Interrater reliability of quantitative ultrasound using force feedback among examiners with varied levels of experience

PeerJ. 2016 Jun 21;4:e2146. doi: 10.7717/peerj.2146. eCollection 2016.

Abstract

Background. Quantitative ultrasound measures are influenced by multiple external factors including examiner scanning force. Force feedback may foster the acquisition of reliable morphometry measures under a variety of scanning conditions. The purpose of this study was to determine the reliability of force-feedback image acquisition and morphometry over a range of examiner-generated forces using a muscle tissue-mimicking ultrasound phantom. Methods. Sixty material thickness measures were acquired from a muscle tissue mimicking phantom using B-mode ultrasound scanning by six examiners with varied experience levels (i.e., experienced, intermediate, and novice). Estimates of interrater reliability and measurement error with force feedback scanning were determined for the examiners. In addition, criterion-based reliability was determined using material deformation values across a range of examiner scanning forces (1-10 Newtons) via automated and manually acquired image capture methods using force feedback. Results. All examiners demonstrated acceptable interrater reliability (intraclass correlation coefficient, ICC = .98, p < .001) for material thickness measures obtained using force feedback. Individual examiners exhibited acceptable reliability with the criterion-based reference measures (ICC > .90, p < .001), independent of their level of experience. The measurement error among all examiners was 1.5%-2.9% across all applied stress conditions. Conclusion. Manual image capture with force feedback may aid the reliability of morphometry measures across a range of examiner scanning forces, and allow for consistent performance among examiners with differing levels of experience.

Keywords: Diagnostic imaging; Morphometry; Quantitative ultrasound; Reliability; Robotics; Sensorimotor feedback.