口腔内超声图像的放射组学分析预测舌癌患者颈淋巴结转移的研究。

Radiomics analysis of intraoral ultrasound images for prediction of late cervical lymph node metastasis in patients with tongue cancer.

机构信息

Department of Oral and Maxillofacial Radiology, Hiroshima University Hospital, Hiroshima, Japan.

Department of Oral and Maxillofacial Radiology, Graduate School of Biomedical and Health Sciences, Hiroshima University, Hiroshima, Japan.

出版信息

Head Neck. 2023 Oct;45(10):2619-2626. doi: 10.1002/hed.27487. Epub 2023 Aug 16.

DOI:10.1002/hed.27487

PMID:37584449

Abstract

BACKGROUND

We investigated the predictability of late cervical lymph node metastasis using radiomics analysis of ultrasonographic images of tongue cancer.

METHODS

We selected 120 patients with tongue cancer who underwent intraoral ultrasonography, 30 of which had late cervical lymph node metastasis. Radiomics analysis was used to extract and quantify the image features. Bootstrap forest (BF), support vector machine (SVM), and neural tanh boost (NTB) were used as the machine learning models, and receiver operating characteristic curve analysis was conducted to determine diagnostic performance.

RESULTS

The sensitivity, specificity, accuracy, and AUC in the validation group were, respectively, 0.600, 0.967, 0.875, and 0.923 for the BF model; 0.700, 0.967, 0.900, and 0.950 for the SVM model; and 0.900, 0.967, 0.950, and 0.967 for NTB model.

CONCLUSIONS

Radiomics analysis and machine learning models using ultrasonographic images of pretreated tongue cancer could predict late cervical lymph node metastasis with high accuracy.

摘要

背景

我们通过对舌癌超声图像的影像组学分析来研究颈淋巴结转移的预测。

方法

我们选择了 120 例接受口腔内超声检查的舌癌患者，其中 30 例有颈淋巴结转移。采用影像组学分析方法提取和量化图像特征。Bootstrap 森林（BF）、支持向量机（SVM）和神经正切提升（NTB）被用作机器学习模型，并进行接收者操作特征曲线分析以确定诊断性能。

结果

在验证组中，BF 模型的敏感性、特异性、准确性和 AUC 分别为 0.600、0.967、0.875 和 0.923；SVM 模型的敏感性、特异性、准确性和 AUC 分别为 0.700、0.967、0.900 和 0.950；NTB 模型的敏感性、特异性、准确性和 AUC 分别为 0.900、0.967、0.950 和 0.967。