SlideShare a Scribd company logo
1 of 25
Presentation Outline
โ€ข Introduction
โ€ข Objectives
โ€ข System Architecture / Ideation Map
โ€ข Module Implementation
โ€ข Results and Discussions
โ€ข Conclusion
โ€ข References
Department of CSE 1
Introduction
๏‚ง Machine Learning(ML) is a branch of Artificial Itelligence(AI) &
Computer Science which focuses on the use of data and
algorithms to imitate the way that humans learn and gradually
improving accuracy
๏‚ง During an echocardiogram, a doctor will generate a real-time
image of the heart. The test monitors ultrasound, high-
frequency sound waves, which are projected through the chest
and bounce back to create an image of your heart.
๏‚ง All the patients suffered heart attacks at some point in the
past. Some are still alive and some are not. The survival and
still-alive variables, when taken together, indicate whether a
patient survived for at least one year following the heart
attack.
๏‚ง Decision Tree is the most important support tool with a tree-
like structure that models probable outcomes, cost of
resources, utilities and possible consequences of supervised
learning.
. Department of CSE 2
Introduction
๏‚ง The problem addressed by past researchers was to predict
from the other variables whether or not the patient will
survive at least one year. The most difficult part of this
problem is correctly predicting that the patient will NOT
survive. (Part of the difficulty seems to be the size of the data
set.
โ€ข Problem Statement : The classification goal is to predict still-
alive using features from 3 to 10
. Department of CSE 3
Objectives
โ€ข The objective of the project is to find how to use machine
learning techniques for analysis and making predictions using
echocardiogram dataset . To find The classification goal is to
predict still-alive using features from 3 to 10 . The survival and
still-alive variables, when taken together, indicate whether a
patient survived for at least one year following the heart
attack.
โ€ข This report examines how to use machine learning techniques
to analyze and predict using echocardiogram dataset.
Department of CSE 4
Objectives
โ€ข This data set is used for exploratory data analysis, data
visualization, dealing with missing values and classification
modelling techniques
โ€ข An ECG should be attached ensuring good tracings to facilitate
the acquisition of complete digital loops. Loops should be
examined and adjusted accordingly in order to ensure a clear
representation of the image acquired.
โ€ข Thetest is useful for diagnosing and monitoring heart problems and
creating treatment plans.
. Department of CSE 5
Objectives
โ€ข There are three main types of echocardiogram tests that a doctor may
perform.
โ€ข A transthoracic echocardiogram is noninvasive. It uses a transducer moved
across your chest to produce the heart image.
โ€ข A transesophageal echo test is performed with a tube transducer in your
throat. This helps to view the heart from a different angle.
โ€ข Last, a stress echocardiogram occurs during exercise on a treadmill or bike.
This is to monitor your heartโ€™s response to physical activity.
Department of CSE 6
System Architecture / Ideation Map
Import all the
necessary Module
s(numpy,pandas,
matplotlib etc.)
Importing the
DataSet (CSV) file
to Jupiter Notebook
Analyse the
Result
Running the
TestCases of the
program
Performing
Exploring Data
Analysis(EDA)
Using Random
Forest Classification
Method
For prediction
Getting Dummies of
the Data
. Department of CSE 7
Project Implementation
โ€ข In this project I will demonstrate how to predict still alive from
3 to 10 in Python using the following steps:
โ€ข Data preparation
โ€ข Exploratory Data Analysis
โ€ข Analyzing the correlation to numerical features
โ€ข Encode categorical features
โ€ข Using Machine Learning Decision Treeโ€™s echocardiogram
Algorithm for getting output
. Department of CSE 8
Project Implementation
โ€ข Hardware & Software Requirements:
โ€ข 4GB RAM
โ€ข Operating System
โ€ข Windows 10,Core i5
โ€ข Anaconda Navigator along with Python 3.8
โ€ข Construction of the Project:The dataset used for this
project relates to direct marketing campaigns of a
Portuguese banking institution. The purpose of building
the models is to predict whether the client will subscribe
for a term deposit. In order to accomplish the project
problem, the proposed solution can be described in
followingsteps:
. Department of CSE 9
Project Implementation
โ€ข Step.1: identifying the problem, collecting, describing the dataset
โ€ข Step.2: Exploring, data cleaning and preparation
โ€ข Step.3: Building the models
โ€ข Step.4: Test and evaluate the performance of the models
โ€ข Metrics : The evaluation metrics proposed are appropriate given
the context of the data, the problem statement, and the intended
solution. Theresulting image of an echocardiogram can show a big picture image of
heart health, function, and strength. For example, the test can show if the heart is
enlarged or has thickened walls. Walls thicker than 1.5cm are considered abnormal.
They may indicate high blood pressure and weak or damaged valves. An
echocardiogram can also measure if your heart is pumping enough blood through your
body.
. Department of CSE 10
Project Implementation
โ€ข survival : The number of months patient survived (has survived, if patient
is still alive). Because all the patients had their heart attacks at different
times, it is possible that some patients have survived less than one year
but they are still alive. Check the second variable to confirm this. Such
patients cannot be used for the prediction task mentioned above.
โ€ข Still-alive : A binary variable that depicts if the patient is still alive (0:
dead at end of survival period, 1: still alive)
โ€ข Age-at-heart-attack :Age in years when heart attack occurred
โ€ข Pericardial-effusion : Pericardial effusion is fluid around the heart (0: no
fluid, 1: fluid)
โ€ข Fractional-shortening : A measure of contracility around the heart, lower
numbers are increasingly abnormal
. Department of CSE 11
Project Implementation
โ€ข Epss : E-point septal separation, another measure of contractility.
Larger numbers are increasingly abnormal
โ€ข Lvdd : Left ventricular end-diastolic dimension. This is a measure of
the size of the heart at end-diastole. Large hearts tend to be sick
hearts
โ€ข Wall-motion-score : A measure of how the segments of the left
ventricle are moving
โ€ข Wall-motion-index : Equals wall-motion-score divided by number
of segments seen. Usually 12-13 segments are seen in an
echocardiogram. (Preferable to use this variable INSTEAD of the
wall motion score)
โ€ข Mult : A derivate variable (suggested that it can be ignored)
โ€ข Name : The Name of the patient (Has benn replaced with "name"
entirely)
.. Department of CSE 12
Project Implementation
โ€ข Group : Group (Has been considered meaningless and
suggested to ignore)
โ€ข Alive-at-1 : It is a Boolean-valued variable derived from the
first two attributes. (0: patient was either dead after 1 year
or had been followed for less than 1 year, 1: patient was alive
at 1 year)
. Department of CSE 13
Methodology
Data Exploration:
โ€ข The data that is used in this project originally comes from the
UCI machine learning repository. This intermediate level data
โ€ข set has 132 rows and 12 columns.
โ€ข The data set provides data that could be used for classifying if
patients will survive for at least one year after a heart attack.
โ€ข All the patients suffered heart attacks at some point in the
past. Some are still alive and some are not.
โ€ข The survival and still-alive variables, when taken together,
indicate whether a patient survived for at least one year
following the heart attack.
. Department of CSE 14
Methodology
โ€ข In this project, we are going to utilize python to develop a
predictive machine learning model! begin by loading our data
and exploring the columns.
โ€ข The problem addressed by past researchers was to predict
from the other variables whether or not the patient will
survive at least one year.
โ€ข The most difficult part of this problem is correctly predicting
that the patient will NOT survive. (Part of the difficulty seems
to be the size of the data set.)
โ€ข This data set is recommended for learning and practicing your
skills in exploratory data analysis, data visualization, dealing
with missing values and classification modelling techniques.
. Department of CSE 15
Methodology
Data reading and Preparation :-
The library can also be used for data-mining and data analysis. The main machine
learning functions that the Scikit-learn library can handle are classification, regression,
clustering, dimensionality reduction, model selection, and preprocessing.
library(ggplot2)
library(caret)
library(rpart)
library(rpart.plot)
library(dplyr)
library(tidyr)
library(DataExplorer)
library(FFTrees)
data <- read.csv("../input/echocardiogram.csv", sep =',', na.strings = c('NA','?'))
str(data)
. Department of CSE 16
Methodology
โ€ข This is the representation of the data sets that can be variables which
were using in the echocardiogram
โ€ข Here we are using all the 123 variables with the values
## 'data.frame': 133 obs. of 13 variables:
## $ survival : num 11 19 16 57 19 26 13 50 19 25 ...
## $ alive : int 0 0 0 0 1 0 0 0 0 0 ...
## $ age : num 71 72 55 60 57 68 62 60 46 54 ...
## $ pericardialeffusion : int 0 0 0 0 0 0 0 0 0 0 ...
## $ fractionalshortening: num 0.26 0.38 0.26 0.253 0.16 0.26 0.23 0.33 0.34 0.14 ...
## $ epss : num 9 6 4 12.1 22 ...
## $ lvdd : num 4.6 4.1 3.42 4.6 5.75 ...
## $ wallmotion.score : num 14 14 14 16 18 12 22.5 14 16 15.5 ...
## $ wallmotion.index : num 1 1.7 1 1.45 2.25 ...
## $ mult : num 1 0.588 1 0.788 0.571 ...
## $ name : Factor w/ 2 levels "","name": 2 2 2 2 2 2 2 2 2 2 ...
## $ group : Factor w/ 4 levels "","1","2","name": 2 2 2 2 2 2 2 2 2 2 ...
## $ aliveat1 : int 0 0 0 0 0 0 0 0 0 0 ...
. Department of CSE 17
Results and Discussion
The Decision tree we used are calculated using sklearn wrapper so can be trusted for
the model performance. Our end goal was to have a Trained model that could beat
the untuned benchmark which it did by a very fine margin. So the solution described
below is satisfactory to our initial expectations. We generated a final model with
above list of tuned parameters.
. Department of CSE 18
Results and Discussion
. Department of CSE 19
Conclusion
โ€ข Machine learning techniques are widely used data analysis
methods in various business and industrial sectors.
โ€ข The main reason for that because ML can build predictive
models to produce better predictions and achieve the desired
level of accuracy, leading to better outcomes.
โ€ข Building the models is an easy and straight forward process.
โ€ข The main challenges in data analysis are data preparation and
cleaning, the selection of appropriate models and attributes
used in their implementation.
. Department of CSE 20
Conclusion
โ€ข The aim of the project is how to use machine learning
techniques for analysis and making predictions using to
determine if the personโ€™s heart is working properly or has significant
problems.
โ€ข Heart defects are generally diagnosed in infants and children, but can be
found in adults too. Echos can even be done during pregnancy to check
for fetal issues.
โ€ข By checking size, pumping strength, and heart rate, your doctor can
find defects. A heart defect can be a hole in a chamber or improper
connections between the heart muscles and arteries.
. Department of CSE 21
References
Here we are providing the reference links which we have followed during the
project work
โ€ข https://www.kaggle.com/loganalive/echocardiogram-dataset-
uci/report#missing-value-treatment--
โ€ข https://specialistdirectinc.com/telecardiology-articles/uncategorized/a-guide
-to-understanding-echocardiogram-results/
โ€ข https://code.datasciencedojo.com/datasciencedojo/datasets/tree/master/E
chocardiogram
โ€ข https://erp.bioscientifica.com/view/journals/echo/2/1/G9.xml
โ€ข https://www.nature.com/articles/s41746-018-0065-x
. Department of CSE 22

More Related Content

Similar to 10_PPT__ML.docx

Stroke Prediction
Stroke PredictionStroke Prediction
Stroke PredictionMamathaGuntu1
ย 
Hybrid Technique for Associative Classification of Heart Diseases
Hybrid Technique for Associative Classification of Heart DiseasesHybrid Technique for Associative Classification of Heart Diseases
Hybrid Technique for Associative Classification of Heart DiseasesJagdeep Singh Malhi
ย 
IRJET- Web-based Application to Detect Heart Attack using Machine Learning
IRJET- Web-based Application to Detect Heart Attack using Machine LearningIRJET- Web-based Application to Detect Heart Attack using Machine Learning
IRJET- Web-based Application to Detect Heart Attack using Machine LearningIRJET Journal
ย 
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbh
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbhfirst review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbh
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbhmithun302002
ย 
IRJET- A System to Detect Heart Failure using Deep Learning Techniques
IRJET- A System to Detect Heart Failure using Deep Learning TechniquesIRJET- A System to Detect Heart Failure using Deep Learning Techniques
IRJET- A System to Detect Heart Failure using Deep Learning TechniquesIRJET Journal
ย 
Short story_2.pptx
Short story_2.pptxShort story_2.pptx
Short story_2.pptxChirudeepGorle
ย 
heart disease predction using machiine learning
heart disease predction using machiine learningheart disease predction using machiine learning
heart disease predction using machiine learningPODILAPRAVALLIKA0576
ย 
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhc
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhcComputational Biomedicine Lab: Current Members, pumpsandpipesmdhc
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhctmhsweb
ย 
Statistical Learning and Model Selection module 2.pptx
Statistical Learning and Model Selection module 2.pptxStatistical Learning and Model Selection module 2.pptx
Statistical Learning and Model Selection module 2.pptxnagarajan740445
ย 
Short story.pptx
Short story.pptxShort story.pptx
Short story.pptxChirudeepGorle
ย 
Lecture 11 Neurologic system disorders.pptx
Lecture 11 Neurologic system disorders.pptxLecture 11 Neurologic system disorders.pptx
Lecture 11 Neurologic system disorders.pptxMesfinShifara
ย 
Heart Disease Prediction Analysis - Sushil Gupta.pptx
Heart Disease Prediction Analysis - Sushil Gupta.pptxHeart Disease Prediction Analysis - Sushil Gupta.pptx
Heart Disease Prediction Analysis - Sushil Gupta.pptxBoston Institute of Analytics
ย 
web-application.pdf
web-application.pdfweb-application.pdf
web-application.pdfouiamouhdifa
ย 
EXPLORATORY DATA ANALYSIS
EXPLORATORY DATA ANALYSISEXPLORATORY DATA ANALYSIS
EXPLORATORY DATA ANALYSISBabasID2
ย 
Enhancement of ecg classification using ga and pso
Enhancement of ecg classification using ga and psoEnhancement of ecg classification using ga and pso
Enhancement of ecg classification using ga and psoeSAT Journals
ย 
Statistics Vocabulary Chapter 1
Statistics Vocabulary Chapter 1Statistics Vocabulary Chapter 1
Statistics Vocabulary Chapter 1Debra Wallace
ย 
CPR machine FYP Mid Progress Presentation Template.pptx
CPR machine  FYP Mid Progress Presentation Template.pptxCPR machine  FYP Mid Progress Presentation Template.pptx
CPR machine FYP Mid Progress Presentation Template.pptxyasir basheer
ย 

Similar to 10_PPT__ML.docx (20)

Stroke Prediction
Stroke PredictionStroke Prediction
Stroke Prediction
ย 
Hybrid Technique for Associative Classification of Heart Diseases
Hybrid Technique for Associative Classification of Heart DiseasesHybrid Technique for Associative Classification of Heart Diseases
Hybrid Technique for Associative Classification of Heart Diseases
ย 
IRJET- Web-based Application to Detect Heart Attack using Machine Learning
IRJET- Web-based Application to Detect Heart Attack using Machine LearningIRJET- Web-based Application to Detect Heart Attack using Machine Learning
IRJET- Web-based Application to Detect Heart Attack using Machine Learning
ย 
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbh
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbhfirst review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbh
first review.pptxgghggggvvvvbbvvvvvhhjjjbbvvvvbbbbbhhhhhhhhhbbh
ย 
IRJET- A System to Detect Heart Failure using Deep Learning Techniques
IRJET- A System to Detect Heart Failure using Deep Learning TechniquesIRJET- A System to Detect Heart Failure using Deep Learning Techniques
IRJET- A System to Detect Heart Failure using Deep Learning Techniques
ย 
Short story_2.pptx
Short story_2.pptxShort story_2.pptx
Short story_2.pptx
ย 
heart disease predction using machiine learning
heart disease predction using machiine learningheart disease predction using machiine learning
heart disease predction using machiine learning
ย 
chap1.ppt
chap1.pptchap1.ppt
chap1.ppt
ย 
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhc
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhcComputational Biomedicine Lab: Current Members, pumpsandpipesmdhc
Computational Biomedicine Lab: Current Members, pumpsandpipesmdhc
ย 
Statistical Learning and Model Selection module 2.pptx
Statistical Learning and Model Selection module 2.pptxStatistical Learning and Model Selection module 2.pptx
Statistical Learning and Model Selection module 2.pptx
ย 
Short story.pptx
Short story.pptxShort story.pptx
Short story.pptx
ย 
Presentation
PresentationPresentation
Presentation
ย 
Lecture 11 Neurologic system disorders.pptx
Lecture 11 Neurologic system disorders.pptxLecture 11 Neurologic system disorders.pptx
Lecture 11 Neurologic system disorders.pptx
ย 
PMED Transition Workshop - Machine Learning Methods to Learn Improved Electro...
PMED Transition Workshop - Machine Learning Methods to Learn Improved Electro...PMED Transition Workshop - Machine Learning Methods to Learn Improved Electro...
PMED Transition Workshop - Machine Learning Methods to Learn Improved Electro...
ย 
Heart Disease Prediction Analysis - Sushil Gupta.pptx
Heart Disease Prediction Analysis - Sushil Gupta.pptxHeart Disease Prediction Analysis - Sushil Gupta.pptx
Heart Disease Prediction Analysis - Sushil Gupta.pptx
ย 
web-application.pdf
web-application.pdfweb-application.pdf
web-application.pdf
ย 
EXPLORATORY DATA ANALYSIS
EXPLORATORY DATA ANALYSISEXPLORATORY DATA ANALYSIS
EXPLORATORY DATA ANALYSIS
ย 
Enhancement of ecg classification using ga and pso
Enhancement of ecg classification using ga and psoEnhancement of ecg classification using ga and pso
Enhancement of ecg classification using ga and pso
ย 
Statistics Vocabulary Chapter 1
Statistics Vocabulary Chapter 1Statistics Vocabulary Chapter 1
Statistics Vocabulary Chapter 1
ย 
CPR machine FYP Mid Progress Presentation Template.pptx
CPR machine  FYP Mid Progress Presentation Template.pptxCPR machine  FYP Mid Progress Presentation Template.pptx
CPR machine FYP Mid Progress Presentation Template.pptx
ย 

Recently uploaded

(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7
(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7
(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7Call Girls in Nagpur High Profile Call Girls
ย 
Kondhwa ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
Kondhwa ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...Kondhwa ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...
Kondhwa ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...tanu pandey
ย 
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...SUHANI PANDEY
ย 
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
ย 
DENR EPR Law Compliance Updates April 2024
DENR EPR Law Compliance Updates April 2024DENR EPR Law Compliance Updates April 2024
DENR EPR Law Compliance Updates April 2024itadmin50
ย 
Proposed Amendments to Chapter 15, Article X: Wetland Conservation Areas
Proposed Amendments to Chapter 15, Article X: Wetland Conservation AreasProposed Amendments to Chapter 15, Article X: Wetland Conservation Areas
Proposed Amendments to Chapter 15, Article X: Wetland Conservation Areas๐Ÿ’ฅVictoria K. Colangelo
ย 
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Moshi Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
ย 
RATING SYSTEMS- IGBC, GRIHA, LEED--.pptx
RATING  SYSTEMS- IGBC, GRIHA, LEED--.pptxRATING  SYSTEMS- IGBC, GRIHA, LEED--.pptx
RATING SYSTEMS- IGBC, GRIHA, LEED--.pptxJIT KUMAR GUPTA
ย 
RA 7942:vThe Philippine Mining Act of 1995
RA 7942:vThe Philippine Mining Act of 1995RA 7942:vThe Philippine Mining Act of 1995
RA 7942:vThe Philippine Mining Act of 1995garthraymundo123
ย 
CSR_Module5_Green Earth Initiative, Tree Planting Day
CSR_Module5_Green Earth Initiative, Tree Planting DayCSR_Module5_Green Earth Initiative, Tree Planting Day
CSR_Module5_Green Earth Initiative, Tree Planting DayGeorgeDiamandis11
ย 
Book Sex Workers Available Pune Call Girls Kondhwa 6297143586 Call Hot India...
Book Sex Workers Available Pune Call Girls Kondhwa  6297143586 Call Hot India...Book Sex Workers Available Pune Call Girls Kondhwa  6297143586 Call Hot India...
Book Sex Workers Available Pune Call Girls Kondhwa 6297143586 Call Hot India...Call Girls in Nagpur High Profile
ย 
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...Pooja Nehwal
ย 
Environmental Science - Nuclear Hazards and Us.pptx
Environmental Science - Nuclear Hazards and Us.pptxEnvironmental Science - Nuclear Hazards and Us.pptx
Environmental Science - Nuclear Hazards and Us.pptxhossanmdjobayer103
ย 
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
ย 
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...SUHANI PANDEY
ย 
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...SUHANI PANDEY
ย 
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...Anamikakaur10
ย 

Recently uploaded (20)

(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7
(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7
(NEHA) Call Girls Navi Mumbai Call Now 8250077686 Navi Mumbai Escorts 24x7
ย 
Kondhwa ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
Kondhwa ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...Kondhwa ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...
Kondhwa ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
ย 
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 8005736733 Starting From 5K to...
ย 
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Magarpatta Call Me 7737669865 Budget Friendly No Advance Booking
ย 
DENR EPR Law Compliance Updates April 2024
DENR EPR Law Compliance Updates April 2024DENR EPR Law Compliance Updates April 2024
DENR EPR Law Compliance Updates April 2024
ย 
Proposed Amendments to Chapter 15, Article X: Wetland Conservation Areas
Proposed Amendments to Chapter 15, Article X: Wetland Conservation AreasProposed Amendments to Chapter 15, Article X: Wetland Conservation Areas
Proposed Amendments to Chapter 15, Article X: Wetland Conservation Areas
ย 
Climate Change
Climate ChangeClimate Change
Climate Change
ย 
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Moshi Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Moshi Call Me 7737669865 Budget Friendly No Advance Booking
ย 
RATING SYSTEMS- IGBC, GRIHA, LEED--.pptx
RATING  SYSTEMS- IGBC, GRIHA, LEED--.pptxRATING  SYSTEMS- IGBC, GRIHA, LEED--.pptx
RATING SYSTEMS- IGBC, GRIHA, LEED--.pptx
ย 
Green Marketing
Green MarketingGreen Marketing
Green Marketing
ย 
Deforestation
DeforestationDeforestation
Deforestation
ย 
RA 7942:vThe Philippine Mining Act of 1995
RA 7942:vThe Philippine Mining Act of 1995RA 7942:vThe Philippine Mining Act of 1995
RA 7942:vThe Philippine Mining Act of 1995
ย 
CSR_Module5_Green Earth Initiative, Tree Planting Day
CSR_Module5_Green Earth Initiative, Tree Planting DayCSR_Module5_Green Earth Initiative, Tree Planting Day
CSR_Module5_Green Earth Initiative, Tree Planting Day
ย 
Book Sex Workers Available Pune Call Girls Kondhwa 6297143586 Call Hot India...
Book Sex Workers Available Pune Call Girls Kondhwa  6297143586 Call Hot India...Book Sex Workers Available Pune Call Girls Kondhwa  6297143586 Call Hot India...
Book Sex Workers Available Pune Call Girls Kondhwa 6297143586 Call Hot India...
ย 
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...
Call Girls in Sakinaka Agency, { 9892124323 } Mumbai Vashi Call Girls Serivce...
ย 
Environmental Science - Nuclear Hazards and Us.pptx
Environmental Science - Nuclear Hazards and Us.pptxEnvironmental Science - Nuclear Hazards and Us.pptx
Environmental Science - Nuclear Hazards and Us.pptx
ย 
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Jejuri Call Me 7737669865 Budget Friendly No Advance Booking
ย 
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...
VVIP Pune Call Girls Vishal Nagar WhatSapp Number 8005736733 With Elite Staff...
ย 
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Wagholi ( Pune ) Call ON 8005736733 Starting From 5K to ...
ย 
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...
Call Now โ˜Ž๏ธ๐Ÿ” 9332606886 ๐Ÿ”ย Call Girls โค Service In Muzaffarpur Female Escorts ...
ย 

10_PPT__ML.docx

  • 1. Presentation Outline โ€ข Introduction โ€ข Objectives โ€ข System Architecture / Ideation Map โ€ข Module Implementation โ€ข Results and Discussions โ€ข Conclusion โ€ข References Department of CSE 1
  • 2. Introduction ๏‚ง Machine Learning(ML) is a branch of Artificial Itelligence(AI) & Computer Science which focuses on the use of data and algorithms to imitate the way that humans learn and gradually improving accuracy ๏‚ง During an echocardiogram, a doctor will generate a real-time image of the heart. The test monitors ultrasound, high- frequency sound waves, which are projected through the chest and bounce back to create an image of your heart. ๏‚ง All the patients suffered heart attacks at some point in the past. Some are still alive and some are not. The survival and still-alive variables, when taken together, indicate whether a patient survived for at least one year following the heart attack. ๏‚ง Decision Tree is the most important support tool with a tree- like structure that models probable outcomes, cost of resources, utilities and possible consequences of supervised learning. . Department of CSE 2
  • 3. Introduction ๏‚ง The problem addressed by past researchers was to predict from the other variables whether or not the patient will survive at least one year. The most difficult part of this problem is correctly predicting that the patient will NOT survive. (Part of the difficulty seems to be the size of the data set. โ€ข Problem Statement : The classification goal is to predict still- alive using features from 3 to 10 . Department of CSE 3
  • 4. Objectives โ€ข The objective of the project is to find how to use machine learning techniques for analysis and making predictions using echocardiogram dataset . To find The classification goal is to predict still-alive using features from 3 to 10 . The survival and still-alive variables, when taken together, indicate whether a patient survived for at least one year following the heart attack. โ€ข This report examines how to use machine learning techniques to analyze and predict using echocardiogram dataset. Department of CSE 4
  • 5. Objectives โ€ข This data set is used for exploratory data analysis, data visualization, dealing with missing values and classification modelling techniques โ€ข An ECG should be attached ensuring good tracings to facilitate the acquisition of complete digital loops. Loops should be examined and adjusted accordingly in order to ensure a clear representation of the image acquired. โ€ข Thetest is useful for diagnosing and monitoring heart problems and creating treatment plans. . Department of CSE 5
  • 6. Objectives โ€ข There are three main types of echocardiogram tests that a doctor may perform. โ€ข A transthoracic echocardiogram is noninvasive. It uses a transducer moved across your chest to produce the heart image. โ€ข A transesophageal echo test is performed with a tube transducer in your throat. This helps to view the heart from a different angle. โ€ข Last, a stress echocardiogram occurs during exercise on a treadmill or bike. This is to monitor your heartโ€™s response to physical activity.
  • 8. System Architecture / Ideation Map Import all the necessary Module s(numpy,pandas, matplotlib etc.) Importing the DataSet (CSV) file to Jupiter Notebook Analyse the Result Running the TestCases of the program Performing Exploring Data Analysis(EDA) Using Random Forest Classification Method For prediction Getting Dummies of the Data . Department of CSE 7
  • 9. Project Implementation โ€ข In this project I will demonstrate how to predict still alive from 3 to 10 in Python using the following steps: โ€ข Data preparation โ€ข Exploratory Data Analysis โ€ข Analyzing the correlation to numerical features โ€ข Encode categorical features โ€ข Using Machine Learning Decision Treeโ€™s echocardiogram Algorithm for getting output . Department of CSE 8
  • 10. Project Implementation โ€ข Hardware & Software Requirements: โ€ข 4GB RAM โ€ข Operating System โ€ข Windows 10,Core i5 โ€ข Anaconda Navigator along with Python 3.8 โ€ข Construction of the Project:The dataset used for this project relates to direct marketing campaigns of a Portuguese banking institution. The purpose of building the models is to predict whether the client will subscribe for a term deposit. In order to accomplish the project problem, the proposed solution can be described in followingsteps: . Department of CSE 9
  • 11. Project Implementation โ€ข Step.1: identifying the problem, collecting, describing the dataset โ€ข Step.2: Exploring, data cleaning and preparation โ€ข Step.3: Building the models โ€ข Step.4: Test and evaluate the performance of the models โ€ข Metrics : The evaluation metrics proposed are appropriate given the context of the data, the problem statement, and the intended solution. Theresulting image of an echocardiogram can show a big picture image of heart health, function, and strength. For example, the test can show if the heart is enlarged or has thickened walls. Walls thicker than 1.5cm are considered abnormal. They may indicate high blood pressure and weak or damaged valves. An echocardiogram can also measure if your heart is pumping enough blood through your body.
  • 12. . Department of CSE 10
  • 13. Project Implementation โ€ข survival : The number of months patient survived (has survived, if patient is still alive). Because all the patients had their heart attacks at different times, it is possible that some patients have survived less than one year but they are still alive. Check the second variable to confirm this. Such patients cannot be used for the prediction task mentioned above. โ€ข Still-alive : A binary variable that depicts if the patient is still alive (0: dead at end of survival period, 1: still alive) โ€ข Age-at-heart-attack :Age in years when heart attack occurred โ€ข Pericardial-effusion : Pericardial effusion is fluid around the heart (0: no fluid, 1: fluid) โ€ข Fractional-shortening : A measure of contracility around the heart, lower numbers are increasingly abnormal . Department of CSE 11
  • 14. Project Implementation โ€ข Epss : E-point septal separation, another measure of contractility. Larger numbers are increasingly abnormal โ€ข Lvdd : Left ventricular end-diastolic dimension. This is a measure of the size of the heart at end-diastole. Large hearts tend to be sick hearts โ€ข Wall-motion-score : A measure of how the segments of the left ventricle are moving โ€ข Wall-motion-index : Equals wall-motion-score divided by number of segments seen. Usually 12-13 segments are seen in an echocardiogram. (Preferable to use this variable INSTEAD of the wall motion score) โ€ข Mult : A derivate variable (suggested that it can be ignored) โ€ข Name : The Name of the patient (Has benn replaced with "name" entirely) .. Department of CSE 12
  • 15. Project Implementation โ€ข Group : Group (Has been considered meaningless and suggested to ignore) โ€ข Alive-at-1 : It is a Boolean-valued variable derived from the first two attributes. (0: patient was either dead after 1 year or had been followed for less than 1 year, 1: patient was alive at 1 year) . Department of CSE 13
  • 16. Methodology Data Exploration: โ€ข The data that is used in this project originally comes from the UCI machine learning repository. This intermediate level data โ€ข set has 132 rows and 12 columns. โ€ข The data set provides data that could be used for classifying if patients will survive for at least one year after a heart attack. โ€ข All the patients suffered heart attacks at some point in the past. Some are still alive and some are not. โ€ข The survival and still-alive variables, when taken together, indicate whether a patient survived for at least one year following the heart attack. . Department of CSE 14
  • 17. Methodology โ€ข In this project, we are going to utilize python to develop a predictive machine learning model! begin by loading our data and exploring the columns. โ€ข The problem addressed by past researchers was to predict from the other variables whether or not the patient will survive at least one year. โ€ข The most difficult part of this problem is correctly predicting that the patient will NOT survive. (Part of the difficulty seems to be the size of the data set.) โ€ข This data set is recommended for learning and practicing your skills in exploratory data analysis, data visualization, dealing with missing values and classification modelling techniques. . Department of CSE 15
  • 18. Methodology Data reading and Preparation :- The library can also be used for data-mining and data analysis. The main machine learning functions that the Scikit-learn library can handle are classification, regression, clustering, dimensionality reduction, model selection, and preprocessing. library(ggplot2) library(caret) library(rpart) library(rpart.plot) library(dplyr) library(tidyr) library(DataExplorer) library(FFTrees) data <- read.csv("../input/echocardiogram.csv", sep =',', na.strings = c('NA','?')) str(data) . Department of CSE 16
  • 19. Methodology โ€ข This is the representation of the data sets that can be variables which were using in the echocardiogram โ€ข Here we are using all the 123 variables with the values ## 'data.frame': 133 obs. of 13 variables: ## $ survival : num 11 19 16 57 19 26 13 50 19 25 ... ## $ alive : int 0 0 0 0 1 0 0 0 0 0 ... ## $ age : num 71 72 55 60 57 68 62 60 46 54 ... ## $ pericardialeffusion : int 0 0 0 0 0 0 0 0 0 0 ... ## $ fractionalshortening: num 0.26 0.38 0.26 0.253 0.16 0.26 0.23 0.33 0.34 0.14 ... ## $ epss : num 9 6 4 12.1 22 ... ## $ lvdd : num 4.6 4.1 3.42 4.6 5.75 ... ## $ wallmotion.score : num 14 14 14 16 18 12 22.5 14 16 15.5 ... ## $ wallmotion.index : num 1 1.7 1 1.45 2.25 ... ## $ mult : num 1 0.588 1 0.788 0.571 ... ## $ name : Factor w/ 2 levels "","name": 2 2 2 2 2 2 2 2 2 2 ... ## $ group : Factor w/ 4 levels "","1","2","name": 2 2 2 2 2 2 2 2 2 2 ... ## $ aliveat1 : int 0 0 0 0 0 0 0 0 0 0 ... . Department of CSE 17
  • 20. Results and Discussion The Decision tree we used are calculated using sklearn wrapper so can be trusted for the model performance. Our end goal was to have a Trained model that could beat the untuned benchmark which it did by a very fine margin. So the solution described below is satisfactory to our initial expectations. We generated a final model with above list of tuned parameters. . Department of CSE 18
  • 21. Results and Discussion . Department of CSE 19
  • 22. Conclusion โ€ข Machine learning techniques are widely used data analysis methods in various business and industrial sectors. โ€ข The main reason for that because ML can build predictive models to produce better predictions and achieve the desired level of accuracy, leading to better outcomes. โ€ข Building the models is an easy and straight forward process. โ€ข The main challenges in data analysis are data preparation and cleaning, the selection of appropriate models and attributes used in their implementation. . Department of CSE 20
  • 23. Conclusion โ€ข The aim of the project is how to use machine learning techniques for analysis and making predictions using to determine if the personโ€™s heart is working properly or has significant problems. โ€ข Heart defects are generally diagnosed in infants and children, but can be found in adults too. Echos can even be done during pregnancy to check for fetal issues. โ€ข By checking size, pumping strength, and heart rate, your doctor can find defects. A heart defect can be a hole in a chamber or improper connections between the heart muscles and arteries.
  • 24. . Department of CSE 21
  • 25. References Here we are providing the reference links which we have followed during the project work โ€ข https://www.kaggle.com/loganalive/echocardiogram-dataset- uci/report#missing-value-treatment-- โ€ข https://specialistdirectinc.com/telecardiology-articles/uncategorized/a-guide -to-understanding-echocardiogram-results/ โ€ข https://code.datasciencedojo.com/datasciencedojo/datasets/tree/master/E chocardiogram โ€ข https://erp.bioscientifica.com/view/journals/echo/2/1/G9.xml โ€ข https://www.nature.com/articles/s41746-018-0065-x . Department of CSE 22