Bioluminescence Tomography Based on One-Dimensional Convolutional Neural Networks

Bioluminescent tomography (BLT) has increasingly important applications in preclinical studies. However, the simplified photon propagation model and the inherent ill-posedness of the inverse problem limit the quality of BLT reconstruction. In order to improve the reconstruction accuracy of positioning and reconstruction efficiency, this paper presents a deep-learning optical reconstruction method based on one-dimensional convolutional neural networks (1DCNN). The nonlinear mapping relationship between the surface photon flux density and the distribution of the internal bioluminescence sources is directly established, which fundamentally avoids solving the ill-posed inverse problem iteratively. Compared with the previous reconstruction method based on multilayer perceptron, the training parameters in the 1DCNN are greatly reduced and the learning efficiency of the model is improved. Simulations verify the superiority and stability of the 1DCNN method, and the in vivo experimental results further show the potential of the proposed method in practical applications.

Automated summary

The paper introduces a deep-learning optical reconstruction method based on 1DCNN to improve the accuracy and efficiency of BLT reconstruction. By establishing the nonlinear mapping relationship between photon flux density and bioluminescence source distribution, it avoids solving the ill-posed inverse problem iteratively, reducing training parameters and improving model learning efficiency. Simulation and in vivo experimental results demonstrate the superiority and stability of the 1DCNN method in practical applications.