Visualising uncertainty in brightfield to fluorescent image inference with BFNet

AbstractPredicting fluorescently labelled cellular structures from brightfield images is a recent application of convolutional neural networks and has already proven a valuable tool in biological imaging studies. These methods reduce the need for time-consuming manual annotations for supervised datasets, potentially reduce the costs in high-throughput imaging screens, and open up a number of potentially novel analyses. However as with any prediction there can be sources of error and uncertainty. Here we present BFNet, a method to visualise the uncertainty in predicted images by applying Monte Carlo dropout during inference and calculating per-pixel variance as an uncertainty map. Our method demonstrates the ability to highlight regions of an image where prediction is difficult or impossible due to imaging artefacts such as occlusions or out-of-focus images, as well as more general uncertainty when a trained model is applied to new data from different imaging of experimental settings. We have provided a python implementation of the method which is available at github.com/swarchal/bfnet.

Download Full-text

Review for "Towards operational phytoplankton recognition with automated high-throughput imaging and compact convolutional neural networks"

10.5194/os-2020-62-rc1 ◽

2020 ◽

Author(s):

Anonymous

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

High Throughput ◽

High Throughput Imaging

Download Full-text

Towards computer-aided severity assessment via deep neural networks for geographic and opacity extent scoring of SARS-CoV-2 chest X-rays

Scientific Reports ◽

10.1038/s41598-021-88538-4 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

A. Wong ◽

Z. Q. Lin ◽

L. Wang ◽

A. G. Chung ◽

B. Shen ◽

...

Keyword(s):

Neural Networks ◽

Monte Carlo ◽

Lung Disease ◽

Disease Severity ◽

Deep Neural Networks ◽

Cross Validation ◽

X Rays ◽

Computer Aided ◽

Monte Carlo Cross Validation ◽

Validation Experiments

AbstractA critical step in effective care and treatment planning for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the cause for the coronavirus disease 2019 (COVID-19) pandemic, is the assessment of the severity of disease progression. Chest x-rays (CXRs) are often used to assess SARS-CoV-2 severity, with two important assessment metrics being extent of lung involvement and degree of opacity. In this proof-of-concept study, we assess the feasibility of computer-aided scoring of CXRs of SARS-CoV-2 lung disease severity using a deep learning system. Data consisted of 396 CXRs from SARS-CoV-2 positive patient cases. Geographic extent and opacity extent were scored by two board-certified expert chest radiologists (with 20+ years of experience) and a 2nd-year radiology resident. The deep neural networks used in this study, which we name COVID-Net S, are based on a COVID-Net network architecture. 100 versions of the network were independently learned (50 to perform geographic extent scoring and 50 to perform opacity extent scoring) using random subsets of CXRs from the study, and we evaluated the networks using stratified Monte Carlo cross-validation experiments. The COVID-Net S deep neural networks yielded R$$^2$$ 2 of $$0.664 \pm 0.032$$ 0.664 ± 0.032 and $$0.635 \pm 0.044$$ 0.635 ± 0.044 between predicted scores and radiologist scores for geographic extent and opacity extent, respectively, in stratified Monte Carlo cross-validation experiments. The best performing COVID-Net S networks achieved R$$^2$$ 2 of 0.739 and 0.741 between predicted scores and radiologist scores for geographic extent and opacity extent, respectively. The results are promising and suggest that the use of deep neural networks on CXRs could be an effective tool for computer-aided assessment of SARS-CoV-2 lung disease severity, although additional studies are needed before adoption for routine clinical use.

Download Full-text

Using Neural Networks to Deal with Three Phonon Scattering in Phonon Monte Carlo Simulation

2021 International Conference on Computer, Control and Robotics (ICCCR) ◽

10.1109/icccr49711.2021.9349415 ◽

2021 ◽

Author(s):

Wenhui Ni ◽

Minhua Chen

Keyword(s):

Neural Networks ◽

Monte Carlo Simulation ◽

Monte Carlo ◽

Phonon Scattering

Download Full-text

High-throughput phenotyping of two plant-size traits of Eucalyptus species using neural networks

Journal of Forestry Research ◽

10.1007/s11676-021-01360-6 ◽

2021 ◽

Author(s):

Marcus Vinicius Vieira Borges ◽

Janielle de Oliveira Garcia ◽

Tays Silva Batista ◽

Alexsandra Nogueira Martins Silva ◽

Fabio Henrique Rojo Baio ◽

...

Keyword(s):

Neural Networks ◽

High Throughput ◽

Mean Squared Error ◽

Vegetation Indices ◽

Plant Size ◽

Eucalyptus Species ◽

Forest Inventories ◽

Spectral Bands ◽

High Throughput Phenotyping ◽

Input Variables

AbstractIn forest modeling to estimate the volume of wood, artificial intelligence has been shown to be quite efficient, especially using artificial neural networks (ANNs). Here we tested whether diameter at breast height (DBH) and the total plant height (Ht) of eucalyptus can be predicted at the stand level using spectral bands measured by an unmanned aerial vehicle (UAV) multispectral sensor and vegetation indices. To do so, using the data obtained by the UAV as input variables, we tested different configurations (number of hidden layers and number of neurons in each layer) of ANNs for predicting DBH and Ht at stand level for different Eucalyptus species. The experimental design was randomized blocks with four replicates, with 20 trees in each experimental plot. The treatments comprised five Eucalyptus species (E. camaldulensis, E. uroplylla, E. saligna, E. grandis, and E. urograndis) and Corymbria citriodora. DBH and Ht for each plot at the stand level were measured seven times in separate overflights by the UAV, so that the multispectral sensor could obtain spectral bands to calculate vegetation indices (VIs). ANNs were then constructed using spectral bands and VIs as input layers, in addition to the categorical variable (species), to predict DBH and Ht at the stand level simultaneously. This report represents one of the first applications of high-throughput phenotyping for plant size traits in Eucalyptus species. In general, ANNs containing three hidden layers gave better statistical performance (higher estimated r, lower estimated root mean squared error–RMSE) due to their greater capacity for self-learning. Among these ANNs, the best contained eight neurons in the first layer, seven in the second, and five in the third (8 − 7 − 5). The results reported here reveal the potential of using the generated models to perform accurate forest inventories based on spectral bands and VIs obtained with a UAV multispectral sensor and ANNs, reducing labor and time.

Download Full-text

Reconfigurable Binary Neural Network Accelerator with Adaptive Parallelism Scheme

Electronics ◽

10.3390/electronics10030230 ◽

2021 ◽

Vol 10 (3) ◽

pp. 230

Author(s):

Jaechan Cho ◽

Yongchul Jung ◽

Seongjoo Lee ◽

Yunho Jung

Keyword(s):

Neural Networks ◽

High Throughput ◽

Deep Neural Networks ◽

State Of The Art ◽

Throughput Performance ◽

Adaptive Parallelism ◽

Sensor Applications ◽

Binary Neural Network ◽

Target Layer ◽

Network Topologies

Binary neural networks (BNNs) have attracted significant interest for the implementation of deep neural networks (DNNs) on resource-constrained edge devices, and various BNN accelerator architectures have been proposed to achieve higher efficiency. BNN accelerators can be divided into two categories: streaming and layer accelerators. Although streaming accelerators designed for a specific BNN network topology provide high throughput, they are infeasible for various sensor applications in edge AI because of their complexity and inflexibility. In contrast, layer accelerators with reasonable resources can support various network topologies, but they operate with the same parallelism for all the layers of the BNN, which degrades throughput performance at certain layers. To overcome this problem, we propose a BNN accelerator with adaptive parallelism that offers high throughput performance in all layers. The proposed accelerator analyzes target layer parameters and operates with optimal parallelism using reasonable resources. In addition, this architecture is able to fully compute all types of BNN layers thanks to its reconfigurability, and it can achieve a higher area–speed efficiency than existing accelerators. In performance evaluation using state-of-the-art BNN topologies, the designed BNN accelerator achieved an area–speed efficiency 9.69 times higher than previous FPGA implementations and 24% higher than existing VLSI implementations for BNNs.

Download Full-text

Inverse design of glass structure with deep graph neural networks

Nature Communications ◽

10.1038/s41467-021-25490-x ◽

2021 ◽

Vol 12 (1) ◽

Author(s):

Qi Wang ◽

Longfei Zhang

Keyword(s):

Neural Networks ◽

Monte Carlo ◽

Configuration Space ◽

Glass Structure ◽

Stable State ◽

Specific Property ◽

Inverse Design ◽

Data Driven ◽

Common Belief ◽

Graph Neural Networks

AbstractDirectly manipulating the atomic structure to achieve a specific property is a long pursuit in the field of materials. However, hindered by the disordered, non-prototypical glass structure and the complex interplay between structure and property, such inverse design is dauntingly hard for glasses. Here, combining two cutting-edge techniques, graph neural networks and swap Monte Carlo, we develop a data-driven, property-oriented inverse design route that managed to improve the plastic resistance of Cu-Zr metallic glasses in a controllable way. Swap Monte Carlo, as a sampler, effectively explores the glass landscape, and graph neural networks, with high regression accuracy in predicting the plastic resistance, serves as a decider to guide the search in configuration space. Via an unconventional strengthening mechanism, a geometrically ultra-stable yet energetically meta-stable state is unraveled, contrary to the common belief that the higher the energy, the lower the plastic resistance. This demonstrates a vast configuration space that can be easily overlooked by conventional atomistic simulations. The data-driven techniques, structural search methods and optimization algorithms consolidate to form a toolbox, paving a new way to the design of glassy materials.

Download Full-text

Neural network based stochastic design charts for settlement prediction

Canadian Geotechnical Journal ◽

10.1139/t04-096 ◽

2005 ◽

Vol 42 (1) ◽

pp. 110-120 ◽

Cited By ~ 17

Author(s):

M A Shahin ◽

M B Jaksa ◽

H R Maier

Keyword(s):

Neural Networks ◽

Monte Carlo ◽

Artificial Neural Networks ◽

Shallow Foundations ◽

Granular Soils ◽

Traditional Methods ◽

Settlement Prediction ◽

Design Charts ◽

Artificial Neural ◽

Level Of Risk

Traditional methods of settlement prediction of shallow foundations on granular soils are far from accurate and consistent. This can be attributed to the fact that the problem of estimating the settlement of shallow foundations on granular soils is very complex and not yet entirely understood. Recently, artificial neural networks (ANNs) have been shown to outperform the most commonly used traditional methods for predicting the settlement of shallow foundations on granular soils. However, despite the relative advantage of the ANN based approach, it does not take into account the uncertainty that may affect the magnitude of the predicted settlement. Artificial neural networks, like more traditional methods of settlement prediction, are based on deterministic approaches that ignore this uncertainty and thus provide single values of settlement with no indication of the level of risk associated with these values. An alternative stochastic approach is essential to provide more rational estimation of settlement. In this paper, the likely distribution of predicted settlements, given the uncertainties associated with settlement prediction, is obtained by combining Monte Carlo simulation with a deterministic ANN model. A set of stochastic design charts, which incorporate the uncertainty associated with the ANN method, is developed. The charts are considered to be useful in the sense that they enable the designer to make informed decisions regarding the level of risk associated with predicted settlements and consequently provide a more realistic indication of what the actual settlement might be.Key words: settlement prediction, shallow foundations, neural networks, Monte Carlo, stochastic simulation.

Download Full-text