Real Time Estimation Method for both the Location and the Size of Tumor Using NIR Fluorescence Images

Abstract

A reconstruction technique, such as Fluorescence Molecular Tomography or Fluorescence Diffusion Optical Tomography, permits us to detect a tumor genesis, cancer detection and so on using the Photo Dynamic Diagnosis(PDD) images. However, these techniques require huge computational cost and the introduction into intra operation has some difficulty about the surgical possibility in real time and narrow field. In this study, we propose a method to identify the fluorescence source simulated as a tumor in real time process using Near Infrared(NIR) fluorescence images, the optimization approach with Weighted Residual Method (WRM). In order to verify the location accuracy of the proposed method, we demonstrated the tumor identification using tissue phantom and a fluorescence tube filled with fluorescence dye (AlPcS). The absolute errors between the true positions and the estimated positions in the experiments were less than 1mm. The estimated time was about seventy mill-seconds. To confirm the validity for living bodies, we carried out the experimental identification using small animals. The fluorescent micro-tubes (diameter 0.4mm) filled with AlPcS were inserted in the brains of the euthanized mice. The errors between the true positions and the estimated positions in the animal experiments were about 1.6mm. Consequently, we confirm that this proposed method is effective as an assisted system in a navigation surgery.