ABSTRACT
Objective. Low-count positron emission tomography (PET) imaging is an efficient way to promote more widespread use of PET because of its short scan time and low injected activity. However, this often leads to low-quality PET images with clinical image reconstruction, due to high noise and blurring effects. Existing PET image restoration (IR) methods hinder their own restoration performance due to the semi-convergence property and the lack of suitable denoiser prior.Approach. To overcome these limitations, we propose a novel deep plug-and-play IR method called Deep denoiser Prior driven Relaxed Iterated Tikhonov method (DP-RI-Tikhonov). Specifically, we train a deep convolutional neural network denoiser to generate a flexible deep denoiser prior to handle high noise. Then, we plug the deep denoiser prior as a modular part into a novel iterative optimization algorithm to handle blurring effects and propose an adaptive parameter selection strategy for the iterative optimization algorithm.Main results. Simulation results show that the deep denoiser prior plays the role of reducing noise intensity, while the novel iterative optimization algorithm and adaptive parameter selection strategy can effectively eliminate the semi-convergence property. They enable DP-RI-Tikhonov to achieve an average quantitative result (normalized root mean square error, structural similarity) of (0.1364, 0.9574) at the stopping iteration, outperforming a conventional PET IR method with an average quantitative result of (0.1533, 0.9523) and a state-of-the-art deep plug-and-play IR method with an average quantitative result of (0.1404, 0.9554). Moreover, the advantage of DP-RI-Tikhonov becomes more obvious at the last iteration. Experiments on six clinical whole-body PET images further indicate that DP-RI-Tikhonov successfully reduces noise intensity and recovers fine details, recovering sharper and more uniform images than the comparison methods.Significance. DP-RI-Tikhonov's ability to reduce noise intensity and effectively eliminate the semi-convergence property overcomes the limitations of existing methods. This advancement may have substantial implications for other medical IR.
Subject(s)
Image Processing, Computer-Assisted , Positron-Emission Tomography , Signal-To-Noise Ratio , Positron-Emission Tomography/methods , Image Processing, Computer-Assisted/methods , Humans , Deep Learning , Phantoms, ImagingABSTRACT
Attention deficit/hyperactivity disorder(ADHD) is a behavioral disorder syndrome found mainly in school-age population. At present, the diagnosis of ADHD mainly depends on the subjective methods, leading to the high rate of misdiagnosis and missed-diagnosis. To solve these problems, we proposed an algorithm for classifying ADHD objectively based on convolutional neural network. At first, preprocessing steps, including skull stripping, Gaussian kernel smoothing, et al., were applied to brain magnetic resonance imaging(MRI). Then, coarse segmentation was used for selecting the right caudate nucleus, left precuneus, and left superior frontal gyrus region. Finally, a 3 level convolutional neural network was used for classification. Experimental results showed that the proposed algorithm was capable of classifying ADHD and normal groups effectively, the classification accuracies obtained by the right caudate nucleus and the left precuneus brain regions were greater than the highest classification accuracy(62.52%) in the ADHD-200 competition, and among 3 brain regions in ADHD and the normal groups, the classification accuracy from the right caudate nucleus was the highest. It is well concluded that the method for classification of ADHD and normal groups proposed in this paper utilizing the coarse segmentation and deep learning is a useful method for the purpose. The classification accuracy of the proposed method is high, and the calculation is simple. And the method is able to extract the unobvious image features better, and can overcome the shortcomings of traditional methods of MRI brain area segmentation, which are time-consuming and highly complicate. The method provides an objective diagnosis approach for ADHD.
Subject(s)
Attention Deficit Disorder with Hyperactivity , Attention Deficit Disorder with Hyperactivity/diagnosis , Brain , Caudate Nucleus , Humans , Magnetic Resonance Imaging/methods , Nerve Net , Prefrontal CortexABSTRACT
OBJECTIVE: To study the radiobiological effects of fast neutron/photon mixed irradiation on human cancer cell in vitro and to discuss the mechanism in relation with cell cycle and apoptosis, thus to provide experimental support for the further application of fast neutron radiotherapy of cancer. METHODS: Exponentially growing human nasopharyngeal cancer cell line CNE-1 was irradiated in vitro with 35 MeV p-->Be fast neutron and 6 MV-X ray in grading doses (0 cGy, 40 cGy, 80 cGy, 120 cGy, 160 cGy, 240 cGy, 320 cGy and 400 cGy for neutron, and 0 cGy, 100 cGy, 200 cGy, 300 cGy, 400 cGy, 600 cGy, 800 cGy and 1000 cGy for X ray). Clonogenic assay was performed, and relative biological effectiveness (RBE) of fast neutron was determined with D(10) by means of cell survival curves. Isoeffective doses of 35 MeV p-->Be fast neutron and 6 MV-X ray were obtained according to the RBE. The cells were assigned into two irradiation regimens, (1) the one-week-fractionation regimen, which adopted the radiation pattern of X x 5, N x 2 and X-N-X-X-N. After irradiation the clonogenic assay was performed to compare their survival fractions; (2) the two-dose regimen, with the radiation pattern of X + N, N + X and X + X. Flow cytometry was done at different time points after irradiation to analyze cell cycle distribution and apoptosis. Fast neutron dose was delivered on Tuesday and Friday, and all the other irradiation intervals were 24 h. RESULTS: The RBE of fast neutron to X ray in CNE-1 cells according to the D(10) ratio was 2.40. The neutron isoeffective dose for a single dose of 200 cGy of 6 MV-X ray was approximately 80 cGy. In clonogenic assay, the cell survival fractions were significantly lower in X-N-X-X-N group (0.0079) than those in X x 5 (0.018) and N x 2 (0.017) groups. The flow cytometry suggested a higher percentage of apoptotic cells after mixed irradiation, and different sequence of X ray and neutron irradiations caused varying changes in cell cycle arrest. CONCLUSION: Mixed irradiation of fast neutron and X ray showed a synergic effect in vitro on CNE-1 cell killing. Cell cycle arrest and apoptosis may play some role in the radiation damage repair mechanisms of mixed beam irradiation.