K-Nearest Neighbor Based Locally Connected Network for Fast Morphological Reconstruction in Fluorescence Molecular Tomography.

Hui Meng,Yuan Gao,Xin Yang,Jie Tian,Kun Wang

doi:10.1109/tmi.2020.2984557

Abstract

Fluorescence molecular tomography (FMT) is a highly sensitive and noninvasive imaging modality for three-dimensional visualization of fluorescence probe distribution in small animals. However, the simplified photon propagation model and ill-posed inverse problem limit the improvement of FMT reconstruction. In this work, we proposed a novel K-nearest neighbor based locally connected (KNN-LC) network to improve the performance of morphological reconstruction in FMT. It directly builds the inverse process of photon transmission by learning the mapping relation between the surface photon intensity and the distribution of fluorescent source. KNN-LC network cascades a fully connected (FC) sub-network with a locally connected (LC) sub-network, where the FC part provides a coarse reconstruction result and LC part fine-tunes the morphological quality of reconstructed result. To assess the performance of our proposed network, we implemented both numerical simulation and in vivo studies. Furthermore, split Bregman-resolved total variation (SBRTV) regularization method and inverse problem simulation (IPS) method were utilized as baselines in all comparisons. The results demonstrated that KNN-LC network achieved accurate reconstruction in both source localization and morphology recovery in a short time. This promoted the in vivo application of FMT for visualizing the distribution of biomarkers inside biological tissue.

Highlights

F LUORESCENCE molecular imaging (FMI) is a widely applied optical imaging modality for numerous preclinical applications [1]–[3]
In vivo fluorescence molecular tomography (FMT) reconstruction of probe distribution in glioma mouse models were carried out to verify the practicality of K-nearest neighbor based locally connected (KNN-locally connected (LC)) network
Based on the priori knowledge that the adjacent voxels tend to have similar fluorescence intensity, we built a KNN-LC network by constructing a LC sub-network based on K-nearest neighbor and cascading it behind a fully connected (FC) sub-network

Summary

Introduction

F LUORESCENCE molecular imaging (FMI) is a widely applied optical imaging modality for numerous preclinical applications [1]–[3]. Fluorescence molecular tomography (FMT) has been developed to trace the fluorescent source location and recover the distribution of fluorescence probes [7]. The complexity of photon propagation model and the ill-posedness of inverse problem still limit the development of FMT reconstruction. To alleviate the modeling error and solve the ill-posed inverse problem in FMT reconstruction, many researchers have developed different model-based methods. Numerous regularization terms [12]–[14] were added to the optimization methods to alleviate the ill-posedness of inverse problem. These methods improved the performance of FMT reconstruction, the deviation between simplified photon propagation model and actual process of light propagation still limits the accuracy of FMT reconstruction. The model-based method [15], [16] usually solved the inverse problem by iterative optimization, which was time consuming

Methods

Results

Conclusion