Multispectral bioluminescence tomography (BLT) attracts increasing more attention in the area of small animal studies because multispectral data acquisition could help in the 3D location of bioluminescent sources. Generally, BLT problem is ill-posed and a priori information is indispensable to reconstruction bioluminescent source uniquely and quantitatively. In this paper, we propose a spectrally solved bioluminescence tomography algorithm with an optimal permissible source region strategy. Being the most different from earlier studies, an optimal permissible source region strategy which is automatically selected without human intervention is developed to reduce the ill-posedness of BLT and therefore improves the reconstruction quality. Furthermore, both numerical stability and computational efficiency benefit from the strategy. In the numerical experiments, a heterogeneous phantom is used to evaluate the proposed algorithm and the synthetic data is produced by Monte Carlo method for avoiding the inverse crime. The results demonstrate the feasibility and potential of our methodology for reconstructing the distribution of bioluminescent sources.

No abstract

Diffuse optical tomography (DOT) is a promising noninvasive imaging modality and is capable of providing functional characteristics of biological tissue by quantifying optical parameters. The DOT image reconstruction is ill-posed and ill-conditioned, due to the highly diffusive nature of light propagation in biological tissues and limited boundary measurements. The widely used regularization technique for DOT image reconstruction is Tikhonov regularization, which tends to yield oversmoothed and low-quality images containing severe artifacts. It is necessary to accurately choose a regularization parameter for Tikhonov regularization. To overcome these limitations, we develop a noniterative reconstruction method, whereby optical properties are recovered based on a back-propagation neural network (BPNN). We train the parameters of BPNN before DOT image reconstruction based on a set of training data. DOT image reconstruction is achieved by implementing a single evaluation of the trained network. To demonstrate the performance of the proposed algorithm, we compare with the conventional Tikhonov regularization-based reconstruction method. The experimental results demonstrate that image quality and quantitative accuracy of reconstructed optical properties are significantly improved with the proposed algorithm.

Early brain tumor detection and diagnosis are critical to clinics. Thus segmentation of focused tumor area needs to be accurate, efficient, and robust. In this paper, we propose an automatic brain tumor segmentation method based on Convolutional Neural Networks (CNNs). Traditional CNNs focus only on local features and ignore global region features, which are both important for pixel classification and recognition. Besides, brain tumor can appear in any place of the brain and be any size and shape in patients. We design a three-stream framework named as multiscale CNNs which could automatically detect the optimum top-three scales of the image sizes and combine information from different scales of the regions around that pixel. Datasets provided by Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized by MICCAI 2013 are utilized for both training and testing. The designed multiscale CNNs framework also combines multimodal features from T1, T1-enhanced, T2, and FLAIR MRI images. By comparison with traditional CNNs and the best two methods in BRATS 2012 and 2013, our framework shows advances in brain tumor segmentation accuracy and robustness.

No abstract

Brain tumors have become a leading cause of death around the globe. The main reason for this epidemic is the difficulty conducting a timely diagnosis of the tumor. Fortunately, magnetic resonance images (MRI) are utilized to diagnose tumors in most cases. The performance of a Convolutional Neural Network (CNN) depends on many factors (i.e., weight initialization, optimization, batches and epochs, learning rate, activation function, loss function, and network topology), data quality, and specific combinations of these model attributes. When we deal with a segmentation or classification problem, utilizing a single optimizer is considered weak testing or validity unless the decision of the selection of an optimizer is backed up by a strong argument. Therefore, optimizer selection processes are considered important to validate the usage of a single optimizer in order to attain these decision problems. In this paper, we provides a comprehensive comparative analysis of popular optimizers of CNN to benchmark the segmentation for improvement. In detail, we perform a comparative analysis of 10 different state-of-the-art gradient descent-based optimizers, namely Adaptive Gradient (Adagrad), Adaptive Delta (AdaDelta), Stochastic Gradient Descent (SGD), Adaptive Momentum (Adam), Cyclic Learning Rate (CLR), Adaptive Max Pooling (Adamax), Root Mean Square Propagation (RMS Prop), Nesterov Adaptive Momentum (Nadam), and Nesterov accelerated gradient (NAG) for CNN. The experiments were performed on the BraTS2015 data set. The Adam optimizer had the best accuracy of 99.2% in enhancing the CNN ability in classification and segmentation.

Bioluminescence tomography (BLT) poses a typical ill-posed inverse problem with a large number of unknowns and a relatively limited number of boundary measurements. It is indispensable to incorporate a priori information into the inverse problem formulation in order to obtain viable solutions. In the paper, Bayesian approach has been firstly suggested to incorporate multiple types of a priori information for BLT reconstruction. Meanwhile, a generalized adaptive Gaussian Markov random field (GAGMRF) prior model for unknown source density estimation is developed to further reduce the ill-posedness of BLT on the basis of finite element analysis. Then the distribution of bioluminescent source can be acquired by maximizing the log posterior probability with respect to a noise parameter and the unknown source density. Furthermore, the use of finite element method makes the algorithm appropriate for complex heterogeneous phantom. The algorithm was validated by numerical simulation of a 3-D micro-CT mouse atlas and physical phantom experiment. The reconstructed results suggest that we are able to achieve high computational efficiency and accurate localization of bioluminescent source.

No abstract