High activity and high functional connectivity are mutually exclusive in resting state zebrafish and human brains

Active neurons impact cell types with which they are functionally connected. Both activity and functional connectivity are heterogeneous across the brain, but the nature of their relationship is not known. Here we employ brain-wide calcium imaging at cellular resolution in larval zebrafish to record spontaneous activity of >12,000 neurons in the forebrain. By classifying their activity and functional connectivity into three levels (high, medium, low), we find that highly active neurons have low functional connections and highly connected neurons are of low activity. Intriguingly, deploying the same analytical methods on functional magnetic resonance imaging (fMRI) data from the resting state human brain, we uncover a similar relationship between activity and functional connectivity, that is, regions of high activity are non-overlapping with those of high connectivity. These findings reveal a previously unknown and evolutionarily conserved brain organizational principle that have implications for understanding disease states and designing artificial neuronal networks.

Optimized Identification of High-Grade Prostate Cancer by Combining Different PSA Molecular Forms and PSA Density in a Deep Learning Model

After skin cancer, prostate cancer (PC) is the most common cancer among men. The gold standard for PC diagnosis is based on the PSA (prostate-specific antigen) test. Based on this preliminary screening, the physician decides whether to proceed with further tests, typically prostate biopsy, to confirm cancer and evaluate its aggressiveness. Nevertheless, the specificity of the PSA test is suboptimal and, as a result, about 75% of men who undergo a prostate biopsy do not have cancer even if they have elevated PSA levels. Overdiagnosis leads to unnecessary overtreatment of prostate cancer with undesirable side effects, such as incontinence, erectile dysfunction, infections, and pain. Here, we used artificial neuronal networks to develop models that can diagnose PC efficiently. The model receives as an input a panel of 4 clinical variables (total PSA, free PSA, p2PSA, and PSA density) plus age. The output of the model is an estimate of the Gleason score of the patient. After training on a dataset of 190 samples and optimization of the variables, the model achieved values of sensitivity as high as 86% and 89% specificity. The efficiency of the method can be improved even further by training the model on larger datasets.

Bio-instantiated recurrent neural networks

Biological neuronal networks (BNNs) constitute a niche for inspiration and analogy making for researchers that focus on artificial neuronal networks (ANNs). Moreover, neuroscientists increasingly use ANNs as a model for the brain. However, apart from certain similarities and analogies that can be drawn between ANNs and BNNs, such networks exhibit marked differences, specifically with respect to their network topology. Here, we investigate to what extent network topology found in nature can lead to beneficial aspects in recurrent neural networks (RNNs): i) the prediction performance itself, that is, the capacity of the network to minimize the desired function at hand in test data and ii) speed of training, that is, how fast during training the network reaches its optimal performance. To this end, we examine different ways to construct RNNs that instantiate the network topology of brains of different species. We refer to such RNNs as bio-instantiated. We examine the bio-instantiated RNNs in the context of a key cognitive capacity, that is, working memory, defined as the ability to track task-relevant information as a sequence of events unfolds in time. We highlight what strategies can be used to construct RNNs with the network topology found in nature, without sacrificing prediction capacity and speed of training. Despite that we observe no enhancement of performance when compared to randomly wired RNNs, our approach demonstrates how empirical neural network data can be used for constructing RNNs, thus, facilitating further experimentation with biologically realistic networks topology.

Application of artificial neuronal networks in extracting parameters of solar cells

This paper presents a new neural network-based approach that aims to use the back propagation multilayer perceptual (MLP) propagation algorithm to improve the extraction of parameters from a solar cell based on the single-diode model from the experimentally measured characteristic I(V). The I(V) current function as a model function for the neural network, is used the Lambert function W and I can be expressed as an explicit function. The main five parameters of interest of the function I(V) are the photocurrent, Iph, the saturation current in inverse diode, I0, the ideality factor of the diode, n, the resistance in series, RS and shunt resistance, RSh. We have used the Matlab to find the five parameters of the cell. To verify the proposed approach, we chose two different solar cells made of mono- and polycrystalline silicon. The comparison between the measured values and the results of the proposed model shows great precision. Finally, the values found of the five parameters by our approach are compared with those found by the method of least squares (MLS).