Study of Alternative Imaging Methods for In Vivo Boron Neutron Capture Therapy

Accurate in vivo boron dosimetry is crucial for successfully implementing Boron Neutron Capture Therapy in clinical settings. This investigation uses a Compton camera detector and Monte Carlo algorithms to evaluate different imaging methods for dosimetry and tumor monitoring. The study employed the Maximum Likelihood Expectation Maximization method for dosimetry tomography, while morphological filtering and Deep Learning techniques with Convolutional Neural Networks were explored for tumor monitoring. The research is significant as it addresses the critical need for precise in vivo boron dosimetry in Boron Neutron Capture Therapy. The findings highlight the potential of the Maximum Likelihood Expectation Maximization method for accurately assessing the boron dose and demonstrate the promising results of the Convolutional-Neural-Network-based approach for tumor monitoring. The research emphasizes the importance of optimizing imaging methods and clinical parameters in this treatment, paving the way for improved treatment outcomes and enhanced patient care.