Table 7.

Summary of radiomics in head and neck neoplasms.

Imaging technique	Type of tumor	No. of patients	Predictive model	Test accuracy indices			References
				Sens.	Spec.	Accuracy
CT scan	Thyroid nodule	103	Tumor type cytokeratin 19	93%	73%	87%	(65)
			Tumor type galectin 3	87%	76%	85%
			Tumor type thyroperoxidase	86%	75%	84%
Ultrasound	Thyroid nodule	137 in training set 95 in validation set	Malignant nodules	–	–	93%	(66)
CT scan	HNSCC	136 in training set 95 in validation set	Multivariate HNC signature to predict tumor stage	–	–	80%	(67)
MRI	HNSCC	85 in training set 42 in validation set	Combined T2W and ceT1W radiomic features to predict tumor stage	84%	70%	79%	(68)
			ceT1W radiomic features to predict tumor stage	80%	67%	75%

HNSCC, head and neck squamous cell carcinoma; HNC, head and neck cancer.