DSSAE: Deep Stacked Sparse Autoencoder Analytical Model for COVID-19 Diagnosis by Fractional Fourier Entropy

2022 ◽  
Vol 13 (1) ◽  
pp. 1-20
Author(s):  
Shui-Hua Wang ◽  
Xin Zhang ◽  
Yu-Dong Zhang
Keyword(s):  
Data Augmentation ◽  
State Of The Art ◽  
Community Acquired Pneumonia ◽  
Chest Ct ◽  
Two Dimensional ◽  
The World ◽  
Healthy Control ◽  
Sparse Autoencoder ◽  
Stacked Sparse Autoencoder ◽  
Fractional Fourier Entropy

( Aim ) COVID-19 has caused more than 2.28 million deaths till 4/Feb/2021 while it is still spreading across the world. This study proposed a novel artificial intelligence model to diagnose COVID-19 based on chest CT images. ( Methods ) First, the two-dimensional fractional Fourier entropy was used to extract features. Second, a custom deep stacked sparse autoencoder (DSSAE) model was created to serve as the classifier. Third, an improved multiple-way data augmentation was proposed to resist overfitting. ( Results ) Our DSSAE model obtains a micro-averaged F1 score of 92.32% in handling a four-class problem (COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis, and healthy control). ( Conclusion ) Our method outperforms 10 state-of-the-art approaches.

scholarly journals Deep Fractional Max Pooling Neural Network for COVID-19 Recognition

2021 ◽  
Vol 9 ◽  
Author(s):  
Shui-Hua Wang ◽  
Suresh Chandra Satapathy ◽  
Donovan Anderson ◽  
Shi-Xin Chen ◽  
Yu-Dong Zhang
Keyword(s):  
Neural Network ◽  
Data Augmentation ◽  
State Of The Art ◽  
Model Averaging ◽  
Community Acquired Pneumonia ◽  
Test Set ◽  
Max Pooling ◽  
L2 Norm ◽  
Healthy Control ◽  
Novel Model

Aim: Coronavirus disease 2019 (COVID-19) is a form of disease triggered by a new strain of coronavirus. This paper proposes a novel model termed “deep fractional max pooling neural network (DFMPNN)” to diagnose COVID-19 more efficiently.Methods: This 12-layer DFMPNN replaces max pooling (MP) and average pooling (AP) in ordinary neural networks with the help of a novel pooling method called “fractional max-pooling” (FMP). In addition, multiple-way data augmentation (DA) is employed to reduce overfitting. Model averaging (MA) is used to reduce randomness.Results: We ran our algorithm on a four-category dataset that contained COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis (SPT), and healthy control (HC). The 10 runs on the test set show that the micro-averaged F1 (MAF) score of our DFMPNN is 95.88%.Discussions: This proposed DFMPNN is superior to 10 state-of-the-art models. Besides, FMP outperforms traditional MP, AP, and L2-norm pooling (L2P).

scholarly journals SECONDARY PULMONARY TUBERCULOSIS RECOGNITION BY ROTATION ANGLE VECTOR GRID-BASED FRACTIONAL FOURIER ENTROPY

Fractals ◽  
2021 ◽  
Author(s):  
SHUI-HUA WANG ◽  
YELIZ KARACA ◽  
XIN ZHANG ◽  
YU-DONG ZHANG
Keyword(s):  
Infectious Disease ◽  
Rotation Angle ◽  
State Of The Art ◽  
Training Set ◽  
Feature Extractor ◽  
Sparse Autoencoder ◽  
Stacked Sparse Autoencoder ◽  
Novel Model ◽  
Fractional Fourier Entropy ◽  
Grid Based

Aim: Tuberculosis is an infectious disease caused by Mycobacterium tuberculosis bacteria. This study plans to build a novel deep learning-based model for the accurate recognition of tuberculosis. Methods: We propose a novel model — rotation angle vector grid-based fractional Fourier entropy and deep stacked sparse autoencoder (RAVG-FrFE–DSSAE) — which uses RAVG-FrFE as a feature extractor and harnesses DSSAE as the classifier. Moreover, an 18-way MDA is introduced on the training set to avoid overfitting. Results: Experimental results of 10 runs of 10-fold CV showcase that this proposed RAVG-FrFE–DSSAE algorithm yields a reasonable performance including of 93.68[Formula: see text]±[Formula: see text]1.11% sensitivity, 94.38[Formula: see text]±[Formula: see text]1.11% specificity, 94.35[Formula: see text]±[Formula: see text]1.04% precision, 94.03[Formula: see text]±[Formula: see text]0.69% accuracy, 94.01[Formula: see text]±[Formula: see text]0.70% [Formula: see text]-score, 88.07[Formula: see text]±[Formula: see text]1.38% MCC, 94.01[Formula: see text]±[Formula: see text]0.70% FMI, and 0.9725 AUC, respectively. Conclusions: Our result outperforms the eight state-of-the-art approaches. Besides, the result shows the effectiveness of the 18-way MDA.

Multimedia-Enabled Dot Codes as Communication Aids

2021 ◽  
pp. 331-345
Author(s):  
Shigeru Ikuta ◽  
Masamichi Watanuki ◽  
Shinya Abe
Keyword(s):  
Special Needs ◽  
State Of The Art ◽  
The State ◽  
Digital Information ◽  
Two Dimensional ◽  
Basic Information ◽  
Communication Aids ◽  
Teaching Materials ◽  
The World ◽  
The Creation

Grid Onput is a set of novel two-dimensional codes comprising extremely small dots. The present authors recently developed software to overlap the dot codes on the user's designed sheet, to create a content to replay audios, to create a standalone application to replay multimedia, and to create an application to replay multimedia on iPad. Simply touching the dot codes with a speaking-pen and/or a dot-code reader enables users to directly access the corresponding digital information; a maximum of four mediums can be easily linked to each dot code icon. In collaboration with schoolteachers all over the world, one of the authors, Shigeru Ikuta, has been creating a variety of original self-made content and conducting various activities at both general and special needs schools. This chapter outlines the recent development of the state-of-the-art Grid Onput dot code technology and presents basic information regarding the creation of original teaching materials using newly developed software and the use at both general and special needs schools.

scholarly journals PSSPNN: PatchShuffle Stochastic Pooling Neural Network for an Explainable Diagnosis of COVID-19 with Multiple-Way Data Augmentation

2021 ◽  
Vol 2021 ◽  
pp. 1-18
Author(s):  
Shui-Hua Wang ◽  
Yin Zhang ◽  
Xiaochun Cheng ◽  
Xin Zhang ◽  
Yu-Dong Zhang
Keyword(s):  
Neural Network ◽  
Pulmonary Tuberculosis ◽  
Healthy Subjects ◽  
Data Augmentation ◽  
State Of The Art ◽  
Accurate Diagnosis ◽  
Test Set ◽  
The World ◽  
Stochastic Pooling ◽  
Better Than

Aim. COVID-19 has caused large death tolls all over the world. Accurate diagnosis is of significant importance for early treatment. Methods. In this study, we proposed a novel PSSPNN model for classification between COVID-19, secondary pulmonary tuberculosis, community-captured pneumonia, and healthy subjects. PSSPNN entails five improvements: we first proposed the n-conv stochastic pooling module. Second, a novel stochastic pooling neural network was proposed. Third, PatchShuffle was introduced as a regularization term. Fourth, an improved multiple-way data augmentation was used. Fifth, Grad-CAM was utilized to interpret our AI model. Results. The 10 runs with random seed on the test set showed our algorithm achieved a microaveraged F1 score of 95.79%. Moreover, our method is better than nine state-of-the-art approaches. Conclusion. This proposed PSSPNN will help assist radiologists to make diagnosis more quickly and accurately on COVID-19 cases.

Levenshtein Augmentation Improves Performance of SMILES Based Deep-Learning Synthesis Prediction

2020 ◽  
Author(s):  
Dean Sumner ◽  
Jiazhen He ◽  
Amol Thakkar ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Deep Learning ◽  
Recurrent Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Sequence Similarity ◽  
Learning Models ◽  
Underlying Network

<p>SMILES randomization, a form of data augmentation, has previously been shown to increase the performance of deep learning models compared to non-augmented baselines. Here, we propose a novel data augmentation method we call “Levenshtein augmentation” which considers local SMILES sub-sequence similarity between reactants and their respective products when creating training pairs. The performance of Levenshtein augmentation was tested using two state of the art models - transformer and sequence-to-sequence based recurrent neural networks with attention. Levenshtein augmentation demonstrated an increase performance over non-augmented, and conventionally SMILES randomization augmented data when used for training of baseline models. Furthermore, Levenshtein augmentation seemingly results in what we define as <i>attentional gain </i>– an enhancement in the pattern recognition capabilities of the underlying network to molecular motifs.</p>

First-Principles Evaluation of the Potential of Borophene as a Monolayer Transparent Conductor

2017 ◽  
Author(s):  
Lyudmyla Adamska ◽  
Sridhar Sadasivam ◽  
Jonathan J. Foley ◽  
Pierre Darancet ◽  
Sahar Sharifzadeh
Keyword(s):  
First Principles ◽  
Density Functional ◽  
State Of The Art ◽  
Transparent Electrode ◽  
Visible Range ◽  
Two Dimensional ◽  
Transparent Conductor ◽  
Many Body ◽  
Functional Theory ◽  
Many Body Perturbation Theory

Two-dimensional boron is promising as a tunable monolayer metal for nano-optoelectronics. We study the optoelectronic properties of two likely allotropes of two-dimensional boron using first-principles density functional theory and many-body perturbation theory. We find that both systems are anisotropic metals, with strong energy- and thickness-dependent optical transparency and a weak (<1%) absorbance in the visible range. Additionally, using state-of-the-art methods for the description of the electron-phonon and electron-electron interactions, we show that the electrical conductivity is limited by electron-phonon interactions. Our results indicate that both structures are suitable as a transparent electrode.

State-of-the-Art of Modeling Surface Water Impoundments

1982 ◽  
Vol 14 (1-2) ◽  
pp. 241-261 ◽  
Author(s):  
P A Krenkel ◽  
R H French
Keyword(s):  
Water Quality ◽  
Surface Water ◽  
State Of The Art ◽  
Rate Coefficients ◽  
Two Dimensional ◽  
One Dimensional ◽  
Integral Energy ◽  
Range Of Values ◽  
Dimensional Integral ◽  
Water Impoundment

The state-of-the-art of surface water impoundment modeling is examined from the viewpoints of both hydrodynamics and water quality. In the area of hydrodynamics current one dimensional integral energy and two dimensional models are discussed. In the area of water quality, the formulations used for various parameters are presented with a range of values for the associated rate coefficients.

scholarly journals D-Penicillamine: The State of the Art in Humans and in Dogs from a Pharmacological and Regulatory Perspective

Antibiotics ◽  
2021 ◽  
Vol 10 (6) ◽  
pp. 648
Author(s):  
Michela Pugliese ◽  
Vito Biondi ◽  
Enrico Gugliandolo ◽  
Patrizia Licata ◽  
Alessio Filippo Peritore ◽  
...  
Keyword(s):  
Veterinary Medicine ◽  
State Of The Art ◽  
European Medicines Agency ◽  
Therapeutic Goals ◽  
First Choice ◽  
Regulatory Perspective ◽  
The World ◽  
United States Food ◽  
States Food ◽  
Food And Drugs Administration

Chelant agents are the mainstay of treatment in copper-associated hepatitis in humans, where D-penicillamine is the chelant agent of first choice. In veterinary medicine, the use of D-penicillamine has increased with the recent recognition of copper-associated hepatopathies that occur in several breeds of dogs. Although the different regulatory authorities in the world (United States Food and Drugs Administration—U.S. FDA, European Medicines Agency—EMEA, etc.) do not approve D-penicillamine for use in dogs, it has been used to treat copper-associated hepatitis in dogs since the 1970s, and is prescribed legally by veterinarians as an extra-label drug to treat this disease and alleviate suffering. The present study aims to: (a) address the pharmacological features; (b) outline the clinical scenario underlying the increased interest in D-penicillamine by overviewing the evolution of its main therapeutic goals in humans and dogs; and finally, (c) provide a discussion on its use and prescription in veterinary medicine from a regulatory perspective.

Sign in / Sign up

Export Citation Format

Share Document