Download Report on Python Project and more Lab Reports Computer Vision in PDF only on Docsity!
Chapter 1
Introduction
1.1 Introduction
Health care practices involve collecting all kinds of patient data which would help the doctor correctly diagnose the condition the subject is likely suffering from. This data could be everything from the simple symptoms observed by the subject, initial diagnosis by a physician or a detailed test result from a lab. Thus far this data is only utilized for subjective analysis by a doctor who then ascertains the disease in play using his/her personal medical expertise. We posit that there is definite potential for application of data mining routines on this rich reserve of patient data.
1.2 Motivation
We have always been in a situation where we have medical test reports with us but are unable to visit the doctor due to some or the other reason, but we are still curious to know what the reports mean. These conditions have led us to idea of an automated medical diagnosis tool which help in determining the disease of the person. We are planning to restrict our focus to diabetes disease as it is one of the most common diseases the people face now adays due to their lifestyle.
1.3 Problem Definition
There is a need for a system that can help prevent or detect early signs of diseases. Our main aim of this exploration is to build a web application using various Machine Learning Algorithms which will be used for early detection or prediction of diseases. The main problem will be to develop a system which will provide personalized diagnosis in the form of diet and various exercises to the users if the results are positive for the user.
1.4 Objectives of Project
With the advance in technology and methodology people have begun automating tasks in order to save time and energy. Most of these tasks were repetitive or simple, but nowadays we have begun automating processes which seemed impossible to do so before. Techniques in data mining has led to the case of extracting not only information, but patterns and conclusions from data which previously had to be done manually.
By using a technique of our own, Myfitness, a web application will automate the process of medical diagnosis for users who are in need of it where it isn't possible to consult a doctor. Myfitness will act as the first line of diagnosis.
1.5 Scope of the project
As the lifestyle of people is very unhealthy most of the people are getting diabetes. Diabetes is one of the major causes of deaths in India. We posit that there is a definite need of an application which is able to detect or prevent early sign of diabetes. The system will also be able to provide personalized preliminary treatment in the form of diet and exercise.
1.6 Application of project
The majority of AI use-cases for managing diabetes appear to fall into two major categories: I. Glucose Monitoring Systems: Machine learning algorithms help automate the process of monitoring blood sugar levels and recommend adjustments in care. II. Nutrition Coaching: To help recommend meal options based on the specific diet criteria of the user.
1.7 Project Impact Analysis
As per statistics of population in India there are so many people who have diabetes and there is definite need of a system which will help the people efficiently manage diabetes. Best ways to manage diabetes is to have a controlled diet with regular exercise. Our System will help those people who are prediabetic or having diabetes. The system is giving personalized diagnosis to people in the form of exercise plan and diet plan.
1.8 Expected Outcome
The Project aims to predict whether patient is suffering from diabetes or not and recommends preliminary treatment in the form of personalized exercise plan and diet plan.
Chapter 2
Proposed Work and Literature Review
2.1 Literature Survey
[1] Deeraj Shetty ,Kishor Rit , Sohail Shaikh and Nikita Patil .“ Diabetes Disease
Prediction Using Data Mining ”. International Conference on Innovations in Information, IEEE(2017). This paper discusses a system which uses machine learning model to cluster the dataset of diabetes and find out the accuracy of the model. It mainly focuses on machine learning algorithm deployment of classification. The proposed system uses machine learning model as prediction and web technologies as user interface. In the model selection they got KNN as best model which works on diabetes datasets. Primarily they are maintaining the database for the patient access. Their examination concentrates on this part of Medical conclusion learning design through the gathered data of diabetes and to create smart therapeutic choice emotionally supportive network to help the physicians. [2] Samrat Kumar Dey, Ashraf Hossain and Md. Mahbubur Rahman. “Implementation of a Web Application to Predict Diabetes Disease: An Approach Using Machine Learning Algorithm”. 21st International Conference of Computer and Information Technology (ICCIT) 2018. This study concentrates on various diseases and their prediction with the help of intelligent system. In this paper they have collaborated various disease prediction with web technologies. They have made a web portal for patients to diagnose their disease. In this they proposed work was done on data collection on Diabetes from various reports then preprocessing is done using machine learning techniques. Their Our main aim of this exploration is to build a web application based on the higher prediction accuracy of some powerful machine learning algorithm. We have used a benchmark dataset namely Pima Indian which is capable of predicting the onset of diabetes based on diagnostics manner. With an accuracy of 82.35% prediction rate Artificial Neural Network (ANN) shows a significant improvement of accuracy which drives us to develop an nteractive Web Application for Diabetes Prediction.
[3] Pahulpreet Singh Kohli and Shriya Arora. “Application of Machine Learning in Disease Prediction”. 4th International Conference on Computing Communication and Automation (ICCCA) 2018. This paper focuses on prediction of various deadly diseases such as Heart, Breast and Diabetes. They have collected datasets from Kaggle and used Data Mining for removing missing values from it. They have proposed all types of supervised machine algorithms for classification. As they have predicted various deadly disease therefore the accuracy was checked in all algorithms. The feature selection for each dataset was accomplished by backward modeling using the p-value test. The results of the study strengthen the idea of the application of machine learning in early detection of diseases. [4] V.Swathi Lakshmi, V.Nithya, K.Sripriya , C. Preethi and K. Logeshwari. “Prediction of Diabetes Patient Stage Using Ontology Based Machine Learning System”. Proceeding of International Conference on Systems Computation Automation and Networking 2019. This paper proposed the work methodology of automated system working on the intelligence system for predicting diabetes disease. l related problems. In this paper a proposed method to identify patient with diabetes disease risk level is indentified. In this work diabetes patient risk level is been detected by using ontology and machine learning technique. Ontology holds disease symptoms, causes and treatments. In machine learning, naïve base algorithm is used to make decision on patient record also it defines possibilities of risk level. The proposed algorithm will be evaluated against the following metrics namely confusion matrix, precision level, mean and this proposed work is found to have better prediction level when compared with existing work. [5] Santosh Rani and Dr. Sandeep Kautish. “Association clustering and Time Series based data mining in Continuous data for Diabetes Prediction”. Proceedings of the Second International Conference on Intelligent Computing and Control Systems (ICICCS 2018) IEEE Xplore****. The project proposed that large amount of health related data is being produced in various levels of health system. Due to the size of the data it will be difficult to process the data and then extract the analysis. But with the use of different machine learning based approaches data can be processed. Machine
2.2 Proposed Work
In this project we have divided the entire project into four major modules. The Modules are namely Registration, Diagnosis, Diagnosis History and Visualization modules Modules
1. Registration Module The Registration module is used for registration of new users. The user needs to login after the registration to use the system. 2.Diagnosis Module This module is the brains of the system which is responsible for identifying the diseases based on the symptoms. It takes input from the user with the help of a simple form and uses the knowledgebase containing information about the diseases and their symptoms and finally provides the result using the K-nearest neighbor algorithm. 3. Exercise Module In this module the exercise plan is generated whenever the user fills the analysis form. The exercise plan generated is stored in the database with date and time so that the user can refer to that in the future as well. 4. Diet Module In this module the diet plan is generated whenever the user fills the analysis form. The diet plan generated is stored in the database with date and time so that the user can refer to that in the future as well. 5. Visualization module The module provides visualizations based on the diagnosis data collected. These visualizations will be in the form of bar graphs. We provide the visualization of both the input parameters as well as the visualization of the results generated.
6. Appointment Module This module deals with the appointments. The patients will be able to book appointments with the doctors and the doctors will be able to view the booked appointments.
Performance Since it is a web application it always requires an active internet connection. The application is very light so once the application gets installed, all data can be downloaded and installed at once from the server. Supportability The web application is light weight which enhances its performance and it is platform independent. In future there will be a need for developing android and ios versions of this application. Interface The web application has a very interactive interface. The interface is very user friendly. Legal All copyrights and legal expects are preserved by the non-profitable organization and all terms and conditions are needed to be followed by the users.
3.1.4 Gap Specification
In the current system the patients can book appointment with the doctor and the details of the appointment are displayed to both the patients and the doctor. But there is a possibility that some times the doctor might not be available for that appointment. So a feature can be added which will be used to confirm the appointments. Once the appointments are booked by the patients the appointments are needed to be confirmed by the doctor, and on confirmation only the patients should visit the doctor in person. The admin is also having limited functions in the current system so a feature can be added so that he can also remove the doctors from the system, currently this operation has to be done by the database administrator. In the current system there is no option of live chat with the Doctor. A feature can be added wherein the patients can interact with the doctors in the real time in form of chat or Video Calling. Currently the patients can interact with the doctors only through the Mobile number provided in the details section.
The current system only deals with diabetes disease, in future a system can be developed which is able to handle multiple diseases and provide efficient preliminary treatment to its users.
3.1.5 Current System
The current system are all web based and mobile applications which provide similar features at high cost. Some applications provides recommendation features for free but the features are not up to the mark. Some applications provides good content but they lack interactive interface. In the current system the main problem is that the users need to be well versed with the technological aspects in order to handle the system. Our System provides an interactive web design which is easy to use and understand and handle even by the naïve users. The current system do not easily provide the details of the doctors. The users need to search the web page minutely to find out the details of the doctor. Our system provides transparency between the patient and the doctor by providing the details easily to the patient.
3.2 Feasibility Study(FAD)
Wt User 1 User 2 Description Writing new application in- house using new company standard Java and SQL Server database Rewrite or reuse current in- house applications. Operational Feasibility 30% It^ fully^ supports^ user- required functionality Supports Services requirement. Current business process would have to be modified to take advantage of software. Score: 60 Score: 75 Technical Feasibility 30% Solution requires writing application in PHP. It should be relatively easy to find programmers with Java and SQL experience. Management is more concerned about acquisition of recent technology. SQL Server is the current company standard for database. Score: 85 Score: 70 Schedule 20% We have ample amount of 3 – 5 months Feasibility time to develop the project.
Figure.3.3.1: Spiral Model The spiral model is similar to the incremental model, with more emphasis placed on risk analysis. The spiral model has five phases: Communication, Planning, Modelling, Construction and deployment. A software project repeatedly passes through these phases in iterations. Communication: Tasks required to establish effective communication between developer and customer. Planning:^ Tasks^ required^ to^ define^ resources,^ timelines,^ and^ other^ project^ related information. Modelling:^ Tasks required in building one or more representations of the application. Construction:^ Tasks^ required to construct, test, install Deployment :^ Tasks required to deliver the software, getting feedbacks etc. The baseline spiral in the planning phase. requirements are gathered and risk is asserted Each Subsequent spiral builds on the baseline spiral. In case of My Fitness, the aim of each spiral would
both algorithm and the knowledge base until it is in the satisfactory accuracy range .each iteration will also improve the graph in terms of representing the different disease Each iteration will represent the different disease case based on location and year.
3 .3. 2 Technology
Technology used : PHP v7.0.0^ :Hypertext Preprocessor^ (or simply^ PHP) is a^ general-purpose programming language originally designed for web development. It was originally created by Rasmus Lerdorf in 1994; the PHP reference implementation is now produced by The PHP Group. PHP originally stood for Personal Home Page , but it now stands for the recursive initialism PHP: Hypertext Preprocessor. XAMPP^ v5.6.31^ :^ XAMPP^ is a^ free and open-source^ cross-platform^ web^ server^ solution stack package developed by Apache Friends, consisting mainly of the Apache HTTP Server, MariaDB database, and interpreters for scripts written in the PHP and Perl programming languages. Since most actual web server deployments use the same components as XAMPP, it makes transitioning from a local test server to a live server possible BOOTSTRAP v 3.3.6 Bootstrap is a free and open-source CSS framework directed at responsive, mobile-first front-end web development. It contains CSS- and (optionally) JavaScript-based design templates for typography, forms, buttons, navigation and other interface components. MYSQL v8.0 : MySQL is free and open-source software under the terms of the GNU General Public License, and is also available under a variety of proprietary licenses. MySQL was owned and sponsored by the Swedish company MySQL AB, which was bought by Sun Microsystems (now Oracle Corporation).[8]^ In 2010, when Oracle acquired Sun, Widenius forked the open-source MySQL project to create MariaDB.
Figure.3.4.1 Gantt Chart The fig 3.4.1 is depicting the Gantt Chart in which the actual project planning of our groupis presented for a period of 8 months from July 2019 to February 2020
Figure 3.5.1 Level 0 DFD The Doctor can perform operations like Check booked appointments and the Patient can perform operations like Check Visualization, Fill Analysis form, Check History, Book Appointment, View Booked Appointment, View Exercise plan and View Diet Plan.
Level 1 DFD
Figure 3.5.2 depicts the level 1 DFD of Admin in which you can see the detailed operations which the admin can perform. Figure 3.5.2 Level 1 DFD (Admin)
When the admin will login to the system then the system will first verify the password and then the admin can perform operations like Adding the details of new Doctor in the system and also View the list of Doctors already added in the system. Figure 3.5.3 depicts the level 1 DFD of Doctor in which you can see the detailed operations which the Doctor can perform. Figure 3.5.3 Level 1 DFD (Doctor) When the Doctor will login to the system then the system will first verify the password and then the Doctor can perform operations like viewing the Booked appointments. Figure 3.5.4 depicts the level 1 DFD of Patient in which you can see the detailed operations which the Patient can perform. When the Patient will login to the system then the system will first verify the password and then the Patient will be able to perform different operations.
