Skip to main content
. Author manuscript; available in PMC: 2022 Nov 2.
Published in final edited form as: Brainlesion. 2022 Jul 22;12962:151–167. doi: 10.1007/978-3-031-08999-2_12

Fig. 4.

Fig. 4.

Qualitative comparison of results for one of the subjects with high resolution T1 scans across the 3 visualization slices. “GT” is the ground truth mask, “PT-FP32” is the mask generated by the original PyTorch FP32 model, “OV-FP32” is the output of the optimized model in FP32, and “OV-INT8” is the output of the optimized model after quantizing to INT8.