PREDICTING THE COMPLICATIONS FOR LIPID

PATIENTS USING MACHINE LEARNING

GUIDED BY, PRESENTED BY,

MR.M.PRAKASH KUMAR M.E., SOWMIKA.B (611720104071)
AP/CSE SRILEKHA.R (611720104072)

SRIMATHI.M (611720104073)
VISALI.K.J (611720104086)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

 OBJECTIVE

The Main Objective of this project is to predicting the lipids in people

by using web application based flask Framework.
 Provides new approach to concealed patterns in the data.
 Helps avoid human biasness.
 Reduce the cost of medical tests.
 ABSTRACT

 This study proposes a Flask API-based system to forecast diseases,

focusing on lipids disease prediction.
 Various machine learning models like Logistic Regression, K-Nearest
Neighbors, Support Vector Machine, Decision Trees, Random Forest,
Gradient Boosting Classifier, and XG Boost are utilized.
 Unlike many existing models that focus on one disease at a time, this
approach considers all contributing factors, enabling a comprehensive
analysis of potential impacts.
 INTRODUCTION

 In recent years, the integration of AI and healthcare has shown promising

potential, particularly in the field of lipids diseases.
 The liver, the largest organ in the body, plays a crucial role in digestion and
detoxification. Viral infections and alcohol consumption can lead to liver
damage and life-threatening conditions such as cirrhosis and diabetes.
 Lipids diseases, including hepatitis, cirrhosis, and diabetes, are major
global health concerns, causing approximately 2 million deaths annually.
 Cirrhosis alone accounted for one million deaths in 2010, with millions
more affected by lipid-related illnesses.
 PROBLEM STATEMENT
 The problem statement involves developing a machine learning model to
predict complications in patients with lipid disorders.
 This could include the identification of risk factors, early detection of
potential complications,
 And providing actionable insights for healthcare professionals to intervene
and manage lipid-related health issues effectively.
 EXISTING SYSTEM
 Traditional disease risk prediction involves supervised machine learning
algorithms trained on labeled data to classify patients into high-risk and
low-risk groups.
 However, existing models often lack accuracy. Machine learning methods
have been extensively used to develop decision support systems for disease
prediction using medical datasets.
 Lipids disease, a type of lipid-related damage or illness, encompasses over
a hundred variations.
 In this process, the Naïve Bayes algorithm is employed for classification
tasks, aiming to find conventional patterns in disease categorization.
 DISADVANTAGES

 Existing system performance is very low.

 Computationally complex.
 More Execution time required to generate results.
 Naïve bayes classifier cannot give a accurate value.
 PROPOSED SYSTEM

The proposed system utilizes patient lipids condition data for improved
accuracy, preprocessing the medical reports and applying machine learning
algorithms like SVM.
 Various packages and libraries are imported, and SVM employs image
classification and recognition methods, consisting of multiple layers of
artificial neurons.
 This aids ophthalmologists in identifying lipids retinopathy and
detecting abnormalities caused by lesions. Machine learning enhances the
effectiveness of diabetic retinopathy detection.
 ADVANTAGES

 Naive Bayes Algorithm gives high Accuracy with the help of

SVM.
 Computationally less complex
 Predict the disease in less time.
 SYSTEM ARCHITECTURE

DATA COLLECTION DATA

EDA
PREPROCESSING

MODEL
KNN svm
IMPLEMENTATION

PREDICTION PREDICTION
Input values
GRAPH RESULT
 SYSTEM MODULES

Module 1: Dataset collection

Module 2: Pre-Processing
Module 3: Data Visualization
Module 4: Model Implementation
Module 5: Prediction
 MODULE DESCRIPTION

 Dataset Collection :The research used data from the UCI Machine
Learning Repository after cleaning and preprocessing.
 The dataset includes lipids patient data, with a majority of male patients.
Eleven parameters were initially included, with ten chosen for analysis and
one as the target class.
 Utilizing UCI repository data ensures accuracy. Effective data collection
and storage are crucial for developing practical AI and machine learning
solutions for specific business problems.
 MODULE DESCRIPTION

Data Pre-Processing:
We analyzed lipids patient’s data whereas samples are lipids patient and
samples are non-lipids patients.
Moreover, from the lipids patient’s dataset are male samples and female
samples were taken for analysis.
The heatmap appear to have some correlated parameters.
Some of these columns have a low correlation.
 Therefore, we omitted some of the features for better prediction of
lipids disease.
 MODULE DESCRIPTION

Data Visualization:

 Data visualization is the representation of data through use of

common graphics, such as charts, plots, info graphics, and even
animations.
 Here, we use the EDA concept to visualize the data based on the
diabetics disease Patient of the System.
 These visual displays of information communicate complex data
relationships and data-driven insights in a way that is easy to
understand.
 MODULE DESCRIPTION
Model Implementation:
 Here, we use the Machine Learning model to predict the result of the
System. Machine Learning models can be understood as a program that has
been trained to find patterns within new data and make predictions. These
models are represented as a mathematical function that takes requests in
the form of input data, makes predictions on input data, and then provides
an output in response.
 Here we use the lot of the machine learning algorithm like Logistic
regression, Random forests, Decision Tree, SVM, kNN, XG Boost,
Gradient Boosting Classifier.
 Logistic regression can deal with any number of numerical as well as absolute
factors. Strategic Regression processes the connection between the element
factors by surveying probabilities (p) utilizing an underlying logistic function.
 Random forests or random decision forests are an ensemble learning technique
for classification, regression and different assignments that works by
developing a huge number of decision trees at training time and yielding the
class that is the method of the classes.
 Decision Tree calculation has a place with the supervised learning algorithms .
In contrast to other supervised learning algorithms, a decision tree algorithm
can be utilized for taking care of regression and classification issues as well.
 SVM is a supervised learning calculation. It can utilize for both grouping or relapse
issues however generally it is utilized in characterization issues.
 KNN is one of the most fundamental occasion-based classification algorithms in
Machine Learning.. A KNN sorts an example to the class that is most decided among K
neighboring. K is a limitation for adjusting the classification algorithms.
 XG Boost The XGBoost or Extreme Gradient Boosting algorithm is a decision tree
based machine learning algorithm which uses a process called boosting to help improve
performance. Since it’s introduction, it’s become of one of the most effective machine
learning algorithms and regularly produces results that outperform most other
algorithms, such as logistic regression, the random forest model and regular decision
trees.
XGBoost has frameworks for various languages, including Python, and
it integrates nicely with the commonly used scikit-learn machine learning
framework used by Python data scientists. It can be used to solve
classification and regression problems, so is suitable for the vast majority
of common data science challenges.
Gradient Boosting Classifier : Gradient Boosting is a functional
gradient algorithm that repeatedly selects a function that leads in the
direction of a weak hypothesis or negative gradient so that it can
minimize a loss function. Gradient boosting classifier combines several
weak learning models to produce a powerful predicting model.
 SOFTWARE REQUIREMENTS

Operating System : Windows 7 or later

Simulation tool : Python -3.6, flask
Documentation : Ms-office
 HARDWARE REQUIREMENTS

RAM : 2 GB.
Processor : I5 and Above
Hard disk space : 2 GB (minimum) free space available.
Screen resolution: 1024 x 768 or higher.
HOME PAGE CODING
<!DOCTYPE html>
<html lang="en">
<head>
<title>Webpage Design</title>
<link rel="stylesheet" href="style.css">
</head>
<style>{
margin: 0;
padding: 0;
}.main{
width: 100%;
background: linear-gradient(to top,
rgba(0,0,0,0.5)50%,rgba(0,0,0,0.5)50%), url(1.jpg);
background-position: center;
background-size: cover;
height: 100vh;
}
<div class="main">
<div class="navbar“>
<div class="icon">
<h2 class="logo"> LIPID_DETECTION</h2>
</div>
<div class="menu">
 <li><a href="#">ABOUT</a></li>
<li><a href="/register">REGISTER</a></li>
<li><a href="/login">LOGIN</a></li>
<li><a href="#">CONTACT</a></li>
</ul>
</div>
<div class="search">
<input class="srch" type="search" name="" placeholder="Type To
text">
<a href="#"> <button class="btn">Search</button></a>
</div>
</div>
<div class="content">
<h1>WELCOME <br><span>LIPID_DETECTION</span> <br>Course</h1>
<p class="par">
<br> a quae totam ipsa illum minus laudantium?</p>
<button class="cn"><a href="#">JOIN US</a></button>
<div class="icons">
<a href="#"><ion-icon name="logo-facebook"></ion-icon></a>
<a href="#"><ion-icon name="logo-instagram"></ion-icon></a>
<a href="#"><ion-icon name="logo-twitter"></ion-icon></a>
<a href="#"><ion-icon name="logo-google"></ion-icon></a>
<a href="#"><ion-icon name="logo-skype"></ion-icon></a>
</div> </div>
<script src="https://unpkg.com/[email protected]/dist/ionicons.js">
</script>
</body>
</html>
 CONCLUSION
 The aim is to create a diagnosis system for chorionic lipids infection using five
supervised machine learning classifiers.
 LR classifier achieved 75% accuracy in predicting lipids disease. The application
predicts lipids infection early and alerts health conditions.
 Exploring more algorithms can enhance accuracy. Trained models are saved as
Pickle files and integrated into a Flask web page.
 REFERENCES
[1] Dietterich TG.(2000), Ensemble methods in machine learning. In:
Proceedings of Multiple Classifier System, vol. 1857, Springer (2020), pp. 1–15.
[2] Zhi-Hua Zhou, Ensemble Methods: Foundations and Algorithms, Machine
Learning & Pattern Recognition Series, 2022.
 [3] Yongjun Piao, Minghao Piao, Keun Ho Ryu, Multiclass detection
classification using a feature subsetbased ensemble from microRNA expression
profiles, Computers in Biology and Medicine 80 (2021) 39– 44.
[4] ManjeevanSeera, Chee Peng Lim, A hybrid intelligent system for medical
data classification, Expert Systems with Applications 41 (2022) 2239– 2249.
[5] Michael J Sorich. An intelligent model for daibetics disease diagnosis.
Artificial Intelligence in Medicine 2021..
THANK YOU