Bapuji Institute of Engineering and Technology
Department of Computer Science & Engineering
PROJECT WORK PHASE – 2 (18CSP83)
“LIVER CIRRHOSIS PREDICTION USING MACHINE LEARNING”
COORDINATOR:
UNDER THE GUIDANCE OF: Dr. NIRMALA C R Ph.D
Prof. ANU C S MTech (PhD)
HEAD OF DEPARTMENT OF COMPUTER
DEPARTMENT OF COMPUTER SCIENCE
AND ENGINEERING
Presented by: SCIENCE AND ENGINEERING
& DEAN PLACEMENT
AKASH M J
4BD20CS007
ABHISHEK I HANCHINAMANI
4BD20CS003
VENKATESHA H
4BD19CS411
KARTHIK S BADIGER
Contents
1. Abstract
2. Introduction
3. Literature Survey
4. Existing System
5. Problem Statement
6. Proposed System
7. Objectives
8. System Requirements and Specification
9. Methodology
10.Dataset
11. Samples
12.Conclusion
Abstract
This research presents a machine learning approach for the prediction of liver cirrhosis, a critical medical
condition with significant public health implications. Leveraging a comprehensive dataset comprising
clinical and demographic features, our model integrates advanced algorithms to accurately forecast the
likelihood of cirrhosis development. Key features include liver function tests, albumin and globulin ratio,
total and direct bilirubin, presence of edema, spiders etc. The model demonstrates promising predictive
capabilities, offering a valuable tool for early detection and intervention. The findings contribute to the
growing field of predictive healthcare analytics, fostering proactive strategies for managing and mitigating
liver cirrhosis risks.
Introduction
Liver cirrhosis is a late stage of scarring (fibrosis) of the liver caused by many forms of liver diseases and
conditions, such as hepatitis and chronic alcoholism. Each time your liver is injured, it tries to repair itself. In
the process, scar tissue forms. As the cirrhosis progresses, more and more scar tissue forms, making it difficult
for the liver to function.
Causes:
Chronic Alcoholism: Long-term excessive alcohol consumption is a leading cause.
Chronic Viral Hepatitis: Hepatitis B, C, and D can cause cirrhosis.
Non-Alcoholic Fatty Liver Disease (NAFLD): Accumulation of fat in the liver.
Autoimmune Hepatitis: The immune system attacks the liver.
Genetic Diseases: Such as cystic fibrosis or Wilson's disease.
Symptoms:
Early stages may be asymptomatic.
As cirrhosis progresses, symptoms may include fatigue, weakness, easy bruising, swelling in the
legs and abdomen, and confusion.
Complications:
Cirrhosis can lead to complications such as portal hypertension, which can cause varices (enlarged
veins) and ascites (fluid buildup in the abdomen).
Liver cancer (hepatocellular carcinoma) is a risk in advanced cirrhosis.
Literature Review
Author Title Methodology Merits
Jianxia Wen, Xing Research Progress and Accuracy 55%.
Support Vector Machine.
Chen et.al 2022 Treatment Status of Liver
Cirrhosis with
Hypoproteinemia
Manjula Devarakonda Health Care Automation Random Forest Accuracy 63%
Venkata, Sumalatha
Lingamgunta et.al 2022
C.Geetha, Evaluation based Approaches Support Vector Machine. Variable Accuracy.
Dr.AR.Arunachalam for Liver Disease Prediction
2021 using Machine Learning
Algorithms
Md. Fazle Rabbi, S.M. Prediction of Liver Disorder Logistic regression, Decision Accuracy 70 %
Mahedy Hasan et.al Using Machine Learning Tree, Random Forest
2020 Algorithm
Literature summary
The literature reviews discussed various aspects of liver disease prediction, treatment, and healthcare
automation methodologies using machine learning algorithms. They provided insights into the
research progress and treatment status of liver cirrhosis, with a particular focus on hypoproteinemia's
role in its progression and management. Additionally, the reviews explored healthcare automation
methodologies, highlighting the use of machine learning techniques such as Support Vector Machine
and Random Forest for predictive modeling, with reported accuracies of 55% and 63%, respectively.
Furthermore, evaluation-based approaches for liver disease prediction using machine learning
algorithms were discussed, including Logistic Regression, Decision Tree, and Random Forest, with
emphasis on their effectiveness in predicting liver disorders.
Gaps Identified
Accuracy of the earlier systems are less.
Cost of diagnosis of liver cirrhosis is high.
The result of diagnosis are general (not specific).
Have low accuracy in predicting the stage.
EXISTING SYSTEM
The existing system for liver cirrhosis assessment often relies on traditional diagnostic methods, including
clinical evaluation, liver function tests, imaging studies, and invasive procedures like liver biopsy. While
these approaches are fundamental, they may have limitations in terms of early detection and widespread
applicability. Machine learning has emerged as a complementary tool to enhance the existing system. By
leveraging vast datasets and advanced algorithms, ML models can analyze diverse patient data to identify
patterns and subtle indicators associated with liver cirrhosis. This offers the potential for earlier and more
accurate predictions, enabling timely interventions and personalized healthcare strategies.
Problem Statement
The liver cirrhosis has become a common disease around the world. The death rate due to the disease is
becoming alarming. Early detection of the disease may reduce the complication of the disease misfortune on
patients. The ease of use of inventive technologies such as the one anticipated in this research may help in
alleviating the troubles of holdup in the uncovering and treatment of liver cirrhosis. The Machine learning
techniques are used to predict whether patient is positive or negative for the disease. One more significant
drive behind this is Predicting Stage also in which stage patient is there.
PROPOSED SYSTEM
The proposed system for liver cirrhosis prediction through machine learning is designed to revolutionize current
diagnostic approaches. This system integrates a diverse array of patient data to develop a robust predictive model
using advanced machine learning algorithms. The primary objective is to predict the liver cirrhosis along with its
stages. Overall, this proposed system represents a significant step toward more accurate and user-friendly liver
cirrhosis prediction system.
Objectives
1. To design a model to analyse various patient data and predict the presence of liver cirrhosis.
2. To design a model to predict the stage of liver cirrhosis using Ensemble Classification algorithm
for model creation.
3. To design front End application using flask for user usage.
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
The software required for the development of this project is:
1. Application Required : PyCharm.
2. Operating System : Windows 10(and other higher version)
3. Programming Language : Python
HARDWARE REQUIREMENTS:
The hardware required for the development of this project is:
4. Processor : Intel Core i3 Processor or higher.
5. System Type : 64-Bit Operating System.
6. RAM : 8 GB RAM Minimum.
METHODOLOGY
Data Pre-
Dataset Testing data
processing
Dataset Description
Dataset Pre-Processing
Training data
Testing data
Training data Classification
Classification model model
Cirrhosis prediction
Stage Cirrhosis
Prediction Prediction
DATASET
SAMPLE
Conclusion
The integration of machine learning into liver cirrhosis prediction holds significant promise for advancing
healthcare practices. By focusing on early detection, improving diagnostic accuracy, and tailoring
interventions based on individualized risk assessments, this approach has the potential to enhance patient
outcomes and streamline healthcare workflows. The development of predictive models holds the potential
to revolutionize clinical practice by enabling healthcare providers to identify high-risk individuals, tailor
treatment plans, and improve patient outcomes.
THANK YOU
References
[1] Jianxia Wen, Xing Chen, Shizhang Wei et.al “Research Progress and Treatment
Status of Liver Cirrhosis with Hypoproteinemia”, Published in 2022 by
Hindawi.
[2] Manjula Devarakonda Venkata, Sumalatha Lingamgunta et.al “Health Care
Automation” Published in October 2022 by IEEE.
[3] C.Geetha, et.al “Evaluation based Approaches for Liver Disease Prediction using
Machine Learning Algorithms” Published on 27 Jan 2021 by IEEE.
[4] Md. Fazle Rabbi, S. M. Mahedy Hasan, et.al “Prediction of Liver Disorders using
Machine Learning Algorithms” Published in 2020 by IEEE.
