TABLE 2.

Mean Dice value for each tissue outcome prediction method with and without applying additional filtering operations to each tissue outcome prediction for the purpose of noise reduction.

	Tmax thresholding	Random decision forest	Param-UNet	RC-Simple	RC-Causal	CTC-Causal
Noise-removal post-processing	0.248 (0.217)	0.262 (0.213)	0.284 (0.229)	0.277 (0.228)	0.286 (0.229)	0.297 (0.235)
No noise-removal post-processing	0.249 (0.214)	0.233 (0.199)	0.287 (0.229)	0.276 (0.228)	0.286 (0.228)	0.296 (0.234)

Bold cells indicate the models chosen to represent each architecture in subsequent analyses.

Abbreviated model names correspond to: Tmax thresholding, random decision forests (RDF), deep learning from perfusion parameter maps (Param-UNet), deep learning from deconvolved residual curves with convolutional (RC-Simple) or causal convolutional (RC-Causal) feature extraction, and deep learning from source concentration-time curves (CTC-Causal).