Automatic segmentation of cerebral white matter hyperintensities using only 3D FLAIR images

Magnetic Resonance (MR) white matter hyperintensities have been shown to predict an increased risk of developing cognitive decline. However, their actual role in the conversion to dementia is still not fully understood. Automatic segmentation methods can help in the screening and monitoring of Mild Cognitive Impairment patients who take part in large population-based studies. Most existing segmentation approaches use multimodal MR images. However, multiple acquisitions represent a limitation in terms of both patient comfort and computational complexity of the algorithms. In this work, we propose an automatic lesion segmentation method that uses only three-dimensional fluid-attenuation inversion recovery (FLAIR) images. We use a modified context-sensitive Gaussian mixture model to determine voxel class probabilities, followed by correction of FLAIR artifacts. We evaluate the method against the manual segmentation performed by an experienced neuroradiologist and compare the results with other unimodal segmentation approaches. Finally, we apply our method to the segmentation of multiple sclerosis lesions by using a publicly available benchmark dataset. Results show a similar performance to other state-of-the-art multimodal methods, as well as to the human rater.     

Magnetic resonance image reconstruction using trained geometric directions in 2D redundant wavelets domain and non-convex optimization

Reducing scanning time is significantly important for MRI. Compressed sensing has shown promising results by undersampling the k-space data to speed up imaging. Sparsity of an image plays an important role in compressed sensing MRI to reduce the image artifacts. Recently, the method of patch-based directional wavelets (PBDW) which trains geometric directions from undersampled data has been proposed. It has better performance in preserving image edges than conventional sparsifying transforms. However, obvious artifacts are presented in the smooth region when the data are highly undersampled. In addition, the original PBDW-based method does not hold obvious improvement for radial and fully 2D random sampling patterns. In this paper, the PBDW-based MRI reconstruction is improved from two aspects: 1) An efficient non-convex minimization algorithm is modified to enhance image quality; 2) PBDW are extended into shift-invariant discrete wavelet domain to enhance the ability of transform on sparsifying piecewise smooth image features. Numerical simulation results on vivo magnetic resonance images demonstrate that the proposed method outperforms the original PBDW in terms of removing artifacts and preserving edges.     

Nano‐Ruby: A Promising Fluorescent Probe for Background‐Free Cellular Imaging

Bioprobes based on fluorescent ruby nanoparticles, which are suitable for ultrasensitive imaging, are reported. A stable aqueous/buffer colloid, permitting facile conjugation to proteins, is produced by femtosecond laser ablation of ruby and the nanoparticles (mean size 17 nm) are photostable, with long lifetime (1–4 ms) 694 nm emission. With time‐gating complete (>20 dB) suppression of cell autofluorescence and suppression of exogenous fluorophores is observed. Nanoparticles are imaged in as‐grown cells and those immunolabeled with quantum dots. Immunoassay binding to target biomolecules is also demonstrated.     

PEGylated Luminescent Gold Nanoclusters: Synthesis,Characterization, Bioconjugation,and Application to One‐ and Two‐Photon Cellular Imaging

Biocompatible, near‐infrared luminescent gold nanoclusters (AuNCs) are synthesized directly in water using poly(ethylene glycol)‐dithiolane ligands terminating in either a carboxyl, amine, azide, or methoxy group. The ≈1.5 nm diameter AuNCs fluoresce at ≈820 nm with quantum yields that range from 4–8%, depending on the terminal functional group present, and display average luminescence lifetimes approaching 1.5 μs. The two‐photon absorption (TPA) cross‐section and two‐photon excited fluorescence (TPEF) properties are also measured. Long‐term testing shows the poly(ethylene glycol) stabilized AuNCs maintain colloidal stability in a variety of media ranging from saline to tissue culture growth medium along with tolerating storage of up to 2 years. DNA and dye‐conjugation reactions confirm that the carboxyl, amine, and azide groups can be utilized on the AuNCs for carbodiimide, succinimidyl ester, and Cu^I‐assisted cycloaddition chemistry, respectively. High signal‐to‐noise one‐ and two‐photon cellular imaging is demonstrated. The AuNCs exhibit outstanding photophysical stability during continuous‐extended imaging. Concomitant cellular viability testing shows that the AuNCs also elicit minimal cytotoxicity. Further biological applications for these luminescent nanoclustered materials are discussed.     

基于NIR高光谱成像技术的长枣虫眼无损检测

为了研究快速识别虫眼枣与正常枣的有效方法,利用特征波长主成分分析法结合波段比算法进行虫眼枣识别。首先,利用NIR高光谱成像系统采集130个长枣(50个正常、80个虫眼枣)图像,提取并分析不同类型长枣特征区域的平均光谱曲线,对970~1 670 nm范围内的光谱数据进行主成分分析,确定7个特征波长(990,1 028,1 109,1 160,1 231,1 285,1 464 nm)。然后,对长枣图像做主成分分析,选择PC2图像进行虫眼识别,虫眼与正常枣的识别率分别为67.5%、100%。为了进一步提高虫眼枣的识别率,采用波段比(R1231/R1109)对未识别的虫眼枣进行再次识别,识别率提高到90%。结果表明,基于NIR高光谱成像技术的检测方法对虫眼枣识别是可行的,同时也为多光谱成像技术应用于在线检测长枣品质提供了理论依据。     

Predicting future duration from present age: A critical assessment

Using a temporal version of the Copernican principle, Gott has proposed a statistical predictor of future longevity based on present age (Gott III, J. R., 1993, Nature, 363, 315) and applied the predictor to a variety of examples, including the longevity of the human species. Although Gott's proposal contains a grain of truth, it does not have the universal predictive power that he attributes to it.     

Gain Assisted Raman Induced Classical and Quantum Talbot Imaging

2D Raman generated classical and quantum Talbot imaging is proposed in a three-level gain assisted system. First, a 2D Raman induced grating (RIG) will be constructed by modulating transmission function in the weak probe channel using a strong control standing wave field. Furthermore, RIG will diffract a probe field, by shining a probing light beam on an optically generated lattice within a rubidium vapor cell. This study uses gain assisted Raman medium [Nature 406 , 277 (2000)] to examine classical and quantum Talbot imaging. In the case of Raman-induced classical imaging, the diffraction pattern repeats itself at planes with integer multiple Talbot lengths. Additionally, by taking into account entangled photon pairs, the scenario of Raman-induced quantum imaging is investigated. This study also looks at the RIG's amplitude and phase information with adjustable image size variation. As a result of the gain feature and zero absorption, this system is anticipated to be more suitable from the perspective of application. This analysis may pave the way for further research into the Talbot effect's nonlinear and quantum dynamical properties. Also, it provide a non-destructive, lensless method for imaging very cold atoms or molecules.     

Polarimetric-Spectral-Sensitive Compressed Ultrafast Photography Enables Real-Time Intensity Visualization of Time-Varying Polarized Ultrafast Laser Pulses

Ultrafast laser pulses with complex spatiotemporal and polarization couplings have shown great application prospects in coherent control, particle acceleration, molecular dynamics excitation, and chiral or anisotropic detection. It is very important to observe the polarization-coupled ultrafast laser pulses for characterizing the performance of laser source and understanding the laser-matter interaction, but there is currently no detection technique that can completely characterize them, including the spatio-, spectral-, and polarimetric-temporal information. In this work, a polarimetric-spectral-sensitive has been compressed ultrafast photography (PS-CUP) to realize the real-time intensity visualization of time-varying polarized (TP) ultrafast laser pulses. PS-CUP can not only distinguish the polarization state of a TP pulse, but also realize the spatio-, spectral- and polarimetric-temporal measurement. To demonstrate the ability of polarization resolving detection, PS-CUP is used to successfully distinguish between the sinistral and dextral TP pulses and measure the angle and degree of linear polarization of a dextral TP pulse, respectively, and the experimental results are highly consistent with theoretical simulations. Owing to the unprecedented multidimensional detection ability, PS-CUP is prospected to provide a powerful tool for the diagnosis of various complex ultrafast laser pulses and the detection of ultrafast chiral optics.     

三种降采样成像策略的性能优化以及与传统傅里叶望远镜的比较

为了获得成像质量较好且成像时间较少的新型傅里叶望远镜成像策略,本文比较了三种降采样成像策略(压缩感知方法 (CS)、低频全采样方法 (LF)和变密度随机采样方法 (VD))与传统傅里叶望远镜(FT)在图像质量和成像时间上的差异.分析方法如下:利用传统FT外场实验所获得的目标频谱数据作为基础,三种降采样方法 (LF, VD和CS)分别按照各自的采样模式和重构方法实现目标图像的重构;通过直观观察和Strehl比两种方法比较三种降采样方法与传统FT在图像质量上的差异;通过分析成像时间的组成要素,初步比较三种降采样方法与传统FT在成像时间上的差异.分析表明:1)压缩感知方法的图像质量优于其他两种降采样方法 (LF和VD),但略低于传统成像结果; 2)压缩感知方法在成像质量上略低于传统FT,但在成像时间上却明显小于传统FT; 3)分析中采用的外场数据均含噪声,这说明上述三种降采样重构过程对噪声有较好的鲁棒性.综合上述分析结果可以看出,基于压缩感知的傅里叶望远镜(CS-FT)是在实际含噪情况下可大幅减少成像时间的优良成像策略.     

鬼成像中一些数学问题

鬼成像是一种与传统成像方式不同的通过光场涨落的高阶关联获得图像信息的新型成像方式。近年来,相比传统成像方式,鬼成像所拥有的一些优点如高灵敏度、超分辨能力、抗散射等,使其在遥感、多光谱成像、热X射线衍射成像等领域得到广泛研究。随着对鬼成像的广泛研究,数学理论和方法在其中发挥的作用愈显突出。例如,基于压缩感知理论,可以进行鬼成像系统采样方式优化、图像重构算法设计及图像重构质量分析等研究工作。本文旨在探索鬼成像中的一些有趣的数学问题,主要包括:系统预处理方法、光场优化及相位恢复问题。对这些问题的研究既可以丰富鬼成像理论,又能推动它在实际应用中的发展。