Deep Learning for Medical Imaging: COVID-19 Detection-MATLAB-MATHWORKS- Dr. Barath Narayanan

Barath Narayanan

Posted by Johanna Pingel, March 18, 2020

I'm pleased to publish another post from Barath Narayanan, University of Dayton Research Institute (UDRI), LinkedIn Profile. Co-author: Dr. Russell C. Hardie, University of Dayton (UD) Dr. Barath Narayanan graduated with MS and Ph.D. degree in Electrical Engineering from the University of Dayton (UD) in 2013 and 2017 respectively. He currently holds a joint appointment as a Research Scientist at UDRI's Software Systems Group and as an Adjunct Faculty for the ECE department at UD. His research interests include deep learning, machine learning, computer vision, and pattern recognition. In this blog, we are applying a Deep Learning (DL) based technique for detecting COVID-19 on Chest Radiographs using MATLAB.

 Background

 Coronavirus disease (COVID-19) is a new strain of disease in humans discovered in 2019 that has never been identified in the past. Coronavirus is a large family of viruses that causes illness in patients ranging from common cold to advanced respiratory syndromes such as Middle East Respiratory Syndrome (MERS-COV) and Severe Acute Respiratory Syndrome (SARS-COV). Many people are currently affected and are being treated across the world causing a global pandemic. In the United States alone, 160 million to 214 million people could be infected over the course of the COVID-19 epidemic (https://www.nytimes.com/2020/03/13/us/coronavirus-deaths-estimate.html). Several countries have declared a national emergency and have quarantined millions of people. Here is a detailed article on how coronavirus affects people: https://www.nytimes.com/article/coronavirus-body-symptoms.html

Detection and diagnosis tools offer a valuable second opinion to the doctors and assist them in the screening process. This type of mechanism would also assist in providing results to the doctors quickly. In this blog, we are applying a Deep Learning (DL) based technique for detecting COVID-19 on Chest Radiographs using MATLAB.

The COVID-19 dataset utilized in this blog was curated by Dr. Joseph Cohen, a postdoctoral fellow at the University of Montreal. Thanks to the article by Dr. Adrian Rosebrock for making this chest radiograph dataset reachable to researchers across the globe and for presenting the initial work using DL. Note that we solely utilize the x-ray images. You should be able to download the images from the article directly. After downloading the ZIP files from the website and extracting them to a folder called "Covid 19", we have one sub-folder per class in "dataset". Label "Covid" indicates the presence of COVID-19 in the patient and "normal" otherwise. Since, we have equal distribution (25 images) of both classes, there is no class imbalance issue here.

K-fold Validation

As you already know that there is a limited set of images available in this dataset, we split the dataset into 10-folds for analysis i.e. 10 different algorithms would be trained using different set of images from the dataset. This type of validation study would provide us a better estimate of our performance in comparison to typical hold-out validation method.

 We adopt ResNet-50 architecture in this blog as it has proven to be highly effective for various medical imaging applications [1,2].

 LINK
https://blogs.mathworks.com/deep-learning/2020/03/18/deep-learning-for-medical-imaging-covid-19-detection/

Development of 3kW Hybrid ESS with Function of Emergency Power Supply- Seok-Hyun Yang,Min-Jae Kim , Se-Wan Choi and Jun-Seok Cho- 비상전원 기능을 갖는 3kW급 하이브리드 ESS 개발

LINK :  http://www.mediafire.com/file/zy858eq2f4qpdzn/Development_of_3kW_Hybrid_ESS_with_Function_of_Eme.pdf/file

Analysis and Design of PV based Energy Storage System with Improved MPPT Algorithm-개선된 MPPT 기법을 적용한 PV 기반 에너지 저장 시스템의 설계와 해석

Analysis and Design of PV based Energy Storage System with Improved MPPT Algorithm-개선된 MPPT 기법을 적용한 PV 기반 에너지 저장 시스템의 설계와 해석 

 ABSTRACT

Analysis and Design of PV based Energy Storage System with improved MPPT Algorithm Paeng, Seong Il Department of Electrical Engineering Graduate School of Konkuk University

 Research and use of renewable energy is becoming more active due to the demand for reduction of carbon dioxide, which is the main cause of depletion of fossil fuels and global climate change, and regulation of the generation. However, Renewable energy using sunlight or wind power has a problem that it is adversely affecting the system due to irregularity of solar radiation or wind power, or it is difficult to operate efficiently. To overcome this problem, PV ESS (Photovoltaic Energy Storage System), which is used after storing power using a battery, has been actively studied. The operation method and control method of PV ESS linked with solar power generation and ESS are proposed into single phase and three phase in this PhD thesis. In the single phase system, during the daytime, the electric power generated from the photovoltaic power generation is used to supply power to the LED lamp, charge the battery, and transmit power to the grid when surplus power is generated. In the night, the - xii - LED lamp is supplied by the power stored in the battery, and the battery is charged by the grid power in the late night. In the daytime mode, technical improvements in Maximum Power Point Tracking(MPPT) algorithm is proposed for single phase photovoltaic Energy Storage System(PV ESS). The three-phase system performs a power leveling function to improve daytime and nighttime power uneven consumption and a function to level out the irregularities of solar power generation power. Therefore, during the day when the solar power is less than the set power, the battery replenishes the power. In the nighttime, battery power is used to transmit power to the system as much as the set value to be used in the load, and the battery is charged using surplus power of the night time. In case of power failure, three phase PV ESS is used as UPS (Uninterruptable Power Supply) function. This paper will contain two major sections which are briefed below. First, a single phase 3.3kW, 600W PV ESS system is designed and a control algorithm according to operating mode is proposed. And a three phase 15kW PV ESS system is designed and a control algorithm according to operation mode is proposed. Second, the MPPT algorithm is explained and the developed MPPT algorithms is proposed. Second, there are various MPPT algorithms in the paper which are reviewed and new MPPT algorithm based the P&O(Perturb and Observe) method is proposed. The proposed algorithms is carried out in simulation and have been validated in the prototype.

Keyword : Photovoltaic, Energy Storage System, Maximum Power Point Tracking

Photovoltaic Power Generation Control System Using MPPT and Android Programming-Pusan National University Department of Electrical and Computer Engineering Author Mallavarapu Sindhu

Thesis for the degree of Master of Science Photovoltaic Power Generation Control System Using MPPT and Android Programming
Supervisor Hee-Je Kim August 2017
The Graduate School Pusan National University Department of Electrical and Computer Engineering Author Mallavarapu Sindhu
LINK:http://www.mediafire.com/file/jc0t4c3pf1ry8lg/Photovoltaic_Power_Generation_Control_System_Using_MPPT_and_Android.pdf/file

A study on simulator or solar generation and MPPT to improve generation efficiency-태양광 발전시스템 모의를 위한 장치 및 발전 효율을 높일 수 있는 MPPT 연구-KWANGWOON UNIVERSITY

A study on simulator or solar generation and MPPT to improve generation efficiency 태양광 발전시스템 모의를 위한 장치 및 발전 효율을 높일 수 있는 MPPT 연구

 Abstract

 Photovoltaic Generation Simulators and MPPT Strategy to Improve Photovoltaic Power Generation Efficiency in Partial Shadows In this paper, the study is performed to supplement MPPT efficiency in partial solar irradiation and can be simulated with solar power plant power. In order to produce photovoltaic inverter and research of photovoltaic module, a simulation device capable of outputting power of photovoltaic power generation is needed. The characteristic curve differs for each photovoltaic cell and the characteristic curve has a nonlinear characteristic. A power device capable of outputting a curve corresponding to the characteristic curve is needed. We have studied about the characteristic curves of these photovoltaic cells and studied mathematical techniques for expressing curves that can be output according to the module characteristics. In order to use this system effectively, we have implemented a simulation and a device to convert a system AC power to a DC power output from a solar cell using a single phase inverter. In order to maximize the power generation efficiency of the solar photovoltaic system, it is necessary to control the output power at the maximum power point in nonlinear output characteristics. This is called maximum power follow-up control (MPPT). Various MPPT algorithms have been used in the past, but the mismatching phenomenon occurs due to shadow due to the installation location of the PV power plant and surrounding environment. If a mismatching phenomenon occurs, it fails to output at the maximum power point, thereby deteriorating the solar power generation efficiency. In order to increase this, several inverters are used in one inverter to control the solar array partly, and the necessary algorithms are proposed. The power conversion is performed to compare the efficiency of the multiple - boost converter in the mismatching. The validity of the proposed algorithm is verified through simulations and instruments.

LINK: http://www.mediafire.com/file/dujwqa17opuipa5/A_study_on_simulator_or_solar_generation_and_MPPT_to_improve_generation_efficiency.pdf/file