Learning curve and interobserver variance in quantification of the optical coherence tomography attenuation...

Area of Science:

• Biomedical Optics
• Medical Imaging Analysis
• Gynecologic Pathology

Background:

• Optical coherence tomography (OCT) is a non-invasive imaging technique used for visualizing tissue microstructure.
• Quantifying tissue optical properties, such as the attenuation coefficient (μOCT), is crucial for OCT-based tissue characterization and classification.
• Understanding the learning curve and interobserver variability in μOCT determination is essential for its clinical application.

Purpose of the Study:

• To quantify the learning curve and interobserver variance associated with determining the attenuation coefficient (μOCT) from optical coherence tomography (OCT) images.
• To assess the feasibility of routine μOCT determination for vulvar tissue classification without extensive specialized training.

Main Methods:

• Novice students and OCT experts determined μOCT values from normal and diseased vulvar tissue images.
• Measurements were taken at five distinct time points, with students receiving inter-session feedback.
• Expert consensus was used as the reference standard for comparison.

Main Results:

• Interobserver variance in μOCT determination was found to be smaller for diseased vulvar tissues compared to normal tissues.
• The discrepancy between expert consensus and student-derived μOCT values increased as μOCT values decreased.
• The learning curve indicated that proficiency in μOCT determination was achieved relatively quickly.

Conclusions:

• Routine determination of μOCT for vulvar tissue classification does not necessitate extensive training.
• The method shows potential for reliable tissue classification, particularly for diseased tissues, with minimal observer experience.
• Further validation in larger cohorts may confirm the clinical utility of simplified μOCT analysis.