A new technique of tissue stiffness (compliance) assessment: its reliability, accuracy and comparison with an existing method

Abstract

Objective: To assess the reliability and accuracy of a new method of tissue stiffness (TS) assessment, automated stiffness assessment (ASA), and compare these findings with previously published results of manual stiffness assessment (MSA).

Design: Descriptive study.

Setting: Human Performance Laboratory, University of Calgary.

Intervention: ASA was used to collect stiffness measurements from three foam surfaces of different stiffness characteristics and two control surfaces. Control surface one (CS1) was rigid and could not be displaced; control surface two (CS2) was rigid but could be displaced. The three foam surfaces and CS1 had been previously tested using MSA.

Main outcome measure: Reliability and accuracy were assessed for each separate electronic component by root mean square analysis (RMS) and linear regression/calibration curves, respectively. The reliability of ASA (all electronic components working in concert) was determined by computing intraclass correlation coefficients (ICC); the accuracy of ASA was assessed by studying relative deformations of the control surfaces.

Results: Each electronic component used in ASA was found to have a calculated RMS of less than 0.03% of the mean, whereas the R2-value for any of these separate components was never lower than 0.99. For ASA, the median ICC for all surfaces tested was 0.99. The mean relative deformation collected from CS1 was 0.34 +/- mm at 44.0 N of input force whereas the mean relative deformation collected from CS2 was 0.008 +/- 0.013 mm. The median ICC for MSA found in a previous investigation was 0.005 and the mean displacement recorded from the control surface with MSA was 1.28 +/- 0.57 mm at 49.10 N.

Conclusion: ASA seems to be extremely reliable and accurate. When ASA and MSA were used to assess identical test surfaces, the results of MSA were poor when compared with those obtained by ASA; we therefore suggest that conclusions reached in prior research employing MSA should be considered with caution. It is our recommendation that remotely controlled, electronic signal gathering procedures such as ASA be the method of choice when assessing TS.