Visual-Health-Management-with-AI
hits:303    addtime: 2023/12/11



  © 2023

  Visual-Health-Management-with-AI

  Edit: Prof. Dr. Wen-Feng Wang, Lalit Mohan Patnaik and Bin Hu

  Lalit Mohan Patnaik would like to thank the National Academy of Sciences India(NASI), Allahabad, India for the support and to the Director, National Institute of Advanced Studies (NIAS), Bengaluru, India for  providing the infrastructure facilities to carry out this work.

  ISBN: 978-1-80430-001-5

  DOI: 10.56157/JPNT3977

  Publication Date: December 11, 2023

Method of obtaining


Description

Functional analysis was born in the early 20th century. Functional analysis was developed from the study of variational problems, integral equations and theoretical physics. The theoretical value of this book lies in the comprehensive application of functional theory, geometry and modern mathematics to study the functional, operator and limit theory on infinite dimensional vector space. Functional analysis can be regarded as analytic geometry and mathematical analysis of infinite dimensional vector space. Therefore, the practical significance of this book lies in that functional analysis is applied in mathematical physical equations, probability theory, computational mathematics and other branches, and it is also a mathematical tool to study physical systems with infinite degrees of freedom.As a research and development of theories and methods for simulating, extending, and expanding human intelligence, artificial intelligence (AI) is changing the world. This book consists of 8 chapters, which is also the latest attempts of my team in health management with AI.

Chapter 1 mainly uses Python as the basic language for analysis, and uses the in-depth learning algorithm of convolutional neural networks based on the online query of the MCNN algorithm to carry out certain personal epidemic prevention and control now that the epidemic situation is open. Through collecting data such as photos, learning analysis is conducted, and through compiling relevant Python code on Pycharm, the data is analyzed to obtain a population distribution map, Calculate the population density through a given function and compare it with a given population density to determine whether the population is clustered and the degree of aggregation. The risk level of the photo site is determined by the degree of aggregation of the population. This allows for simple protection of oneself, including monitoring a certain place. Chapter 2 will establish a game model from the perspective of game theory, aiming at controlling the COVID-19 epidemic with appropriate and scientific allocation of funds and minimum cost of epidemic prevention. Chapter 3 will establish a game model from the perspective of game theory, aiming at controlling the COVID-19 epidemic with appropriate and scientific allocation of funds and minimum cost of epidemic prevention.Chapter 1 mainly uses Python as the basic language for analysis, and uses the in-depth learning algorithm of convolutional neural networks based on the online query of the MCNN algorithm to carry out certain personal epidemic prevention and control now that the epidemic situation is open. Through collecting data such as photos, learning analysis is conducted, and through compiling relevant Python code on Pycharm, the data is analyzed to obtain a population distribution map, Calculate the population density through a given function and compare it with a given population density to determine whether the population is clustered and the degree of aggregation. The risk level of the photo site is determined by the degree of aggregation of the population. This allows for simple protection of oneself, including monitoring a certain place. Chapter 2 will establish a game model from the perspective of game theory, aiming at controlling the COVID-19 epidemic with appropriate and scientific allocation of funds and minimum cost of epidemic prevention. Chapter 3 will establish a game model from the perspective of game theory, aiming at controlling the COVID-19 epidemic with appropriate and scientific allocation of funds and minimum cost of epidemic prevention.

Chapter 4 introduces the research background and significance of artificial intelligence, involving deep reinforcement learning (DQN) of artificial intelligence algorithms. For the health management of COVID-19 patients, this chapter mainly uses the travel analysis of COVID-19 patients to establish the artificial intelligence training environment model, and also uses Python programming methods to achieve the training of artificial intelligence algorithms, and finally obtains a valuable solution. This chapter first explains artificial intelligence, and then envisions a simplified training environment to study the feasibility of algorithms for solving such problems. Finally, the feasibility was demonstrated through experiments and training of the algorithm. The design of Chapter 5 is mainly based on the Arduino control board, equipped with various hardware and corresponding software programming implementations to assist families in dealing with epidemics. The hardware system of the designed home epidemic prevention system mainly includes: Arduino UNO main control board, tracking and obstacle avoidance module, temperature and humidity measurement module, human fall detection monitoring module, and medication reminder module. Among them, the fall detection module is a human fall intelligent monitoring system built using Python, Opencv, and MediaPipe, which can observe in real-time if a person is falling, standing, or sitting.

Chapter 6 aims to study how to use Python for data analysis, and proposes a prediction model based on deep learning. We have used the requests library in Python for data crawlers, Pandas, NumPy, and Matplotlib to process and visualize data, and used LS-SVM to solve model parameters using the least squares method. According to the experimental results, my proposed model has achieved better prediction results than traditional models on multiple data sets. My research results provide strong support for the application of Python in the field of data analysis, and can play a role in many application scenarios. Future research can further optimize model performance or apply it to a wider range of fields.

Chapter 7 aims to judge the risk degree of COVID-19 infection by collecting the information of face and whether the person has been infected with COVID-19. Face detection is carried out by yoloface algorithm, and then face recognition is carried out to extract the face information. According to whether the person has been infected with COVID-19, the information is stored in the corresponding database to complete the information entry. When a person's face is detected again, the program displays the person's risk of contracting COVID-19. The emergence of the novel coronavirus has a great impact on the society, but also has a great impact on people's life, so that people's lifestyle has changed. With the continuous development of deep learning, the technology of face recognition has been rapidly developed. Because of the advantages of face recognition, such as no contact, low cost and long distance, it has been more and more widely used in life. Face brushing has become a common thing for people to go out and enter the community, units, schools, etc., need to be verified by face brushing, thus reducing the flow of people. This subject aims to judge the risk degree of COVID-19 infection by collecting the information of face and whether the person has been infected with COVID-19. Face detection is carried out by yoloface algorithm, and then face recognition is carried out to extract the face information. According to whether the person has been infected with COVID-19, the information is stored in the corresponding database to complete the information entry. When a person's face is detected again, the program displays the person's risk of contracting COVID-19.

Chapter 8 collected experimental data on postprandial blood glucose changes, experimental data on drug control of blood glucose, experimental data on exercise control of blood glucose, and simulation data sets shared by video anchors on the Internet, combined with random forest, decision tree, logical regression and other machine learning algorithms, to build mathematical models for hyperglycemia risk early warning and sugar control strategy research. The results showed that the machine learning model was used to establish the early warning mechanism of hyperglycemia risk, which could provide more active and effective suggestions for diabetes patients to develop blood glucose control strategies, and provide active guidance and help for patient groups and individuals who lack awareness of self-health-management.

Throughout this book, we aim to help the readers to understand why visual health management can employ AI as an important tool, and how to utilize this tool in the researches on practical engineering projects.


About the authors

      Wen-Feng Wang is a full professor in Shanghai Institute of Technology. He is a key talent in Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences (2018- ), a reviewer of many SCI journals, including some top journals - Water Research, Science China-Information Sciences, Science of the Total Environment, Environmental Pollution, IEEE Transactions on Automation Science and Engineering and etc. He served as a keynote speaker of AMICR2019, IACICE2020, OAES2020, AICS2021, 3DIT-MSP&DL2020 and etc. He is also the scientist in chief of RealMax in the research field of intelligent computing and big data.


      Lalit Mohan Patnaik has served as the President of the Advanced Computing and Communications Society, and the Computational Intelligence Society of India. He is a Fellow of The World Academy of Sciences, Indian National Science Academy, Indian Academy of Sciences, National Academy of Sciences, Indian National Academy of Engineering, Maharashtra Academy of Sciences. His name appears in Asia’s Who's Who of Men and Women of Achievement; Directory of Distinguished Computer Professionals of India. For over one hundred and seventy International Conferences, he has served as General Chair or Program Chair or Steering Committee Chair; most of them sponsored by the Institute of Electrical and Electronics Engineers (IEEE).


      Bin Hu is an engineer and a counselor at Changsha Normal University, ACM Member, IEEE Member, GCDF, he is the guest editor of EI-indexed journals or book series: International Journal of Reasoning-based Intelligent Systems (IJRIS) , International Journal of Information Systems and Supply Chain Management (IJISSCM), International Journal of Information Systems in the Service Sector(IJISSS) and Smart Innovation, Systems and Technologies(SIST), the technical editor of International Journal of Computer and Communication Technology (IJCCT). He published many articles in high standard journals like Knowledge-based System (KS), Journal of Intelligent and Fuzzy Control (JIFS), Enterprise Information System(EIS), Financial Innovation(FI) and etc.


Table of  Contents

Visual Health Management with AI
Editorial Committee
Preface
Chapter 1. Deep Learning for Epidemic Prevention and Control
1 Introduction
1.1 Research background
1.2 Research topics and trends
1.3 Research direction of our project
1.4 Research problems and objectives
2 Some Preliminaries
2.1 The learning system
2.1.1 Programming idea
2.1.2 Mechanical learning
2.1.3 Convolutional neural network
2.1.4 Trends in deep learning models
2.2 Crowd counting algorithm
2.2.1 Traditional machine learning
2.2.2 Multi
column convolutional networks
3 Modelling Approach
3.1 Overview of the algorithm
3.2 Generating density from an image
3.2.1 Crowd image representation
3.2.2 Crowd density map conversion
3.3 Construction of a density map
3.4 Establish a deep counting model
3.5 The principles for training the model
4 Technological Implementation
4.1 Collection of learning data
4.2 Photo resolution reduction
4.3 Generating an annotation file
4.4 Generate ground_truth map
4.5 Programming to determine risk levels
5 Unresolved Issues
Chapter 2. Design of a Cost Game Model for Epidemic Prevention
1 Introduction
1.1 The motivation
1.2 Research backgrounds
2 Some Preliminaries
2.1 Compiler configuration
2.2 The overall design steps
2.3 The programming steps
2.4 Python Library Configuration
3 Technical implementation
3.1 The kernel code module
3.2 Analysis of the results
3.3 Summary and discussion
4. Unresolved Issues
Chapter 3. Approaches to Infer the Risk of COVID-19 Infection
1 Introduction
1.1 Research background
1.2 Research topics and trends
1.3 Organization of this chapter
2 Theory for Programming
2.1 The Bayes Theorem
2.2 The Bayesian networks
2.3 The considered problem
2.4 Utilization of Python
2.4.1 Big data technology
2.4.2 Tools and function modules
2.4.3 Technical development process
2.5 Fundamentals of web crawlers
3. Technology for Data Processig
3.1 Tools needed for web crawlers
3.1.1 Request Library Overview
3.1.2 Overview of the parsing library
3.2 Programming and parsing
3.2.1 Call the required modules
3.2.2 Performance of the project
3.3 COVID-19 data processing
3.3.1 Overview of the JSON file
3.3.2 Overview of the data in Excel
3.3.3 JSON to excel
3.3.4 Excel data processing
4 Our Major Advances
4.1 Construction of a dataset
4.2 Three aspects for analyses
4.3 Construction of Bayesian network
4.4 Programming of the project model
4.5 Interpretation of programming code
4.6 Final Validation of programming results
5 Unresolved Issues
Chapter 4. Health Management Based on Artificial Intelligence
1 Introduction
1.1 Research backgrounds
1.2 Research topics and trends
1.3 Concretizing data problems
1.4 The main idea of this chapter
2 Design of the System
2.1 Principle of DQN algorithm
2.2 Consideration of the interface
2.3 Python and tkinter library description
3 The Programming Work
3.1 The Complete Code
3.2 Code interpretation
3.3 Summary of the code
4 Code debugging process
4.1 Control placement problem
4.2 Special control positions to be set
4.3 The lines to be set in the right place
4.4 The type of input field to be comparable
4.5 Corresponding results from the input data
5 Summary of the Key Points
5.1 New rounds of exploration
5.2 Conditions for planning consideration
6 The Concluding Remarks
Chapter 5. Design of Household Epidemic Prevention Systems
1 Introduction
1.1 Motivation of the project
1.2 The new stage of epidemic prevention
1.3 Household epidemic prevention system
2 Principles to design the system
2.1 Analyses of family epidemic prevention
2.2 The modules to be equipped in the system
2.2.1 Electronic prototyping tools
2.2.2 The characteristics of Arduino
2.3 Our Schemes to be equipe these modules
2.3.1 The tracking module scheme
2.3.2 Obstacle avoidance module scheme
2.3.3 Temperature and humidity measurement
3 Algorithm for Remote Care of Patients
3.1 Sensor-based care algorithm
3.2 Utilization of environmental information
3.3 Technologies based on visual perception
3.4 Implementation of the considered project
4 Unresolved Issues
Chapter 6. Intelligent Search and Reasoning for Outbreaks
1 Introduction
1.1 Research backgrounds
1.2 Research topics and trends
1.3 Research objectives of this chapter
1.4 Main ideas and research methods
2 Python crawler Technology
2.1 Big data processing
2.2 Python crawler Basics
2.3 Python data analysis library
3 Model and algorithm selection
3.1 Least squares vector machine
3.2 Construction of the learning system
3.3 Least squares support vector machine
4 Code implementation phase
4.1 Baidu search index crawler
4.2 Building an early warning model
4.3 Experimental results and analysis
5 Unresolved Issues
Chapter 7. Video Recognition of Risk in New Crown Infection
1 Introduction
1.1 Research backgrounds
1.2 Research topics and trends
2 Modelling Approaches
2.1 Requirement analyses
2.1.1 Video identification technology
2.1.2 Feasibility of the facial recognition
2.1.3 Our methods for image processing
2.2 Developing the running environment
2.1.1 Development environment
2.2.2 Choice of development tools
2.2.3 Installing the runtime environment
2.3 Mechanisms of the learning system
2.3.1 Extract the features of our faces
2.3.2 Risk levels of different populations
2.4 Function library configuration
2.4.1 Knowledge of deep learning
2.4.2 Introduction of the torch library
3 Technical Implementation
3.1 Face detection algorithm
3.1.1 Face-related research
3.1.2 The YOLOface algorithm
3.1.2 Non-maximum suppression
3.2 The Programming modules and code
4 The Major Results
4.1 Interface of the learning system
4.2 The risk degree of COVID-19 infection
5 Concluding Remarks
Chapter 8. Hyperglycemia Risk Warning and Sugar Control
1 Introduction
1.1 Research backgrounds
1.2 The focus of the project
1.3 Organization of this chapter
2 Acquisition of the Rough Data
2.1 Postprandial blood glucose data
2.2 Medication and exercise data
3 Mathematical Modeling Process
3.1 Basic modeling ideas
3.2 Linear trend equation
3.3 Parameter optimization technique
4 Blood Sugar Change Experiment
4.1 Programming ideas
4.2 The programming work
4.3 Experiments with the data
5 Research on Sugar Control Strategy
5.1 Hyperglycemia risk warning
5.1.1 Early warning Methods
5.1.2 Programming implementation
5.1.3 Early warning of hyperglycemia risk
5.2 The Choice of Sugar Control Strategy
5.2.1 Decision-making Method
5.2.2 Programming implementation
5.2.3 Active exercise intervention strategy
6 Concluding Remarks
References

IAVAE (London, United Kingdom) unless otherwise stated