Deep Learning-Based Prediction of Final Infarct Core from CT Perfusion Data: A Comparison to the Clinical Standard.

It is notable that the 50% probability band was the most accurate for the DL model even though the Bland–Altman analysis showed a trend toward overestimation, and the correlation coefficient for predicted and actual core volumes was stronger for the 90% probability band map than the 50% probability band map.

also in 53,322 other papers