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

Table 4

Different clinical applications of radiomics in ovarian tumors.

References	Case numbers	Radiomic method	Results
[130]	187	US	US-based radiomics could be efficiently used for developing the classification stage in ovarian tumor
[131]	59	CT	CT features of serous borderline tumors were distinct from low-grade serous carcinomas
[132]	Not mentioned	OCT	3D texture analysis of OCT was useful for quantitatively characterizing ovarian tissue
[133]	38	OCT	OCT-based radiomics had the potential to classify different subtypes of ovarian tissue
[134]	10	SHG	SHG texture analysis had the potential for ovarian cancer classification
[135]	10	SHG	3D SHG texture analysis achieved high accuracy for classifying high-grade cancer tissue and normal ovarian tissue
[136]	8	SHG	Metastatic tissue images features were distinct from that of healthy tissues
[137]	91	CT	CT-based radiomics had the potential to predict responses of ovarian cancer patients to chemotherapy
[138]	120	CT	CT-based radiomic features computed from both spatial and frequency domains had a reliable prediction ability of tumor response to postsurgical chemotherapy
[139]	364	CT	Radiomic prognostic vector (RPV) could be exploited to personalized therapy of epithelial ovarian cancer (EOC) and had the potential to apply in other cancer types
[140]	38	CT	Quantitative metrics noninvasively capturing spatial intersite heterogeneity may predict outcomes in patients with HGSOC
[141]	142	CT	Radiomic signature was potential prognostic markers that may allow for individualized evaluation of patients with advanced HGSOC