Scribd
Upload
279 views

Medicine Recommendation System

Uploaded by

dasabhoy238
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
279 views

Medicine Recommendation System

Uploaded by

dasabhoy238
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

MEDICINE RECOMMENDATION

SYSTEM
This system leverages machine learning to provide personalized medicine
recommendations for patients. It aims to improve patient care by
suggesting the most appropriate treatments and medications.
INTRODUCTION
The task is about developing a personalized medicine
recommendation system using machine learning. It focuses on
utilizing patient data (e.g., medical history, symptoms) to
recommend optimal medications. The system addresses key
challenges like drug-drug interactions, patient-specific
factors, and treatment personalization. Machine learning
models are trained on healthcare datasets to improve prediction
accuracy and reduce adverse effects. We are using algorithms,
model training, and system implementation for recommending
medicines based on individual health profiles.
RECOMMENDATION SYSTEMS TILL DATE
Medicine recommendation systems that already exist aim to provide personalized treatment
recommendations, help with drug prescriptions, and offer guidance on appropriate medications based
on patient data. Some notable existing systems include:

IBM Watson for Oncology:


Uses AI to analyze patient data, including medical records and genetic information, to suggest
personalized cancer treatment plans. It compares patient information against established treatment
protocols.
Largely focused on oncology and cancer treatments; can be costly for widespread usage.

Doctor AI:
A deep learning model that predicts future medical events, including recommended medications, based
on patient history.
Relies on large volumes of historical data, which might limit its effectiveness for first-time patients or
underrepresented medical conditions.

MediFind:
Recommends doctors and treatments for patients based on their symptoms and conditions. It focuses
on finding the best treatment paths for rare diseases.
Primarily a doctor and treatment finder, rather than providing direct medication recommendations.

Ada Health:
Uses patient symptoms to generate possible diagnoses and recommend treatments or medications.
Focuses more on symptom analysis and diagnostics, rather than on detailed medication
recommendations.
FLOWCHART OF THE MEDICINE RECCOMENDATION SYSTEM
User

Enter
symptoms

Processing is done
by the trained
model

Disease
prediction
Descriptio
n Diet
Precautio Workou
ns t
Medicatio
n
TECH-STACK DISCUSSION:
The project aims to develop a medical recommendation system that predicts diseases based on
patient symptoms and suggests appropriate medications, precautions, workouts, and diet plans.

Dataset:
The dataset contains symptoms and diseases, used for multi-class disease classification. This
includes patient records with symptom descriptions and the corresponding disease diagnoses.
Data Preprocessing:
•Cleaned the dataset, encoded categorical variables, and normalized numerical features.
•Strings (e.g., symptoms) were encoded into integers for model training.
Model Training:
•Various machine learning models were implemented, including Support Vector Classifier (SVC),
Random Forest, and Gradient Boosting.
•Trained and tested multiple models, achieving 100% accuracy with the Support Vector Classifier
due to clean data.
•The trained model was saved for future use.
Disease Prediction:
•Developed a predictive model where patient symptoms are passed as input, and the model predicts
the likely disease.
•The predicted disease is then compared with the actual data for evaluation.
Recommendation System:
•Created a system that provides medication, precautions, workouts, and diet
recommendations based on the predicted disease.
•Developed functions to retrieve and display disease descriptions, medications, and
related information.

Front-End Development:
•Built the web interface using Flask and HTML/CSS (Bootstrap).
•Multiple HTML pages were created for the user interface, including forms where users
can input their symptoms.
•Implemented interactive features like buttons and pop-up windows to enhance user
experience.

Integration:
•Connected the back-end (Flask) with the front-end, passing predicted diseases and
related recommendations.
•Used the Jinja template to display information dynamically on the web pages.

The final system allows users to input symptoms, predicts the disease, and displays the
corresponding recommendations. The interface provides a smooth interaction between
the user input and the disease details.
Thank You

Prepared by:
Sneha Dey
Abhoy Das
Abhishek Saha

You might also like