I am excited to share that we have just released the latest version of our MIRTorch: A PyTorch-based Michigan Image reconstruction toolbox. The work is inspired by MIRT, a well-acclaimed toolbox for medical imaging reconstruction developed by Professor Jeff Fessler and his group.
Update of MIRTorch
The new release of MIRTorch brings a host of new features for computerized tomography in nuclear medicine and radiology, such as forward-backward projectors for single photon emission computerzied tomography (SPECT), branchless distance-driven (bdd) CT, and iterative algorithms like maximum-likelihood expectation-maximization (MLEM). With this new release, MIRTorch offers enhanced capabilities for deep learning in multimodal imaging modalities like SPECT/CT, PET/CT and PET/MRI that can provide researchers and practitioners with more advanced AI tools for medical image reconstruction.