Using nnUNet V1 to infer the predicted results, it does not work in the final step

[angleboy8]

When the pretrained model generated by nnUNetV1 is used, nnUNetV1 does not work in the final step that exports the segmentation result. It will always remain in this state. I think that this is because the size of the input CT sample is too large, namely, 103.5MB. The RAM of my PC is 32GB and the GPU is 4090. Please how to resolve this issue. Thank you for your help very much. Looking forward to your reply.