The practical clinical role of machine learning models with different algorithms in predicting prostate cancer local recurrence after radical prostatectomy

Table 1 MRI sequences and parameters for radiomics analysis

Parameters	T2WI	DWI	DCE
Repetition time (ms)	6980.0	6540.0	4.2
Echo time (ms)	104.00	64.00	1.34
Layer thickness (mm)	3	3	3
Interlayer spacing (mm)	0	0	0
Field of view (mm × mm)	200 × 200	220 × 220	240 × 240
Imaging matrix	384 × 384	130 × 130	224 × 224
B-value (s/mm²)	/	50, 1000, 1500	/
FlipAngle	120	180	12
PixelSpacing (mm)	0.52 × 0.52	1.69 × 1.69	1.07 × 1.07

After T2WI, T1WI and DWI acquisitions, we administered a dose of gddiethylenetriaminepenta-acetic acid (0.05 mmol/kg; 3mL/sec; Magnevist, Bayer, Berlin, Germany) through a 20G antecubital intravenous line by using an MR-compatible injector (Spectris; Medrad, Pittsburgh, PA) to acquire DCE images. T2WI = T2-weighted imaging; T1WI = T1-weighted imaging; DWI = diffusion-weighted imaging; DCE = dynamic contrast-enhanced

ISSN: 1470-7330