. The present findings question whether a similar phenomenon would be observed in other neuroscience domains. Group results may be replicable, but could be driven by a subgroup of the sample.

show abstract

“…In contrast to the BVDB, several adjustments were made to the data acquisition protocol. Calibration was performed to reduce variability in pain ratings [ 33 ] by using the “method of staircase” [ 34 ] (p. 400) and performed twice to ensure further robustness by averaging the results of the two trials. In addition, the duration of the stimuli windows was raised to 10 s to ensure enough recording time of the EDA signal, as previous studies showed that peak values normally occur “between 3 and 6 s poststimulus” [ 35 ], and thus longer recording times than the ones chosen in BVDB could be beneficial.…”

Section: Methodsmentioning

confidence: 99%

Explainable Artificial Intelligence (XAI) in Pain Research: Understanding the Role of Electrodermal Activity for Automated Pain Recognition

Gouverneur

Shirahama

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

Artificial intelligence and especially deep learning methods have achieved outstanding results for various applications in the past few years. Pain recognition is one of them, as various models have been proposed to replace the previous gold standard with an automated and objective assessment. While the accuracy of such models could be increased incrementally, the understandability and transparency of these systems have not been the main focus of the research community thus far. Thus, in this work, several outcomes and insights of explainable artificial intelligence applied to the electrodermal activity sensor data of the PainMonit and BioVid Heat Pain Database are presented. For this purpose, the importance of hand-crafted features is evaluated using recursive feature elimination based on impurity scores in Random Forest (RF) models. Additionally, Gradient-weighted class activation mapping is applied to highlight the most impactful features learned by deep learning models. Our studies highlight the following insights: (1) Very simple hand-crafted features can yield comparative performances to deep learning models for pain recognition, especially when properly selected with recursive feature elimination. Thus, the use of complex neural networks should be questioned in pain recognition, especially considering their computational costs; and (2) both traditional feature engineering and deep feature learning approaches rely on simple characteristics of the input time-series data to make their decision in the context of automated pain recognition.

show abstract

To Calibrate or not to Calibrate? A Methodological Dilemma in Experimental Pain Research

Cited by 13 publications

References 80 publications

Intra‐ and interindividual reliability of muscle pain induced by an intramuscular injection of hypertonic saline injection into the quadriceps

Intra‐ and interindividual reliability of muscle pain induced by an intramuscular injection of hypertonic saline injection into the quadriceps

Interindividual variability and individual stability of pain- and touch-related neuronal gamma oscillations

Explainable Artificial Intelligence (XAI) in Pain Research: Understanding the Role of Electrodermal Activity for Automated Pain Recognition

Contact Info

Product

Resources

About