学科服务 更多>>

关键词： Feature screening   Missing at random   Missing covariate   Pearson Chi-Square statistic   Sure screening property  

摘要： Most existing feature screening methods assume that data are fully observed. It is quite a challenge to develop screening methods for incomplete data since the traditional missing data analysis techniques cannot be directly applied to ultrahigh dimensional case. A two-step model-free feature screening procedure for ultrahigh dimensional categorical data when some covariate values are missing at random is developed. For each covariate with missing data, the first step screens out the variables in the unspecified propensity function. In the second step, screening statistics such as the adjusted Pearson Chi-Square statistics can be calculated by leveraging the variables obtained in the first step and the special structure of categorical data. Sure screening properties are established for the proposed method. Finite sample performance is investigated by simulation studies and a real data example. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Milan system for reporting cytopathology   salivary gland   atypia of undetermined significance   cytology   fine-needle aspiration   risk of malignancy   salivary gland  

摘要： Background Many prior institutional and multi-institutional studies have applied the Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) retrospectively to their specimens to determine the risk of malignancy (ROM) of each category. Most of these studies focused on general assessment of the system and risk classification. However, there seems to be less focus on the category of atypia of undetermined significance (AUS) that could be attributed to the low number of cases that could fit into this category. Herein, we present a bi-institutional experience with this category. Methods A computerized search of the databases was performed to identify all salivary gland fine-needle aspiration (FNA) in two institutions over a period of 12 years. The final diagnosis of each case was reclassified based on MSRSGC, and histology follow-up was retrieved. Results Sixty AUS cases (out of 1560 salivary gland FNA) were identified with a rate of 3.8%. Forty cases (66%) had a subsequent tissue material. Correlation with histology revealed that the estimated ROM is 37.5% (15/40) and the overall ROM is 25% (15/60). Fifty percent of the cases had a prominent lymphoid component and most commonly represented lymphomas, reactive lymph node or sialadenitis. Conclusion The AUS category is a heterogeneous group of lesions with predominant lymphoid-rich entities. Some variability exists between institutions with most having higher ROM than the suggested 20% by the MSRSGC atlas.

关键词： Three-dimensional displays   Geometry   Shape   Solid modeling   Estimation   Image reconstruction   Training   Monocular pose estimation   monocular depth estimation   point-cloud estimation   geometry reconstruction  

摘要： We propose an unsupervised method to learn the 3D geometry of object categories by looking around them. Differently from traditional approaches, this method does not require CAD models or manual supervision. Instead, using only video sequences showing object instances from a moving viewpoint, the method learns a deep neural network that can predict several aspects of the 3D geometry of such objects from single images. The network has three components. The first is a Siamese viewpoint factorization network that robustly aligns the input videos and learns to predict the absolute viewpoint of the object from a single image. The second is a depth estimation network that performs monocular depth prediction. The third is a shape completion network that predicts the full 3D shape of the object from the output of the monocular depth prediction module. While the three modules solve very different task, we show that they all benefit significantly from allowing networks to perform probabilistic predictions. This results in a self-assessment mechanism which is crucial for obtaining high quality predictions. Our network achieves state-of-the-art results on viewpoint prediction, depth estimation, and 3D point cloud estimation on public benchmarks.

关键词： Uncanny valley   Human-robot interaction   Psychology   Social interaction   Categorical perception  

摘要： Android robots that resemble humans closely, but not perfectly, can provoke negative feelings of dislike and eeriness in humans (the "Uncanny Valley" effect). We investigated whether category confusion between the perceptual categories of "robot" and "human" contributes to Uncanny Valley aversion. Using a novel, validated corpus of 182 images of real robot and human faces, we precisely estimated the shape of the Uncanny Valley and the location of the perceived robot/human boundary. To implicitly measure confusion, we tracked 358 participants' mouse trajectories as they categorized the faces. We observed a clear Uncanny Valley, though with some interesting differences from standard theoretical predictions;the initial apex of likability for highly mechanical robots indicated that these robots were still moderately dislikable, and the Uncanny Valley itself was positioned closer to the mechanical than to the human-like end of the spectrum. We also observed a pattern of categorization suggesting that humans do perceive a categorical robot/human boundary. Yet in contrast to predictions of the category confusion mechanism hypothesis, the locations of the Uncanny Valley and of the category boundary did not coincide, and mediation analyses further failed to support a mechanistic role of category confusion. These results suggest category confusion does not explain the Uncanny Valley effect.

关键词： Industrial land use intensity   Location quotient   Dominant industries  

摘要： Facing with scarce urban land resources, quantitative evaluating the industrial land use intensity is crucial to sustainable urban land use. In this study, ten cities with different development levels in China, Zhenjiang, Weinan, Guangzhou, Tonghua, Tongliao, Liaocheng, Taiyuan, Puyang, Zhuzhou, were chosen as the study cases. The industrial land use intensity (ILUI) of these cities were assessed by a comprehensive evaluation framework, and dominant industries were recognized by the method of Location Quotient. The results showed that a conspicuous unbalanced ILUI index was observed in these cities, with the highest index in Guangzhou and the lowest in Liaocheng. Moreover, the total scale, internal structure, layout and economic intensity of industrial land did not coordinate with each other, which indicated that inner-city land use feature varies. In addition, the Water conservancy, environment and public facilities management industry, the Information transmission, software and information technology services industry and the Renting and Commercial Services industry were recognized as the common dominant industries among these cities. This study could provide an insight of the gap between current industrial land use pattern and balanced intensive industrial land use pattern, to use the urban land more sustainable in the future.

关键词： T1   Pancreas head cancer   AJCC   Stage  

摘要： Background The new T1 pancreatic cancer by the eighth edition of the AJCC staging system discards the concept of "extension beyond the pancreas" and focuses on size only. Furthermore, the new T1 is divided into T1a, T1b, and T1c based on size. The evidence pertaining to these changes has not been evaluated. This is to evaluate the feasibility of the new T1 definition in the pancreas head cancer cohort. Methods Data from 540 patients with T1 pancreatic ductal adenocarcinoma as defined by the eighth edition were collected from Korea, Japan, and the USA. Invasive IPMNs were excluded. Survival analyses were performed. Results Of the 540 patients, 181 patients were T1 according to the seventh edition and 359 were down-staged to T1 from the former T3 because the concept of "extension beyond the pancreas" was discarded. The 5-year survival rate and the median survival of T1 patients were 30.6% and 27 months, respectively. Comparing tumors that extend beyond the pancreas (new T1) and those confined within the pancreas (original T1), the latter showed significantly longer median survival (43 vs. 24 months, p < 0.001). In terms of T1a/b/c, there were no significant differences in survival. Using MaxStat, subdividing into two groups using 1.1 cm as the cut-off value, yielded significantly discrete prognostic groups (p < 0.001). Conclusion The new T1 definition may be more practical, but the implications of the concept of "extension beyond the pancreas" should be re-investigated. Further, the subcategorization of T1a/b/c may not be adequate and may require revision or deletion.

关键词： Community of inquiry   Learning communities   Construct validity   Improving classroom teaching   Distance education and online learning  

摘要： In online or blended environments, the Community of Inquiry (CoI) framework sets that a meaningful educational experience derives from the interrelation of teaching, social and cognitive presences. Each presence is subdivided, resulting in a structure in ten categories at the basis for the CoI survey instrument. Although the survey structure in three presences was repeatedly validated in the literature, the categories within presences were not and were consequently investigated in this study. High internal consistencies between items of each category demonstrated that the structure is reliable. Its convergent and discriminant validity were confirmed using multi-group confirmatory factor analyses, that further allowed to reinforce the construct validation by evaluating its factorial invariance across independent samples collected in two universities and varying in gender and age (n(1) = 343;n(2) =420). To assess its discriminant validity and because high estimated correlations between categories were observed, alternative structures in eight, seven and six factors were also compared to the original ten-category structure. They were all valid despite yielding inferior fits. The partial structural invariance of CoI structures in categories were also confirmed across groups. Next, a CoI structure in three presences resulted in an insufficient fit to data across independent groups. Much more conclusively, a second-order structure including both presences and categories demonstrated a very good fit to the data, highlighting the importance of categories to reflect students' perceptions. This paper, although presented at a conceptual level, enlightens the potential of studying the influence of categories on each other, learning outcomes, or to identify areas of improvement in online and blended courses by relying on meaningful and trustful categories that further characterize the well-known presences.

关键词： Acoustics  

摘要： Recalibration is a learning process in which perceptual boundaries between speech-sounds adjust through exposure to a supplementary source of information. Using a dichotic-listening methodology, the experiments reported here establish interaural recalibration-in which an ambiguous speech sound in one ear is recalibrated on the basis of a clear sound presented to the other ear. This demonstrates a previously unknown form of recalibration and shows that location-specific recalibration occurs even when people are unaware of location differences between the sounds involved.

关键词： categorization   EEG   fMRI   human   occipital cortex  

摘要： Categorization allows organisms to generalize existing knowledge to novel stimuli and to discriminate between physically similar yet conceptually different stimuli. Humans, nonhuman primates, and rodents can readily learn arbitrary categories defined by low-level visual features, and learning distorts perceptual sensitivity for category-defining features such that differences between physically similar yet categorically distinct exemplars are enhanced, whereas differences between equally similar but categorically identical stimuli are reduced. We report a possible basis for these distortions in human occipitoparietal cortex. In three experiments, we used an inverted encoding model to recover population-level representations of stimuli from multivoxel and multielectrode patterns of human brain activity while human participants (both sexes) classified continuous stimulus sets into discrete groups. In each experiment, reconstructed representations of to-be-categorized stimuli were systematically biased toward the center of the appropriate category. These biases were largest for exemplars near a category boundary, predicted participants' overt category judgments, emerged shortly after stimulus onset, and could not be explained by mechanisms of response selection or motor preparation. Collectively, our findings suggest that category learning can influence processing at the earliest stages of cortical visual processing.