A total of 80 ultrasound still images, derived from 40 individual cases, were systematically evaluated by the radiologists. Each case contributed two images—one acquired using the classic linear probe (CLP) and the other obtained with the high-frequency probe (HFP)—ensuring a direct comparison of imaging quality and diagnostic consistency between the two modalities.

The overall inter-reader agreement across all participants was found to be "substantial" for both ultrasound probes, with a kappa coefficient (κ) of 0.71 for CLP and a slightly higher κ of 0.78 for HFP. This indicates that while both probes facilitated a relatively high level of consistency among readers, the use of high-frequency ultrasound led to a modest improvement in agreement, suggesting better lesion characterization and diagnostic confidence.

When stratified by experience level, senior radiologists demonstrated a "substantial" level of agreement when interpreting images obtained with the CLP (κ = 0.78). However, when utilizing the HFP, their agreement reached an "almost perfect" level (κ = 0.86), highlighting the added benefit of enhanced resolution and detail provided by high-frequency ultrasound in the hands of experienced practitioners.

For radiology residents, inter-reader agreement was lower overall but still showed notable improvements with the use of HFP. When assessing images obtained with the CLP, the agreement was "moderate" (κ = 0.59), reflecting a higher degree of variability in interpretation. However, with the HFP, the agreement improved to a "substantial" level (κ = 0.73), indicating that even less experienced readers benefited from the superior imaging capabilities of high-frequency ultrasound, which likely facilitated more accurate lesion differentiation. [Table 1]