Lessons From Deep Neural Networks for Studying the Coding Principles of Biological Neural Networks

Covariance properties under natural image transformations for the generalised Gaussian derivative model for visual receptive fields

The property of covariance, also referred to as equivariance, means that an image operator is well-behaved under image transformations, in the sense that the result of applying the image operator to a transformed input image gives essentially a similar result as applying the same image transformation to the output of applying the image operator to the original image. This paper presents a theory of geometric covariance properties in vision, developed for a generalised Gaussian derivative model of receptive fields in the primary visual cortex and the lateral geniculate nucleus, which, in turn, enable geometric invariance properties at higher levels in the visual hierarchy. It is shown how the studied generalised Gaussian derivative model for visual receptive fields obeys true covariance properties under spatial scaling transformations, spatial affine transformations, Galilean transformations and temporal scaling transformations. These covariance properties imply that a vision system, based on image and video measurements in terms of the receptive fields according to the generalised Gaussian derivative model, can, to first order of approximation, handle the image and video deformations between multiple views of objects delimited by smooth surfaces, as well as between multiple views of spatio-temporal events, under varying relative motions between the objects and events in the world and the observer. We conclude by describing implications of the presented theory for biological vision, regarding connections between the variabilities of the shapes of biological visual receptive fields and the variabilities of spatial and spatio-temporal image structures under natural image transformations. Specifically, we formulate experimentally testable biological hypotheses as well as needs for measuring population statistics of receptive field characteristics, originating from predictions from the presented theory, concerning the extent to which the shapes of the biological receptive fields in the primary visual cortex span the variabilities of spatial and spatio-temporal image structures induced by natural image transformations, based on geometric covariance properties.

Label-free macrophage phenotype classification using machine learning methods

Macrophages are heterogeneous innate immune cells that are functionally shaped by their surrounding microenvironment. Diverse macrophage populations have multifaceted differences related to their morphology, metabolism, expressed markers, and functions, where the identification of the different phenotypes is of an utmost importance in modelling immune response. While expressed markers are the most used signature to classify phenotypes, multiple reports indicate that macrophage morphology and autofluorescence are also valuable clues that can be used in the identification process. In this work, we investigated macrophage autofluorescence as a distinct feature for classifying six different macrophage phenotypes, namely: M0, M1, M2a, M2b, M2c, and M2d. The identification was based on extracted signals from multi-channel/multi-wavelength flow cytometer. To achieve the identification, we constructed a dataset containing 152,438 cell events each having a response vector of 45 optical signals fingerprint. Based on this dataset, we applied different supervised machine learning methods to detect phenotype specific fingerprint from the response vector, where the fully connected neural network architecture provided the highest classification accuracy of 75.8% for the six phenotypes compared simultaneously. Furthermore, by restricting the number of phenotypes in the experiment, the proposed framework produces higher classification accuracies, averaging 92.0%, 91.9%, 84.2%, and 80.4% for a pool of two, three, four, five phenotypes, respectively. These results indicate the potential of the intrinsic autofluorescence for classifying macrophage phenotypes, with the proposed method being quick, simple, and cost-effective way to accelerate the discovery of macrophage phenotypical diversity.

Natural and Artificial Intelligence: A brief introduction to the interplay between AI and neuroscience research

Neuroscience and artificial intelligence (AI) share a long history of collaboration. Advances in neuroscience, alongside huge leaps in computer processing power over the last few decades, have given rise to a new generation of in silico neural networks inspired by the architecture of the brain. These AI systems are now capable of many of the advanced perceptual and cognitive abilities of biological systems, including object recognition and decision making. Moreover, AI is now increasingly being employed as a tool for neuroscience research and is transforming our understanding of brain functions. In particular, deep learning has been used to model how convolutional layers and recurrent connections in the brain's cerebral cortex control important functions, including visual processing, memory, and motor control. Excitingly, the use of neuroscience-inspired AI also holds great promise for understanding how changes in brain networks result in psychopathologies, and could even be utilized in treatment regimes. Here we discuss recent advancements in four areas in which the relationship between neuroscience and AI has led to major advancements in the field; (1) AI models of working memory, (2) AI visual processing, (3) AI analysis of big neuroscience datasets, and (4) computational psychiatry.