Temperature-aware postmortem brain MRI simulator

Abstract

Background

Postmortem brain MRI contrast depends on tissue temperature, which varies widely across imaging settings and postmortem interval. Consequently, image appearance diverges from in-vivo references, and a normal postmortem reference distinct from the living brain is needed. However, a comprehensive, temperature-aware educational reference is lacking. This study aimed to build a temperature-aware postmortem brain MRI simulator that enables interactive exploration of sequence parameters and provides a normal postmortem reference for education and protocol design.

Postmortem brain MRI contrast depends on tissue temperature, which varies widely across imaging settings and postmortem interval. Consequently, image appearance diverges from in-vivo references, and a normal postmortem reference distinct from the living brain is needed. However, a comprehensive, temperature-aware educational reference is lacking. This study aimed to build a temperature-aware postmortem brain MRI simulator that enables interactive exploration of sequence parameters and provides a normal postmortem reference for education and protocol design.

Methods

Data from 36 postmortem brains were used to fit tissue-specific linear relations between temperature and T1, T2, T2*, and proton density in white matter (WM), gray matter (GM), basal ganglia (BG), and cerebrospinal fluid (CSF). Voxelwise signals were computed from analytical (Bloch based) equation for spin echo (SE), inversion recovery SE, and gradient echo. A 3D brain was formed from brain maps registered in the Montreal Neurological Institute's 152 brain template (MNI152) after slight modification. The simulator displays synchronized triplanar (axial, coronal, sagittal) views and an intersection centered MPR slice/ presets (T1WI, T2WI, FLAIR, T2*WI) configure TR/TE/TI and flip angle. Temperatures spanning 0–50 °C were modeled. Python was used to model temperature-dependent relaxation and simulate voxelwise signals (NumPy/SciPy) and to implement the interactive MPR visualization (PyQt/Matplotlib).

Data from 36 postmortem brains were used to fit tissue-specific linear relations between temperature and T1, T2, T2*, and proton density in white matter (WM), gray matter (GM), basal ganglia (BG), and cerebrospinal fluid (CSF). Voxelwise signals were computed from analytical (Bloch based) equation for spin echo (SE), inversion recovery SE, and gradient echo. A 3D brain was formed from brain maps registered in the Montreal Neurological Institute's 152 brain template (MNI152) after slight modification. The simulator displays synchronized triplanar (axial, coronal, sagittal) views and an intersection centered MPR slice/ presets (T1WI, T2WI, FLAIR, T2*WI) configure TR/TE/TI and flip angle. Temperatures spanning 0–50 °C were modeled. Python was used to model temperature-dependent relaxation and simulate voxelwise signals (NumPy/SciPy) and to implement the interactive MPR visualization (PyQt/Matplotlib).

Results

The simulator reproduced the expected WM and GM contrast inversion on T1 weighted SE at low temperatures and identified temperature dependent optimal TI for FLAIR that suppresses CSF while preserving the normal tissue hierarchy across 0 to 50 °C. Temperature driven contrast trends for T1WI, T2WI, FLAIR, and T2*WI were visualized consistently across triplanar views and an intersection-centered MPR slice and matched a normal reference postmortem appearance derived from our dataset.

The simulator reproduced the expected WM and GM contrast inversion on T1 weighted SE at low temperatures and identified temperature dependent optimal TI for FLAIR that suppresses CSF while preserving the normal tissue hierarchy across 0 to 50 °C. Temperature driven contrast trends for T1WI, T2WI, FLAIR, and T2*WI were visualized consistently across triplanar views and an intersection-centered MPR slice and matched a normal reference postmortem appearance derived from our dataset.

Conclusion

We developed a temperature-aware postmortem brain MRI simulator centered on a normal postmortem reference, enabling contrast-focused education and interactive sequence-parameter exploration across temperatures for protocol design.

We developed a temperature-aware postmortem brain MRI simulator centered on a normal postmortem reference, enabling contrast-focused education and interactive sequence-parameter exploration across temperatures for protocol design.