Using texture analysis of ultrasonography images of neck lymph nodes to differentiate metastasis to non-metastasis in oral maxillary gingival squamous cell carcinoma

Abstract

Object

To differentiate between metastatic neck nodes and non-metastatic neck nodes in oral maxillary gingival squamous cell carcinoma, textural analysis of these lymph nodes in ultrasound images was performed in this study.

Methods

Twenty five metastatic neck nodes and 28 non-metastatic neck nodes were enrolled in this study. Seventy eight texture characteristics were retrieved from the US images using the LIFEx software.

The Mann Whitney U test was measurably utilized to survey on the off chance that there was a measurably noteworthy distinction within the textural characteristics between metastatic neck nodes and non-metastatic neck nodes. The capacity of the surface highlights to recognize between metastatic neck nodes and non-metastatic neck nodes was illustrated utilizing the Receiver Operating Characteristic analysis curves (ROC). Youden's J statistic was used to determine the cut-off positions in each ROC curve that maximized sensitivity and specificity.

Results

Zone size non uniformity (ZSNU) highlight appeared the foremost noteworthy contrast between these nodes (p < 0.001).

Strength had Area Under the Curve (AUC) of 0.811, specificity of 0.821 and sensitivity of 0.8, when measured at the cutoff value of 896.344.

Conclusions

Our results come about uncovered that quality highlight may be the finest surface highlight to distinguish from non-metastatic neck nodes and to anticipate metastatic neck nodes in oral maxillary gingival squamous cell carcinoma.