Machine Learning Based on a Multiparametric and Multiregional Radiomics Signature Predicts Radiotherapeutic Response in Patients with Glioblastoma

Table 1

Characteristics of patients in the TCIA and independent test datasets.

Characteristic	TCIA ()	Huadong ()
Ages (years)			0.856
Range	17-80	18-73
Median	57.5	54
Gender, No. (%)				0.847
Female	40 (39.22%)	13 (43.33%)
Male	62 (60.78%)	17 (56.67%)
Status, No. (%)			0.0011
Alive	10 (9.8%)	11 (36.67%)
Dead	92 (90.2%)	19 (63.33%)
KPS			0.2795
	75	21
	17	9
Not reported	10	0
Tumour location			0.377
Frontal lobe	24	12
Temporal lobe	43	11
Parietal lobe	19	4
Occipital lobe	8	3
Insular lobe	6	0
Callosum lobe	2	0
OS (months)			0.6516
Range	1-74.87	3.3-52.43
Median	14.30	14.77
31-gene prediction result
RH	89	—
RR	13	—
Radiomics prediction result			0.8813
RH	85	24
RR	17	6