Radiomics based targeted radiotherapy planning (Rad-TRaP): a computational framework for prostate cancer treatment planning with MRI

Table 1 A description of the two retrospective imaging datasets employed in this work

Dataset	Number of patients	Imaging	Resolution (pixels)	Slice thickness (mm.)
D ₁	12	T2w (small FOV)	192×192 - 256×256	2.0-3.5
	12	T2w (large FOV)	320×320 - 512×512	3.0-5.0
	12	DWI	192×192 - 256×256	2.0-3.5
	12	CT (post-brachytherapy)	512×512	3.0
	1	CT (pre-EBRT)	512×512	3.0
D ₂	11	T2w	256×256 - 512×512	3.0-3.3
	11	DWI	256×256 - 512×512	3.0-3.3

D ₂ was used in the training of radiomics based machine learning classifier. This classifier was used to obtain cancer delineations for D ₁

ISSN: 1748-717X