Machine Learning: Applications and Advanced Progresses of Radiomics in Endocrine Neoplasms

Table 2

Different clinical applications of radiomic models (features) in thyroid cancer.

References	Case numbers	Radiomic method	Results
[29]	55	^18F-FDG-PET/CT	Radiomic features had the potential to diagnose malignant thyroid cancer
[39]	44	MRI	The textural analysis classifies thyroid nodules with high sensitivity and specificity on multi-institutional DW-MRI data sets
[34]	450	US	US-based radiomics had the potential to predict the lymph node metastasis of PTC preoperatively
[35]	189	US	The accuracy of the US-based radiomic method was much higher than that of US examination in the prediction of metastasis of PTC
[36]	43	MRI	Radiomic models may have the potential to differentiate benign from malignant nodules
[37]	527	US	Radiomic features had limited values as a noninvasive biomarker for predicting clinical aggressive behaviors
[38]	400	US	US radiomic features of the primary tumor were associated with lateral cervical lymph node status
[40]	1576	US	A CADx system using CNN-combinations may help radiologists make decisions by overcoming interobserver variability when assessing thyroid nodules on US
[41]	624	CT	Radiomic model had the potential to predict ETE preoperatively in patients with PTC
[42]	768	US	Radiomic features were significantly associated with disease-free survival