Dual-weighted frequency difference magnetic induction tomography for acute intracranial hemorrhage detection

Abstract:
Multi-frequency magnetic induction tomography (mfMIT) is considered a promising technique for the early detection of intracranial hemorrhage. However, the lack of practical and stable algorithms has limited the use of mfMIT in clinical application research. In this study, we developed a dual-weighted frequency difference (DWFD) algorithm to address the limitations of the conventional weighted frequency difference method. This method does not require a priori information about the electrical conductivity of each tissue in the brain and only requires MIT measurements at three excitation frequencies to reconstruct an image of the hemorrhagic target. The performance of the proposed algorithm was verified using a three-dimensional head simulation model with an actual anatomical structure and comparing it with that of the conventional method. The results of the simulation experiments revealed that the DWFD algorithm can extract the hemorrhage tissue information and reconstruct low-noise images in a complex brain model with multiple tissue backgrounds. Through this method, the influence of non-homogeneous tissues in the head, which is observed in the conventional method, is eliminated; thus, the method can be used in further experiments. The DWFD method can facilitate the use of mfMIT in clinical studies for the rapid detection of intracranial hemorrhage.