MAJOR PROJECT PRESENTATION

ON
INTEGRATED DISEASE PREDICTION
PLATFORM

HMR institute of Technology and Management

Under the Supervision of: Submitted by

Mr. Gyanender Kumar Aryan Singh
Department Coordinator Ayush
CSE Department Kabir Mittal
Integrated Disease Prediction Platform

IDP
PLATFORM
 CONTENTS
1. Introduction
2. Scope of work
3. Literature Survey
4. Pitfalls of the work
5. Problem Formulation
6. Objective
7. Tools and Technology
8. Methodology
9. Performance Parameters
10. Result Analysis
11. Comparitive Analysis
12. Conclusion
13. Future Scope
14. Reference
 Model Evaluation:
• Feature Engineering and Selection: • Using appropriate evaluation
Identify and extract relevant metrics (e.g., accuracy, mse ,rse,
features from the integrated data precision, F1-score) to assess the
that are predictive of multiple model's performance.
diseases.

•Design a system to integrate data

• Selecting the most appropriate
from various sources (Electronic
machine learning algorithm for
Health Records, medical images,
Multiple Disease prediction.
genomics data) for a holistic patient
profile.
 3..Beyond Accuracy: Studies go
1.Promising Accuracy: Studies across 2.Algorithm Variety: Different disease
beyond raw accuracy by employing
diabetes, heart disease, and liver disease predictions benefit from diverse machine metrics like precision, recall, and F1
report machine learning models achieving learning algorithms. While ANNs excel in score. This ensures comprehensive
high accuracy in predicting the respective diabetes, SVMs outperform in liver disease, evaluation of model performance,
conditions. Accuracy ranged from 87% for highlighting the need for choosing models considering factors like how effectively
diabetes to 95% for liver disease, showcasing based on specific context and data sets. true positives and negatives are
the potential for early diagnosis and
identified..
intervention
1. Data Availability and Integration: Acquiring and integrating comprehensive
healthcare data from various sources and ensuring its quality and interoperability can be
challenging..
2.Model Generalizability: Developing a model that accurately predicts across diverse
patient populations with varied disease combinations can be difficult.
3.Computational Resources: Implementing and running a complex multi-disease
model might require significant computational resources, posing cost and infrastructure
challenges.
4.Monitoring and Updating: Continuously monitoring model performance and
updating it with new data and evolving disease patterns is crucial to maintain accuracy and
reliability..
1. Current healthcare ML models are siloed: They focus
on single diseases, requiring multiple models and
hindering holistic patient care.
2.Existing models lack accuracy and
comprehensiveness: Suboptimal performance risks
misdiagnosis and may overlook crucial parameters, leading to
false positives/negatives.
 Shift from
single- disease Boost
to multi- prediction
disease accuracy
analysis:

Integrate Streamline
broader workflow and
parameter reduce
sets: complexity
Programming Language
Python Python is a high-level, general-purpose programming language. Its design philosophy
Emphasizes code readability with the use of significant indentation.

Technology
Machine Learning: Machine learning enables a machine to automatically learn from data,
improve
performance from experiences, and predict things without being explicitly programmed.
Libraries and Packages:
Pandas: Python package that offers various data structures and operations for manipulating
numerical
data and time series.
NumPy: NumPy is a general-purpose array processing package that provides tools for handling n-
dimensional arrays.
Seaborn: Seaborn is an amazing visualization library for statistical graphics plotting in Python.
Matplotlib: Matplotlib is a plotting library for the Python programming language and its
numerical
mathematics extension NumPy.
Scikit-learn: It features various classification, regression and clustering algorithms including
support-vector machines, random forests, gradient boosting, k-means.
1. Understanding the problem
Building a Mutiple Disease Prediction App in which paitents can predict multiple diseases
simultaneously with significantly higher accuracy than existing single-disease models, offering earlier
diagnoses and improved patient outcomes
2. Data collection
Collecting the data from relevant resources.
3. Data pre-processing
Removing the unnecessary columns, filling up missing values , changing the data-type of
columns into integer-type, scaling the data by using pandas, numpy, seaborn, scikit-learn kit.
4. Algorithm selection:
Choose appropriate machine learning for individual disease prediction or implement a multi-task learning
approach for simultaneous prediction of multiple diseases
5. Model selection and evaluation
Importing models like linear regression, logistic regression from scikit-learn and
metrics to choose the best machine learning model.
1. Accuracy: It is the ratio of the number of correct predictions to the total number of
predictions made for a dataset.
2. Confusion Matrix: A confusion matrix or error matrix is a table that shows the number of
correct and incorrect predictions made by the model compared with the actual classifications
in the test set or what type of errors are being made.
3. Precision: It is the ratio of True Positives to all the positives predicted by the model.
4. F1-score: It is a single metric that combines both Precision and Recall. The higher the F1
score, the better is the performance of our model. The range for F1-score is [0,1].
 RESULT
ANALYSIS
1. Importing libraries and loading the data from csv file to a
Pandas DataFrame
2. printing the first 5 rows of the dataframe

3.number of rows and columns in the dataframe

4. check relationship between columns

5. Getting more information about the dataset

6. Checking for missing values in each column

7. Getting some statistical measures about the data

8. Distribution of target Variable

1 --> Parkinson's Positive
0 --> Healthy
9. Grouping the data based on the target variable

10. Data Pre-Processing

Splitting the data to training data & Test data

11. Model Training

Support Vector Machine Model

Training the SVM model with training data

Naive bayes Model

Training the naive bayes model with training data

Logistic Regression
K-Nearest Neighbour
Random Forest
1. User Interface:

User interface
2. Diabetes Disease:

Diabetes Disease Input Data

3. Heart Disease Prediction

Heart Disease Prediction

4. Parkinson’s Prediction

Parkinson’s Prediction
5. Liver Disease Prediction

Liver Disease Prediction

6. Jaundice Prediction

Jaundice Prediction
7. Hepatitis Prediction

Hepatitis Prediction
8. Lung Cancer Prediction

Lung Cancer Prediction

In conclusion, Mutiple disease prediction Web App using machine learning
and Streamlit, offering a user-friendly interface and overcoming limitations of
traditional models. The system's unified platform predicts multiple diseases
simultaneously, addressing accuracy challenges in single-disease models for
improved early detection and patient outcomes

1. Breaking Barriers: Our system surpasses single-disease models by

simultaneously analyzing multiple health conditions, offering a comprehensive
view of patient health and boosting efficiency.

2. AI-Powered Accuracy: Employing robust machine learning algorithms

and convolutional neural networks, we achieve superior disease detection
accuracy, leading to earlier diagnoses and better patient outcomes.

3. Streamlined Workflow: Built using user-friendly Streamlit, our

system provides a seamless interface for healthcare professionals,
streamlining the prediction process and enhancing ease of use.
 FUTURE SCOPE

In the coming years, the future scope of disease prediction using the "Integrated
Disease Prediction Platform" involves a concerted effort towards refining algorithmic
methodologies, enhancing data integration techniques, and advancing model
interpretability. Integration of multi-modal data sources, including genomic, proteomic,
and clinical data, will be pivotal for building comprehensive predictive models.
Additionally, the development of novel feature selection algorithms and interpretable
AI techniques will contribute to the creation of more transparent and clinically relevant
predictive models. Real-time disease monitoring, cross-disease prediction capabilities,
and the translation of research findings into clinical practice will further solidify the
platform's impact on personalized healthcare delivery. Moreover, addressing ethical
and regulatory considerations and fostering interdisciplinary collaborations will be
essential for ensuring the responsible and equitable deployment of the Integrated
Disease Prediction Platform in diverse healthcare settings, ultimately leading to
improved patient outcomes and population health management.
Base paper link:

1. [1] S. Aruna; G. Manjula Devi; M. Priyanka "Parkinson’s Disease Prediction using

Machine Learning Techniques", 2021 International Conference on Innovative
Computing and Communications (ICICC)

2. [2] M. S. R. Karthick; S. Rajendran "Jaundice Prediction Using Machine Learning

Techniques", 2021 International Conference on Computational Intelligence and
Sustainable Technologies (ICCIST)

3. [3] F. Khan; S. K. Pal "Hepatitis Disease Prediction Using Machine Learning

Algorithms", 2021 International Conference on Artificial Intelligence and Data
Engineering (AIDE)
[4] P. S. Goyal; S. Gupta; A. Mittal "Lung Cancer Prediction Using Machine Learning
Techniques", 2021 International Conference on Computer Communication and Informatics
(ICCCI)

[5] A. K. Sharma; N. K. Gupta "Chronic Kidney Disease Prediction Using Machine Learning
Algorithms", 2022 IEEE International Conference on Electrical, Computer and
Communication Technologies (ICECCT)

[6] S. K. Mishra; R. K. Gupta; M. Pandey "Breast Cancer Prediction Using Machine

Learning Techniques", 2021 International Conference on Advanced Computing and
Intelligent Engineering (ICACIE)

[7] A. R. Sarker; M. H. Khan "Alzheimer's Disease Prediction Using Machine Learning

Algorithms", 2022 International Conference on Advanced Computing Technologies and
Applications (ICACTA)

[8] R. S. Patel; K. P. Singh; S. K. Mishra "Thyroid Disease Prediction Using Machine

Learning Techniques", 2021 International Conference on Computing, Communication and
Signal Processing (ICCCSP)