THYROID DISEASE CLASSIFICATION

USING MACHINE LEARNING

1.Introduction:

 The Thyroid gland is a vascular gland and one of the most important organs
of the human body.
 The two types of Thyroid disorders are Hyperthyroidism and
Hypothyroidism.
 A thyroid-related Blood test is used to detect this disease but it is often
blurred and noise will be present.
 Data cleansing methods were used to make the data primitive enough for the
analytics to show the risk of patients getting this disease.
 Machine Learning plays a very deciding role in disease prediction.
 Machine Learning algorithms, SVM - support vector machine, Random
Forest Classifier, XGB Classifier and ANN - Artificial Neural Networks are
used to predict the patient’s risk of getting thyroid disease.
 Technical Architecture:
1.1 Overview:
 Define Problem / Problem Understanding
o Specify the business problem
o Business requirements
o Literature Survey
o Social or Business Impact.
 Data Collection & Preparation
o Collect the dataset
o Data Preparation
 Exploratory Data Analysis
o Descriptive statistical
o Visual Analysis
 Model Building
o Training the model in multiple algorithms
o Testing the model
 Performance Testing & Hyperparameter Tuning
o Testing model with multiple evaluation metrics
o Comparing model accuracy before & after applying hyperparameter
tuning
 Model Deployment
o Save the best model
o Integrate with Web Framework
 Project Demonstration & Documentation
o Record explanation Video for project end to end solution
o Project Documentation-Step by step project development procedure
Milestone 1: Define Problem / Problem Understanding

Activity 1: Specify the business problem

 Refer to Project Description
Activity 2: Business requirements
 The business requirements for a machine learning model to predict
thyroid disease include the ability to accurately predict thyroid disease
based on the scan results, Minimize the number of false positives (wrong
thyroid disease confirmations) and false negatives (thyroid is there but
got as not thyroid disease).
 Provide an explanation for the model's decision, to comply with
regulations and improve transparency.
Activity 3: Literature Survey (Student Will Write)
 The thyroid gland is one of the body’s most visible endocrine glands. Its
size is determined by the individual’s age, gender, and physiological
states, such as pregnancy or lactation.
 It is divided into two lobes (right and left) by an isthmus a band of
tissue). It is imperceptible in everyday life yet can be detected when
swallowing.
 The thyroid hormones T4 and T3 are needed for normal thyroid function.
These hormones have a direct effect on the body’s metabolic rate.
 It contributes to the stimulation of glucose, fatty acid, and other molecule
consumption.
  Additionally, it enhances oxygen consumption in the majority of the
body’s cells by assisting in the processing of uncoupling proteins, which
contributes to an improvement in the rate of cellular respiration.
 Thyroid conditions are difficult to detect in test results, and only trained
professionals can do so. However, reading such extensive reports and
predicting future results is difficult.
 Assume a machine learning model can detect the thyroid disease in a
patient. The thyroid disease can then be easily identified based on the
symptoms in the patient’s history.
 Currently, models are evaluated using accuracy metrics on a validation
dataset that is accessible.
Activity 4: Social or Business Impact.
Social Impact:-
 Untreated/undetected thyroid disease is more dangerous at times it
can lead to fatal of the person. So, we can detect it at the earliest then
people can get treatment and get cured.
Business Model/Impact:-
 We can make this application public, offer services as a subscription
based or can collaborate with healthcare centers or specialists.
Milestone 2: Data Collection & Preparation

 ML depends heavily on data. It is the most crucial aspect that makes

algorithm training possible.
 So this section allows you to download the required dataset.
Activity 1: Download the dataset

 There are many popular open sources for collecting the data.
 Eg: kaggle.com, UCI repository, etc.
 In this project, we have used drug200.csv data. This data is
downloaded from kaggle.com.
 Please refer to the link given below to download the dataset.
Link: https://www.kaggle.com/prathamtripathi/drug-classification

Activity 1.1: Importing the libraries

Import the necessary libraries as shown in the image.

Activity 1.2: Read the Dataset

 Our dataset format might be in .csv, excel files, .txt, .json, etc. We
can read the dataset with the help of pandas.
 In pandas, we have a function called read_csv() to read the dataset.
As a parameter, we have to give the directory of the csv file.
Activity 2: Data Pre-processing
As we have understood how the data is, let's pre-process the collected
data.
The download data set is not suitable for training the machine learning
model as it might have so much randomness so we need to clean the dataset
properly in order to fetch good results. This activity includes the following steps.
  Handling missing values
 Descriptive analysis
 Splitting the dataset as x and y
 Handling Categorical Values
 Checking Correlation
 Converting Data Type
 Splitting dataset into training and test set
 Handled Imbalanced Data
 Applying StandardScaler
Activity 2.1: Checking for null values
 For checking the null values, data.isnull() function is used. To sum
those null values we use the .sum() function to it. From the below
image we found that there are no null values present in our dataset.
So we can skip handling the missing value step.
 Removing the Redundant attributes from the dataset.

 Re-mapping the 'target' values to the diagnostic Group

  Re-mapping the 'target' values to the diagnostic Group

 Dropping Null Values

Checking the 'age' is there any above 100 and we drop the age>100.

Activity 2.2: Splitting the data x and y

Splitting the data x and y
 Making 'F' on wherever we have the 'nan' values on data.

Activity 2.3: Converting the Data Type

 Converting the data type from object to float. So that we will get
output properly and Checking info about the data.
 Here, we have the object values are 'TSH', 'T3', 'TT4', 'T4U', 'FTI', 'TBG' and
convert them to float values.

 Then we can check the datatype information about the dataset by code of
x.info()
Activity 2.4: Handling Categorical Values
 As we can see our dataset has categorical data we must convert the categorical
data to integer encoding or binary encoding.
 To convert the categorical features into numerical features we use encoding
techniques. There are several techniques but in our project we are using Ordinal Encoding and
Label Encoding.

• In our project, categorical features are x and y values.

• Here, applying Ordinal Encoding on x values.
• Replacing the nan values with zero (0) values.

• Now, applying Label Encoding on y(Independent variable) value.

Activity 2.5: Splitting data into train and test
Now let’s split the Dataset into train and test sets
Changes: first split the dataset into x and y and then split the data set
Here x and y variables are created. On x variable, data is passed with dropping the
target variable. And my target variable is passed. For splitting training and testing
data we are using the train_test_split() function from sklearn. As parameters, we
are passing x, y, test_size, random_state.
Activity 2.6: Handling Imbalanced Data

Activity 2.7: Applying StandardScaler

• Scaling the features makes the flow of gradient descent smooth and helps
algorithms quickly reach the minima of the cost function.
• Without scaling features, the algorithm may be biased toward the feature
which has values higher in magnitude. it brings every feature in the same
range and the model uses every feature wisely.

• Here, we have the data in array format and we are making it

dataframe(table format).

Activity 2.8: Performing Feature Importance

• The idea behind permutation feature importance is simple. The feature
importance is calculated by noticing the increase or decrease in error when we
permute the values of a feature.
• If permuting the values causes a huge change in the error, it means the feature is
important for our model.
Activity 2.9: Selecting Output Columns
• Before we have this many columns

• After Performing Feature Importance by using 'Permutation Importance' we are

dropping some columns which are not important for 'target'.
Milestone 3: Exploratory Data Analysis
Activity 1: Descriptive analysis
Descriptive analysis is to study the basic features of data with the statistical process. Here
pandas have a worthy function called describe. With this described function we can find
mean, std, min, max and percentile values of continuous features.
Checking info about data by using data_info()

Activity 2: Visual analysis

Visual analysis is the process of using visual representations, such as charts, plots,
and graphs, to explore and understand data. It is a way to quickly identify
patterns, trends, and outliers in the data, which can help to gain insights and make
informed decisions.
Activity 2.1: Checking Correlation.
Here, I'm finding the correlation using HeatMap. It visualizes the data in 2-D
coloured maps making use of colour variations. It describes the related variables
in the form of colours instead of numbers; it will be plotted on both axes.
Here, there is no correlation between columns.

Milestone 4: Model Building

Activity 1: Training the model in multiple algorithms
Now our data is cleaned and it’s time to build the model. We can train our data on
different algorithms. For this project we are applying four classification
algorithms. The best model is saved based on its performance.
Activity 1.1: Random Forest Classifier Model
A function named Random Forest Classifier Model is created and train and test
data are passed as the parameters. Inside the function, the Random Forest
Classifier algorithm is initialized and training data is passed to the model with the
.fit() function. Test data is predicted with the .predict() function and saved in a
new variable. For evaluating the model, accuracy_score and classification report
is done.
Activity 1.2: XGBClassifier model
A function named XGBClassifier model is created and train and test data are
passed as the parameters. Inside the function, the XGBClassifier algorithm is
initialized and training data is passed to the model with the .fit() function. Test
data is predicted with the .predict() function and saved in a new variable. For
evaluating the model, the accuracy score and classification report is done.
Activity 1.3: SVC model
A function named SVC model is created and train and test data are passed as the
parameters. Inside the function, the SVC algorithm is initialized and training data
is passed to the model with .fit() function. Test data is predicted with the .predict()
function and saved in a new variable. For evaluating the model, the accuracy
score and classification report is done.
Activity 1.4 ANN Model
Artificial Neural Networks (ANN) are multi-layer fully-connected neural nets.
They consist of an input layer, multiple hidden layers, and an output layer. Every
node in one layer is connected to every other node in the next layer. We make the
network deeper by increasing the number of hidden layers.
Activity 2: Testing the model
Milestone 5: Performance Testing & Hyperparameter Tuning
Activity 1: Testing model with multiple evaluation metrics
Multiple evaluation metrics means evaluating the model's performance on a test
set using different performance measures. This can provide a more comprehensive
understanding of the model's strengths and weaknesses. We are using evaluation
metrics for classification tasks including accuracy, precision, recall, support and
F1-score.
Activity 1.1: Compare the model
For comparing the above four models, the compareModel function is defined.
Activity 2:Comparing model accuracy before & after applying
hyperparameter tuning
From sklearn, accuracy is used to evaluate the score of the model. On the parameters, we
have given xgb1 (model name), x, y, cv (as 3 folds). Our model is performing well. So,
we are saving the model by pickle.dump().
Note: To understand cross validation, refer to this link.
https://towardsdatascience.com/cross-validation-explained-evaluating-estimator-performa
nce-e51e5430ff85.
Milestone 6: Model Deployment
Activity 1:Save the best model
Saving the best model after comparing its performance using different evaluation
metrics means selecting the model with the highest performance and saving its
weights and configuration. This can be useful in avoiding the need to retrain the
model every time it is needed and also to be able to use it in the future.

Activity 2: Integrate with Web Framework

In this section, we will be building a web application that is integrated to the
model we built. A UI is provided for the uses where he has to enter the values for
predictions. The enter values are given to the saved model and prediction is
showcased on the UI.
This section has the following tasks
• Building HTML Pages
• Building server side script
Activity 2.1: Building Html Pages:
For this project create three HTML files namely
• home.html
• predict.html
• submit.html
and save them in the templates folder.
Let’s see how our home.html page looks like:
Now when you click on predict button from top right corner you will get
redirected to predict.html
Let's look how our predict.html file looks like:
Now when you click on submit button from left bottom corner you will get
redirected to submit.html
Let's look how our submit.html file looks like: it is ['miscellaneous'].
Activity 2.2: Build Python code:
Import the libraries

Load the saved model. Importing the flask module in the project is mandatory. An
object of Flask class is our WSGI application. Flask constructor takes the name of
the current module (__name__) as argument.

Render HTML page:

Here we will be using a declared constructor to route to the HTML page which
we have created earlier.
In the above example, ‘/’ URL is bound with the home.html function. Hence,
when the home page of the web server is opened in the browser, the html page
will be rendered. Whenever you enter the values from the html page the values
can be retrieved using POST Method.
Retrieves the value from UI:

Here we are routing our app to predict() function. This function retrieves all the
values from the HTML page using Post request. That is stored in an array. This
array is passed to the model.predict() function. This function returns the
prediction. And this prediction value will be rendered to the text that we have
mentioned in the submit.html page earlier.
Main Function:

Activity 2.3: Run the application

• Open anaconda prompt from the start menu
• Navigate to the folder where your python script is.
• Now type “python app.py” command
• Navigate to the localhost where you can view your web page.
• Click on the predict button from the top right corner, enter the inputs,
click on the submit button, and see the result/prediction on the web.
Milestone 7: Project Demonstration & Documentation
Below mentioned deliverables to be submitted along with other deliverables
Activity 1:- Record explanation Video for the project end to end solution
Activity 2:- Project Documentation-Step by step project development
procedure
Create document as per the template provided