Search for: All records

Abstract Elephants have long been observed to grip objects with their trunk, but little is known about how they adjust their strategy for different weights. In this study, we challenge a female African elephant at Zoo Atlanta to lift 20–60 kg barbell weights with only its trunk. We measure the trunk’s shape and wrinkle geometry from a frozen elephant trunk at the Smithsonian. We observe several strategies employed to accommodate heavier weights, including accelerating less, orienting the trunk vertically, and wrapping the barbell with a greater trunk length. Mathematical models show that increasing barbell weights are associated with constant trunk tensile force and an increasing barbell-wrapping surface area due to the trunk’s wrinkles. Our findings may inspire the design of more adaptable soft robotic grippers that can improve grip using surface morphology such as wrinkles. 

Despite having a trunk that weighs over 100 kg, elephants mainly feed on lightweight vegetation. How do elephants manipulate such small items? In this experimental and theoretical investigation, we filmed elephants at Zoo Atlanta showing that they can use suction to grab food, performing a behaviour that was previously thought to be restricted to fishes. We use a mathematical model to show that an elephant’s nostril size and lung capacity enables them to grab items using comparable pressures as the human lung. Ultrasonographic imaging of the elephant sucking viscous fluids show that the elephant’s nostrils dilate up to 30 % in radius, which increases the nasal volume by 64 % . Based on the pressures applied, we estimate that the elephants can inhale at speeds of over 150 m s −1 , nearly 30 times the speed of a human sneeze. These high air speeds enable the elephant to vacuum up piles of rutabaga cubes as well as fragile tortilla chips. We hope these findings inspire further work in suction-based manipulation in both animals and robots. 

Fraud detection is of great importance because fraudulent behaviors may mislead consumers or bring huge losses to enterprises. Due to the lockstep feature of fraudulent behaviors, fraud detection problem can be viewed as finding suspicious dense blocks in the attributed bipartite graph. In reality, existing attribute-based methods are not adversarially robust, because fraudsters can take some camouflage actions to cover their behavior attributes as normal. More importantly, existing structural information based methods only consider shallow topology structure, making their effectiveness sensitive to the density of suspicious blocks. In this paper, we propose a novel deep structure learning model named DeepFD to differentiate normal users and suspicious users. DeepFD can preserve the non-linear graph structure and user behavior information simultaneously. Experimental results on different types of datasets demonstrate that DeepFD outperforms the state-of-the-art baselines. 

Abstract Non-small-cell lung cancer (NSCLC) represents approximately 80–85% of lung cancer diagnoses and is the leading cause of cancer-related death worldwide. Recent studies indicate that image-based radiomics features from positron emission tomography/computed tomography (PET/CT) images have predictive power for NSCLC outcomes. To this end, easily calculated functional features such as the maximum and the mean of standard uptake value (SUV) and total lesion glycolysis (TLG) are most commonly used for NSCLC prognostication, but their prognostic value remains controversial. Meanwhile, convolutional neural networks (CNN) are rapidly emerging as a new method for cancer image analysis, with significantly enhanced predictive power compared to hand-crafted radiomics features. Here we show that CNNs trained to perform the tumor segmentation task, with no other information than physician contours, identify a rich set of survival-related image features with remarkable prognostic value. In a retrospective study on pre-treatment PET-CT images of 96 NSCLC patients before stereotactic-body radiotherapy (SBRT), we found that the CNN segmentation algorithm (U-Net) trained for tumor segmentation in PET and CT images, contained features having strong correlation with 2- and 5-year overall and disease-specific survivals. The U-Net algorithm has not seen any other clinical information (e.g. survival, age, smoking history, etc.) than the images and the corresponding tumor contours provided by physicians. In addition, we observed the same trend by validating the U-Net features against an extramural data set provided by Stanford Cancer Institute. Furthermore, through visualization of the U-Net, we also found convincing evidence that the regions of metastasis and recurrence appear to match with the regions where the U-Net features identified patterns that predicted higher likelihoods of death. We anticipate our findings will be a starting point for more sophisticated non-intrusive patient specific cancer prognosis determination. For example, the deep learned PET/CT features can not only predict survival but also visualize high-risk regions within or adjacent to the primary tumor and hence potentially impact therapeutic outcomes by optimal selection of therapeutic strategy or first-line therapy adjustment. 

Abstract In this work, a novel version of macrocyclic arenes, namely leaning pillar[6]arenes, was discovered and it can be considered as a tilted version of a pillar[6]arene with two hydroxy/alkoxy functionalities removed. Through a facile two‐step synthetic approaches, in conjunction with a diversity of post‐modification possibilities, a series of leaning pillar[6]arenes, with good cavity adaptability and enhanced guest‐binding capability, was synthesized, and their self‐assembly in single‐crystal states is presented. DFT calculations demonstrated that the lower rotational barrier of unsubstituted phenylene rings, the uneven electron density centered at the leaning phenyl rings, and the polarization effect along the edge generated by the hydrogen‐bond‐induced orientation of hydroxy groups greatly affected the host‐guest properties, and meanwhile provided an intuitive explanation for the pillar‐like and rigid structure of traditional pillar[6]arenes. Significantly, the crystal structure of cyclo‐oligomeric quinone was obtained by direct oxidation of leaning pillar[6]arenes.