Toward Noninvasive High-Resolution In Vivo pH Mapping in Brain Tumors by <sup>31</sup>P-Informed deepCEST MRI.
Authors
Affiliations (6)
Affiliations (6)
- Institute of Neuroradiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany.
- Institute of Neuroradiology, Goethe University Hospital Frankfurt, Goethe University, Frankfurt am Main, Germany.
- Cooperative Brain Imaging Center (CoBIC), Goethe University Frankfurt, Frankfurt am Main, Germany.
- Department Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.
- Institute of Radiology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
- Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany.
Abstract
The intracellular pH (pH<sub>i</sub>) is critical for understanding various pathologies, including brain tumors. While conventional pH<sub>i</sub> measurement through <sup>31</sup>P-MRS suffers from low spatial resolution and long scan times, <sup>1</sup>H-based APT-CEST imaging offers higher resolution with shorter scan times. This study aims to directly predict <sup>31</sup>P-pH<sub>i</sub> maps from CEST data by using a fully connected neuronal network. Fifteen tumor patients were scanned on a 3-T Siemens PRISMA scanner and received <sup>1</sup>H-based CEST and T1 measurement, as well as <sup>31</sup>P-MRS. A neural network was trained voxel-wise on CEST and T1 data to predict <sup>31</sup>P-pH<sub>i</sub> values, using data from 11 patients for training and 4 for testing. The predicted pH<sub>i</sub> maps were additionally down-sampled to the original the <sup>31</sup>P-pH<sub>i</sub> resolution, to be able to calculate the RMSE and analyze the correlation, while higher resolved predictions were compared with conventional CEST metrics. The results demonstrated a general correspondence between the predicted deepCEST pH<sub>i</sub> maps and the measured <sup>31</sup>P-pH<sub>i</sub> in test patients. However, slight discrepancies were also observed, with a RMSE of 0.04 pH units in tumor regions. High-resolution predictions revealed tumor heterogeneity and features not visible in conventional CEST data, suggesting the model captures unique pH information and is not simply a T1 segmentation. The deepCEST pH<sub>i</sub> neural network enables the APT-CEST hidden pH-sensitivity and offers pH<sub>i</sub> maps with higher spatial resolution in shorter scan time compared with <sup>31</sup>P-MRS. Although this approach is constrained by the limitations of the acquired data, it can be extended with additional CEST features for future studies, thereby offering a promising approach for 3D pH imaging in a clinical environment.