Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems—patient classification, fundamental biological processes and treatment of patients—and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

3D snapshot microscopy enables fast volumetric imaging by capturing a 3D volume in a single 2D camera image and performing computational reconstruction. Fast volumetric imaging has a variety of biological applications such as whole brain imaging of rapid neural activity in larval zebrafish. The optimal microscope design for this optical 3D-to-2D encoding is both sample- and task-dependent, with no general solution known. Deep learning based decoders can be combined with a differentiable simulation of an optical encoder for end-to-end optimization of both the deep learning decoder and optical encoder. This technique has been used to engineer local optical encoders for other problems such as depth estimation, 3D particle localization, and lensless photography. However, 3D snapshot microscopy is known to require a highly non-local optical encoder which existing UNet-based decoders are not able to engineer. We show that a neural network architecture based on global kernel Fourier convolutional neural networks can efficiently decode information from multiple depths in a volume, globally encoded across a 3D snapshot image. We show in simulation that our proposed networks succeed in engineering and reconstructing optical encoders for 3D snapshot microscopy where the existing state-of-the-art UNet architecture fails. We also show that our networks outperform the state-of-the-art learned reconstruction algorithms for a computational photography dataset collected on a prototype lensless camera which also uses a highly non-local optical encoding.

Localisation microscopy often relies on detailed models of point-spread functions. For applications such as deconvolution or PSF engineering, accurate models for light propagation in imaging systems with a high numerical aperture are required. Different models have been proposed based on 2D Fourier transforms or 1D Bessel integrals. The most precise ones combine a vectorial description of the electric field and accurate aberration models. However, it may be unclear which model to choose as there is no comprehensive comparison between the Fourier and Bessel approaches yet. Moreover, many existing libraries are written in Java (e.g., our previous PSF generator software) or MATLAB, which hinders their integration into deep learning algorithms. In this work, we start from the original Richards-Wolf integral and revisit both approaches in a systematic way. We present a unifying framework in which we prove the equivalence between the Fourier and Bessel strategies and detail a variety of correction factors applicable to both of them. Then, we provide a high-performance implementation of our theoretical framework in the form of an open-source library that is built on top of PyTorch, a popular library for deep learning. It enables us to benchmark the accuracy and computational speed of different models and allows for an in-depth comparison of the existing models for the first time. We show that the Bessel strategy is optimal for axisymmetric beams, while the Fourier approach can be applied to more general scenarios. Our work enables the efficient computation of a point-spread function on CPU or GPU, which can then be included in simulation and optimisation pipelines.

Animals are often bombarded with visual information and must prioritize specific visual features based on their current needs. The neuronal circuits that detect and relay visual features have been well studied. Much less is known about how an animal adjusts its visual attention as its goals or environmental conditions change. During social behaviours, flies need to focus on nearby flies. Here we study how the flow of visual information is altered when female Drosophila enter an aggressive state. From the connectome, we identify three state-dependent circuit motifs poised to modify the response of an aggressive female to fly-sized visual objects: convergence of excitatory inputs from neurons conveying select visual features and internal state; dendritic disinhibition of select visual feature detectors; and a switch that toggles between two visual feature detectors. Using cell-type-specific genetic tools, together with behavioural and neurophysiological analyses, we show that each of these circuit motifs is used during female aggression. We reveal that features of this same switch operate in male Drosophila during courtship pursuit, suggesting that disparate social behaviours may share circuit mechanisms. Our study provides a compelling example of using the connectome to infer circuit mechanisms that underlie dynamic processing of sensory signals.

Many animals produce distinct sounds or substrate-borne vibrations, but these signals have proved challenging to segment with automated algorithms. We have developed SongExplorer, a web-browser based interface wrapped around a deep-learning algorithm that supports an interactive workflow for (1) discovery of animal sounds, (2) manual annotation, (3) supervised training of a deep convolutional neural network, and (4) automated segmentation of recordings. Raw data can be explored by simultaneously examining song events, both individually and in the context of the entire recording, watching synced video, and listening to song. We provide a simple way to visualize many song events from large datasets within an interactive low-dimensional visualization, which facilitates detection and correction of incorrectly labelled song events. The machine learning model we implemented displays higher accuracy than existing heuristic algorithms and similar accuracy as two expert human annotators. We show that SongExplorer allows rapid detection of all song types from new species and of novel song types in previously well-studied species.Competing Interest StatementThe authors have declared no competing interest.

How does the mammalian retina detect motion? This classic problem in visual neuroscience has remained unsolved for 50 years. In search of clues, here we reconstruct Off-type starburst amacrine cells (SACs) and bipolar cells (BCs) in serial electron microscopic images with help from EyeWire, an online community of 'citizen neuroscientists'. On the basis of quantitative analyses of contact area and branch depth in the retina, we find evidence that one BC type prefers to wire with a SAC dendrite near the SAC soma, whereas another BC type prefers to wire far from the soma. The near type is known to lag the far type in time of visual response. A mathematical model shows how such 'space-time wiring specificity' could endow SAC dendrites with receptive fields that are oriented in space-time and therefore respond selectively to stimuli that move in the outward direction from the soma.

High-throughput electron microscopy allows recording of lar- ge stacks of neural tissue with sufficient resolution to extract the wiring diagram of the underlying neural network. Current efforts to automate this process focus mainly on the segmentation of neurons. However, in order to recover a wiring diagram, synaptic partners need to be identi- fied as well. This is especially challenging in insect brains like Drosophila melanogaster, where one presynaptic site is associated with multiple post- synaptic elements. Here we propose a 3D U-Net architecture to directly identify pairs of voxels that are pre- and postsynaptic to each other. To that end, we formulate the problem of synaptic partner identification as a classification problem on long-range edges between voxels to encode both the presence of a synaptic pair and its direction. This formulation allows us to directly learn from synaptic point annotations instead of more ex- pensive voxel-based synaptic cleft or vesicle annotations. We evaluate our method on the MICCAI 2016 CREMI challenge and improve over the current state of the art, producing 3% fewer errors than the next best method.

Single-molecule localization fluorescence microscopy constructs super-resolution images by sequential imaging and computational localization of sparsely activated fluorophores. Accurate and efficient fluorophore localization algorithms are key to the success of this computational microscopy method. We present a novel localization algorithm based on deep learning which significantly improves upon the state of the art. Our contributions are a novel network architecture for simultaneous detection and localization, and new loss function which phrases detection and localization as a Bayesian inference problem, and thus allows the network to provide uncertainty-estimates. In contrast to standard methods which independently process imaging frames, our network architecture uses temporal context from multiple sequentially imaged frames to detect and localize molecules. We demonstrate the power of our method across a variety of datasets, imaging modalities, signal to noise ratios, and fluorophore densities. While existing localization algorithms can achieve optimal localization accuracy at low fluorophore densities, they are confounded by high densities. Our method is the first deep-learning based approach which achieves state-of-the-art on the SMLM2016 challenge. It achieves the best scores on 12 out of 12 data-sets when comparing both detection accuracy and precision, and excels at high densities. Finally, we investigate how unsupervised learning can be used to make the network robust against mismatch between simulated and real data. The lessons learned here are more generally relevant for the training of deep networks to solve challenging Bayesian inverse problems on spatially extended domains in biology and physics.