# Physiological framework for non-invasive detection and objective nociception activity in communicative patients: a pilot case study

## TL;DR

This study introduces a non-invasive method to objectively track pain using physiological signals, showing strong correlation with subjective pain scores.

## Contribution

A novel physiologically grounded index, ΔT, derived from bioimpedance data, is proposed as an objective and anticipatory biomarker of nociception.

## Key findings

- ΔT closely follows and often precedes subjective NRS scores with a correlation coefficient of up to 0.86.
- Data augmentation strategies enabled synthetic trajectory generation, revealing consistent lead–lag and causal relationships between ΔT and NRS.

## Abstract

Pain assessment in both communicative and non-communicative patients remains a major clinical challenge due to the inherently subjective nature of conventional tools such as the Numeric Rating Scale (NRS). In this study, we introduce a physiologically grounded and objective index,
ΔT=TS−TD
, derived from fractional-order impedance modeling of nociceptive dynamics. Here,
TD
represents the transduction, and
TS
reflects the transmission. These components are extracted non-invasively using the Anspec-Pro device, which records skin bioimpedance in real-time. A positive
ΔT
indicates enhanced central excitability, while a negative value suggests dominant inhibition. In a case study of postoperative patients, we show that
ΔT
closely follows and often precedes subjective NRS scores, with correlation coefficients reaching up to 0.86 (
p=0.002
). We also introduce a refined index,
ΔTdyn
, which incorporates the trend and local variability of
ΔT
for improved temporal alignment with reported pain. To address the very limited dataset (three patients, nineteen intervals each), we implemented a data augmentation strategy based on autoregressive modeling of
ΔT
and transfer-function mapping to NRS. This approach enabled the generation of synthetic trajectories per patient, thereby enriching the dataset while maintaining physiological plausibility. Analyses of the augmented data revealed consistent lead–lag patterns, correlations, and Granger causality relationships between
ΔT
,
ΔTdyn
, and NRS, suggesting that
ΔT
may serve both as an anticipatory biomarker of nociceptive activity and as a real-time index aligned with subjective pain reporting. Overall, these results provide proof-of-concept that the Anspec-Pro device can support objective, non-invasive nociceptive tracking in clinical environments.