A Stacked Generalization of 3D Orthogonal Deep Learning Convolutional Neural Networks for Improved Detection of White Matter Hyperintensities in 3D FLAIR Images

BACKGROUND AND PURPOSE:Accurate and reliable detection of white matter hyperintensities and their volume quantification can provide valuable clinical information to assess neurologic disease progression. In this work, a stacked generalization ensemble of orthogonal 3D convolutional neural networks, StackGen-Net, is explored for improving automated detection of white matter hyperintensities in 3D T2-FLAIR images.MATERIALS AND METHODS:Individual convolutional neural networks in StackGen-Net were trained on 2.5D patches from orthogonal reformatting of 3D-FLAIR (n = 21) to yield white matter hyperintensity posteriors. A meta convolutional neural network was trained to learn the functional mapping from orthogonal white matter hyperintensity posteriors to the final white matter hyperintensity prediction. The impact of training data and architecture choices on white matter hyperintensity segmentation performance was systematically evaluated on a test cohort (n = 9). The segmentation performance of StackGen-Net was compared with state-of-the-art convolutional neural network techniques on an independent test cohort from the Alzheimer’s Disease Neuroimaging Initiative-3 (n = 20).RESULTS:StackGen-Net outperformed individual convolutional neural networks in the ensemble and their combination using averaging or majority voting. In a comparison with state-of-the-art white matter hyperintensity segmentation techniques, StackGen-Net achieved a significantly higher Dice score (0.76 [SD, 0.08], F1-lesion (0.74 [SD, 0.13]), and area under precision-recall curve (0.84 [SD, 0.09]), and the lowest absolute volume difference (13.3% [SD, 9.1%]). StackGen-Net performance in Dice scores (median = 0.74) did not significantly differ (P = .22) from interobserver (median = 0.73) variability between 2 experienced neuroradiologists. We found no significant difference (P = .15) in white matter hyperintensity lesion volumes from StackGen-Net predictions and ground truth annotations.CONCLUSIONS:A stacked generalization of convolutional neural networks, utilizing multiplanar lesion information using 2.5D spatial context, greatly improved the segmentation performance of StackGen-Net compared with traditional ensemble techniques and some state-of-the-art deep learning models for 3D-FLAIR.

White matter hyperintensities (WMHs) correspond to pathologic features of axonal degeneration, demyelination, and gliosis observed within cerebral white matter.¹ Clinically, the extent of WMHs in the brain has been associated with cognitive impairment, Alzheimer’s disease and vascular dementia, and increased risk of stroke.^2,3 The detection and quantification of WMH volumes to monitor lesion burden evolution and its correlation with clinical outcomes have been of interest in clinical research.^4,5 Although the extent of WMHs can be visually scored,⁶ the categoric nature of such scoring systems makes quantitative evaluation of disease progression difficult. Manually segmenting WMHs is tedious, prone to inter- and intraobserver variability, and is, in most cases, impractical. Thus, there is an increased interest in developing fast, accurate, and reliable computer-aided automated techniques for WMH segmentation.Convolutional neural network (CNN)-based approaches have been successful in several semantic segmentation tasks in medical imaging.⁷ Recent works have proposed using deep learning–based methods for segmenting WMHs using 2D-FLAIR images.^8-11 More recently, a WMH segmentation challenge¹² was also organized (http://wmh.isi.uu.nl/) to facilitate comparison of automated segmentation of WMHs of presumed vascular origin in 2D multislice T2-FLAIR images. Architectures that used an ensemble of separately trained CNNs showed promising results in this challenge, with 3 of the top 5 winners using ensemble-based techniques.¹²Conventional 2D-FLAIR images are typically acquired with thick slices (3–4 mm) and possible slice gaps. Partial volume effects from a thick slice are likely to affect the detection of smaller lesions, both in-plane and out-of-plane. 3D-FLAIR images, with isotropic resolution, have been shown to achieve higher resolution and contrast-to-noise ratio¹³ and have shown promising results in MS lesion detection using 3D CNNs.¹⁴ Additionally, the isotropic resolution enables viewing and evaluation of the images in multiple planes. This multiplanar reformatting of 3D-FLAIR without the use of interpolating kernels is only possible due to the isotropic nature of the acquisition. Network architectures that use information from the 3 orthogonal views have been explored in recent works for CNN-based segmentation of 3D MR imaging data.¹⁵ The use of data from multiple planes allows more spatial context during training without the computational burden associated with full 3D training.¹⁶ The use of 3 orthogonal views simultaneously mirrors how humans approach this segmentation task.Ensembles of CNNs have been shown to average away the variances in the solution and the choice of model- and configuration-specific behaviors of CNNs.¹⁷ Traditionally, the solutions from these separately trained CNNs are combined by averaging or using a majority consensus. In this work, we propose the use of a stacked generalization framework (StackGen-Net) for combining multiplanar lesion information from 3D CNN ensembles to improve the detection of WMH lesions in 3D-FLAIR. A stacked generalization¹⁸ framework learns to combine solutions from individual CNNs in the ensemble. We systematically evaluated the performance of this framework and compared it with traditional ensemble techniques, such as averaging or majority voting, and state-of-the-art deep learning techniques.     

Pathophysiology and management of sensitive skin: position paper from the special interest group on sensitive skin of the International Forum for the Study of Itch (IFSI)

The special interest group on sensitive skin of the International Forum for the Study of Itch previously defined sensitive skin as a syndrome defined by the occurrence of unpleasant sensations (stinging, burning, pain, pruritus and tingling sensations) in response to stimuli that normally should not provoke such sensations. This additional paper focuses on the pathophysiology and the management of sensitive skin. Sensitive skin is not an immunological disorder but is related to alterations of the skin nervous system. Skin barrier abnormalities are frequently associated, but there is no cause and direct relationship. Further studies are needed to better understand the pathophysiology of sensitive skin – as well as the inducing factors. Avoidance of possible triggering factors and the use of well-tolerated cosmetics, especially those containing inhibitors of unpleasant sensations, might be suggested for patients with sensitive skin. The role of psychosocial factors, such as stress or negative expectations, might be relevant for subgroups of patients. To date, there is no clinical trial supporting the use of topical or systemic drugs in sensitive skin. The published data are not sufficient to reach a consensus on sensitive skin management. In general, patients with sensitive skin require a personalized approach, taking into account various biomedical, neural and psychosocial factors affecting sensitive skin.     

Visumax飞秒激光辅助制瓣LASIK术中不透明气泡层及相关因素

目的：研究Visumax飞秒激光辅助的准分子激光原位角膜磨镶术（LASIK）术中不透明气泡层（OBL）产生的类型、相关影响因素及其临床意义。方法：回顾性分析。收集2016 年7 月1 日至8 月20 日在南京医科大学附属眼科医院行飞秒LASIK的患者154 例（300 眼）的临床资料，等效球镜度为（-5.88±1.51）D。飞秒激光采用Zeiss公司 Visumax飞秒激光仪，将术中产生的OBL分为瓣外OBL、瓣内快速OBL和瓣内慢速OBL。300 眼按角膜曲率平均K值分为A组（<42 D）、B组（42～46 D）和C组（>46 D）；按角膜瓣的厚度分为D组（100 μm）、E组（110 μm）和F组（120 μm）；按角膜厚度分为G组（<500 μm）、H组（500～540 μm）和I组（>540 μm）。统计不同角膜曲率、角膜瓣厚度、角膜厚度情况下3 种OBL产生的比例，并采用多元Logistic回归模型进行统计分析。结果：300 眼均发生OBL。以瓣外OBL作为参照，角膜瓣越薄，越容易出现瓣内快速OBL（r=-0.719，P=0.034）和瓣内慢速OBL（r=-0.875，P=0.044）。以瓣内慢速OBL作为参照，角膜曲率越高，越容易出现瓣内快速OBL（r=0.923，P=0.046）；角膜瓣越厚，越容易出现瓣外OBL（r=0.897，P=0.044）。结论：OBL的产生与角膜曲率、角膜瓣厚度、角膜厚度均存在一定的相关性，但瓣外OBL、瓣内快速OBL和瓣内慢速OBL对Visumax飞秒激光辅助的LASIK术后视觉质量的影响及远期疗效仍有待进一步观察研究。     

Loss of function,missense, and intronic variants in NOTCH1 confer different risks for left ventricular outflow tract obstructive heart defects in two European cohorts

Loss of function variants in NOTCH1 cause left ventricular outflow tract obstructive defects (LVOTO). However, the risk conferred by rare and noncoding variants in NOTCH1 for LVOTO remains largely uncharacterized. In a cohort of 49 families affected by hypoplastic left heart syndrome, a severe form of LVOTO, we discovered predicted loss of function NOTCH1 variants in 6% of individuals. Rare or low-frequency missense variants were found in 16% of families. To make a quantitative estimate of the genetic risk posed by variants in NOTCH1 for LVOTO, we studied associations of 400 coding and noncoding variants in NOTCH1 in 1,085 cases and 332,788 controls from the UK Biobank. Two rare intronic variants in strong linkage disequilibrium displayed significant association with risk for LVOTO amongst European-ancestry individuals. This result was replicated in an independent analysis of 210 cases and 68,762 controls of non-European and mixed ancestry. In conclusion, carrying rare predicted loss of function variants in NOTCH1 confer significant risk for LVOTO. In addition, the two intronic variants seem to be associated with an increased risk for these defects. Our approach demonstrates the utility of population-based data sets in quantifying the specific risk of individual variants for disease-related phenotypes.     

深圳市全人口结核病控制实施效果评价

目的　研究深圳市实施包含非户籍人口在内的全人口结核病控制模式。方法 在全人口中推行以发现和彻底治疗传染性肺结核病为主的现代结核病控制和监测策略。结果 从1993年至2001年,登记涂阳肺结核8483例,治愈率96.3%。肺结核疫情明显下降,深圳市户籍人口涂阳患病率下降55.9%,非户籍人口涂阳患病率下降38.6%。结论 实施全人口结核病控制是必要的、可行的和科学的。