Multiscale Bidirectional Input Convolutional and Deep Neural Network for Human Activity Recognition

In this paper, we proposed a multiscale and bidirectional input model based on convolutional neural network and deep neural network, named MBCDNN. In order to solve the problem of inconsistent activity segments, a multiscale input module is constructed to make up for the noise caused by filling. In order to solve the problem that single input is not enough to extract features from original data, we propose to manually design aggregation features combined with forward sequence and reverse sequence and use five cross-validation and stratified sampling to enhance the generalization ability of the model. According to the particularity of the task, we design an evaluation index combined with scene and action weight, which enriches the learning ability of the model to a great extent. In the 19 kinds of activity data based on scene+action, the accuracy and robustness are significantly improved, which is better than other mainstream traditional methods.

A reference catalog of DNA palindromes in the human genome and their variations in 1000 Genomes

A palindrome in DNA is like a palindrome in language, but when read backwards, it is a complement of the forward sequence; effectively, the two halves of a sequence complement each other from its midpoint like in a double strand of DNA. Palindromes are distributed throughout the human genome and play significant roles in gene expression and regulation. Palindromic mutations are linked to many human diseases, such as neuronal disorders, mental retardation, and various cancers. In this work, we computed and analyzed the palindromic sequences in the human genome and studied their conservation in personal genomes using 1000 Genomes data. We found that ~30% of the palindromes exhibit variation, some of which are caused by rare variants. The analysis of disease/trait-associated single-nucleotide polymorphisms in palindromic regions showed that disease-associated risk variants are 14 times more likely to be present in palindromic regions than in other regions. The catalog of palindromes in the reference genome and 1000 Genomes is being made available here with details on their variations in each individual genome to serve as a resource for future and retrospective whole-genome studies identifying statistically significant palindrome variations associated with diseases or traits and their roles in disease mechanisms.

Evaluating spatiotemporal integration of shape cues

ABSTRACTPrior work has shown that humans can successfully identify letters that are constructed with a sparse array of dots, wherein the dot pattern reflects the strokes that would normally be used to fashion a given letter. In the present work the dots were briefly displayed, one at a time in sequence, varying the spatial order in which they were shown. A forward sequence was spatially ordered as though one were passing a stroke across the dots to connect them. Experiments compared this baseline condition to the following three conditions: a) the dot sequence was spatially ordered, but in the reverse direction from how letter strokes might normally be written; b) the dots in each stroke of the letter were displayed in a random order; c) the sequence of displayed dots were chosen for display from any location in the letter. Significant differences were found between the baseline condition and all three of the comparison conditions, with letter recognition being far worse for the random conditions than for conditions that provided consistent spatial ordering of dot sequences. These findings show that spatial order is critical for integration of shape cues that have been sequentially displayed.

Reversal of interleg coupling in backward locomotion implies a prime role of the direction of locomotion.

Conventional descriptions of interleg coupling relate to anatomical definitions such as fore- or hindlegs, right or left legs (i.e. the body is the frame of reference). This convention is obvious for forward walking, where forelegs (in anatomical terms) are also the leading legs (in terms of direction). In backward locomotion, however, the leading legs in terms of direction are the hindlegs in terms of anatomy. What effects do the anatomy and direction of movement have on the sequence of stepping? Our observations on the locomotion of mole rats in a transparent acrylic tunnel revealed that, as in nature, mole rats moved both forwards and backwards. They typically employed a diagonal sequence of steps in forward walking, whereas in backward walking they typically employed a lateral sequence. However, when stepping was described with movement direction as the frame of reference, both forward and backward walking were made up of the same sequence of steps. The same invariant trend was recorded during backward galloping, but to a lesser extent than during walking. We suggest that the backward sequence is simply a reversal of the forward sequence: a hindleg during backward locomotion acts like a foreleg in forward locomotion, while a foreleg acts like a hindleg in forward locomotion. Interleg coupling therefore remains invariant in relation to the direction of locomotion.

Opponent Hues in Visual Masking

Temporal interactions among opponent and non-opponent hues were investigated in a visual masking paradigm in which both backward and forward temporal sequences were employed. Subjective confidence ratings rather than identification thresholds alone served as response indicators for masking sequences. Results indicate that in the backward sequence (test stimulus followed by masking stimulus) a greater masking effect occurred when the stimuli were of non-opponent hue pairs (red-yellow, red-blue, green-yellow, green-blue) than when compared with opponent hue pairs (red-green, yellow-blue). For the forward sequence (test stimulus preceded by masking stimulus) the masking effect was reduced when compared with the backward sequence. These findings appear to reflect the presumed temporal and spatial antagonistic qualities of opponent hue processes as postulated in the Hering model of color vision.