A breathing pattern comprises multiple variable components, including air volume, timing indices, and chest wall movement parameters, coordinated by the integrated function of physiological systems such as neural pathways, vital organs, and respiratory muscles [1]. Pathological alterations in breathing patterns, commonly referred to as dysfunctional breathing, are observed in several respiratory diseases and can serve as markers of disease control, particularly in conditions such as asthma [2, 3]. However, in clinical practice, objective measurement of changes in breathing pattern components remains challenging due to the lack of a monitoring system capable of simultaneously and instantaneously quantifying all key parameters. In clinical practice, the assessment of alterations in breathing patterns often relies on validated questionnaires due to their ease of use and practical applicability. Among the most frequently employed instruments are the Nijmegen Questionnaire, the Self-Evaluation of Breathing Questionnaire (SEBQ), and the manual assessment of respiratory motion (MARM), which primarily depend on patient perceptions and clinician experience [2]. However, these subjective measures lack the capacity to provide precise, quantifiable data on breathing patterns, thus limiting the ability to achieve a comprehensive and holistic respiratory assessment.

Over the years, non-invasive motion-analysis systems have been developed to objectively record thoracoabdominal (TA) movements and quantify breathing pattern components [4, 5]. The most widely used systems include the respiratory inductive plethysmograph (RIP), which measures chest wall motion using inductive coils embedded in two elastic bands, and the optoelectronic plethysmograph (OEP), which utilizes reflective markers placed on specific anatomical landmarks of the chest wall [6]. These systems have been extensively validated in both healthy and clinical populations under various breathing conditions [6].

Structured light plethysmography (SLP) has been a recently developed, portable, fully contactless digital motion-analysis system designed for clinical application. It enables real-time quantification of timing parameters and TA movement-related indices [7]. The system projects a structured black-and-white grid onto the anterior TA surface, spanning from the clavicles to the umbilicus, and captures respiratory-induced distortions of the grid using two auto-calibrated digital cameras [8]. These distortions are processed to generate a one-dimensional movement trace over time. For optimal image capture, subjects are required to wear a tight-fitting white T-shirt. Unlike RIP and OEP, SLP does not require direct contact or active cooperation from the subject, thereby minimizing the observer effect and allowing for objective measurement of natural breathing patterns and their variability over time [9].

SLP has been employed in clinical research to detect changes in breathing pattern components in patients with COVID-19 [10], chronic obstructive pulmonary disease (COPD) [11], neuromuscular disorders [12], and following lung surgery [13]. Its application has also been explored in pediatric populations across a range of respiratory conditions [14]. Notably, SLP has been shown to distinguish timing parameter differences between healthy children and children with asthma, both before and after bronchodilator administration [15]. Despite these promising findings, there remains limited evidence directly comparing SLP measurements to those obtained from reference-standard motion analysis systems under different breathing conditions. Existing validation studies have primarily focused on comparing SLP performance against pneumotachograph, evaluating timing indices at rest, and typically involving small samples of recorded breath cycles [8].

Therefore, the present study was an attempt to compare the measurement accuracy of SLP with another non-invasive motion analysis system, such as RIP. The validity of SLP was aimed to be assessed in different breathing conditions through a breath-by-breath analysis. It was hypothesized that SLP would provide accurate estimates of timing parameters and TA movement-associated indices both at rest and following exercise compared to RIP. Furthermore, it was hypothesized that SLP performance would be comparable to RIP in detecting changes in these parameters after exercise.

Fifty adults (30 males) with mean age (SD) of 29.26 years old (6.79) participated in the study. Although 7 out of 50 participants self-reported history of asthma, no fixed airway obstruction was identified among the studied sample. The demographic data of study’s sample are presented in Table 1. All the participants successfully completed the 10-minute incremental exercise protocol. At rest, 3,395 and after exercise 4,295 breath cycles were identified and analyzed. A statistically significant change in RR was detected from both devices after exercise [RIP: μ –3.57 bpm, SD 2.96 bpm, 95% confidence interval (CI) –4.41 to –2.73, p < 0.05; SLP: μ –3.57 bpm, SD 2.95 bpm, 95% CI –4.41 to –2.73, p < 0.5].

Significant differences in timing parameters between the monitoring methods were not found both at rest and after exercise with the largest mean difference being found for Te at rest (Table 2). Averaged agreement between the methods for all timing parameters (RR, Ti/Ttot, and Te) was found both at rest and after exercise (Figures 1, 2, and 3 respectively). The majority of the differences (> 92%) for all timing parameters fell within 95% limits of agreement (LOA) and were close to zero over different magnitudes of breath cycles (Figures 1, 2, and 3). Similar results were found regarding breath-by-breath agreement and the measurement agreement for the changes in timing parameters after exercise (Tables 2 and 3, and Figures 1, 2, and 3).

The mean bias between the devices for RCampi and ABampi was found to be 0.31 and 0.23 at rest with the LOA ranging from –0.06 to 0.66 and –0.06 to 0.52 respectively. A potential tendency of greater discrepancies between the devices was found over greater magnitudes of these parameters at rest (Figure 4). This was consistent after exercise with both averaged and breath-by-breath measurement agreement being slightly violated at greater magnitudes of TA movements (Figure 4).

Both devices detected an averaged increase in RCampi (RIP: μ –0.26, SD 0.28, 95% CI –0.18 to 0.05; SLP: μ –0.13, SD 0.19, 95% CI –0.18 to 0.07) and ABampi after exercise. However, a systematic bias was found between the devices for the changes in the RCampi and ABampi (see Table 3) with SLP measuring smaller changes in displacements for both regions of TA area over greater magnitudes compared to the RIP (Figures 4 and 5). In contrast, good averaged and breath-by-breath agreement between the devices was found for RCampi/ABampi both at rest and after exercise (Tables 2 and 3; Figure 6).

The present study aimed to evaluate the criterion validity of SLP in comparison to RIP under resting and post-exercise conditions. The findings indicate that SLP can be regarded as a valid tool for monitoring and quantifying timing indices of breathing parameters at rest. These outcomes align with previously published studies demonstrating agreement between SLP and pneumotachograph in assessing timing-related respiratory variables [8]. However, prior evaluations of SLP accuracy were limited to resting conditions and relatively short recording durations (approximately 45 seconds). In contrast, the current study extends this evidence by confirming the sustained measurement accuracy of SLP over longer recording intervals.

This validation study contributes to the existing body of knowledge on the monitoring capabilities of SLP, demonstrating agreement in timing indices with RIP even during rapid breathing induced by submaximal exercise. The SLP was shown to accurately capture timing parameters across a broad range of breathing cycles, as determined through breath-by-breath analysis. The mean difference in RR measurements between RIP and SLP at rest and post-exercise was smaller than the clinically important difference of 2 bpm [26]. Although previous studies have suggested that SLP can detect differences in timing indices between healthy individuals and patients with COPD, indicating sensitivity to pathological changes, these findings were not previously validated against a reference-standard motion analysis system [13]. The current study confirms that SLP is capable of detecting alterations in timing indices of breathing patterns, with results closely matching those obtained using RIP.

Some contradictory results were found regarding the SLP’s ability to measure breathing parameters associated with TA movement compared to the RIP. The ability of the SLP to measure regional TA movements was previously supported in a small feasibility study (n = 15) where the SLP showed a statistically significant post-operative reduction in the RC contribution to TA movement after lung resection surgery [15]. In the present validation study, where direct comparisons of the RCampi and ABampi were made, both devices measured an average increase in RCampi after exercise confirming the SLP’s ability to detect changes in TA motion under a different breathing condition. Nevertheless, the SLP tends to identify smaller regional TA movements compared to the RIP. The difference was greater after exercise, when RCampi and ABampi were found to become larger. It was also noted that changes in the ratio of these breathing parameters measured by the RIP after exercise were almost double those estimated by the SLP.

These discrepancies could have been expected, since these devices use different recording approaches to quantify breathing kinematics. The RIP records changes in the cross-sectional (anteriorly, latterly, and posteriorly) regional compartments of the TA area [4, 6]. On the other hand, the SLP quantifies motion changes of a larger area of the surface of TA, between the clavicles and umbilicus, but only anteriorly [8, 19]. Studies characterizing breathing kinematics between resting and rapid breathing have shown that ribs tend to move upward, forward, and outward to increase the volume of chest wall during rapid breathing in a sitting position [27]. Since the SLP may not be able to detect fully circumferential changes in TA motion, this may result in the discrepancies found between the two methods of the current study.

To date, it is not clear whether the identified differences between the RIP and the SLP for the amplitudes of regional displacements of TA area can be considered clinically meaningful, due to a lack of normative standardized data. In theory, this may cause ambiguity in estimation of air-volume parameters by the SLP. Although this was beyond the scope of the current study, a growth-adjusted reference equations for tidal breathing parameters using SLP have been reported [28]. On the other hand, the SLP was found to be a valid tool to quantify objectively the ratio of the motion of the two compartments of TA area irrespective of the actual magnitude of the RC and AB displacements over a variety of breath sizes obtained at rest and after exercise.

As mentioned above, SLP is a fully contactless motion analysis system that has demonstrated validity and responsiveness in quantifying and recording both timing indices and TA movements, at rest and following exercise. This novel recording method requires minimal participant cooperation, and its non-contact nature theoretically reduces the likelihood of individuals becoming consciously aware of the breathing assessment—an issue commonly encountered with other non-invasive motion analysis systems such as RIP and OEP. It has been previously supported that SLP may minimize measurement-induced alterations in breathing patterns, thereby enabling a more accurate representation of an individual’s natural respiratory behavior [9]. Combined with its ease of use, these features support the potential of SLP as a valuable tool for the objective assessment of breathing patterns in clinical environments.

One of the limitations of the present study is that it examined the SLP’s accuracy in a relatively young healthy population with normal BMI, even though recruitment of a heterogeneous sample was attempted. This leaves open the SLP’s performance in other groups of people such as those with increased BMI or variable lung function. The study mainly focused on the examination of specific timing indices, RCampi, ABampi (expressed in arbitrary units), and thereby future validation studies could further evaluate the SLP’s performance for other physiological respiratory parameters such as air-volume components and breathing asynchronism by estimating the labored breathing index (LBI) as an indicator of breathing asynchronism in other populations in several positions. In the present study, breathing pattern components were recorded only in sitting position, partly because it is one of the common postures for evaluating breathing patterns in clinical practice.

As a result, future research should aim to expand our current knowledge regarding the SLP’s performance for quantifying a series of breathing pattern components in different postures compared to a standard reference motion analysis system. Finally, the reliability of the SLP was not specifically examined due to the primary interest of the current study design. However, previously published evidence has shown that SLP can provide reliable measurements of timing components during two consecutive resting breathing sessions as compared to a pneumotachograph [8]. Further work is needed to test its reliability regarding the measurements of TA movements over time.