0% found this document useful (0 votes)
12 views7 pages

SmartCare: AI Medical Diagnosis System

The document outlines a healthcare web application project that utilizes machine learning and deep learning for accurate disease prediction based on user-input symptoms via text or voice. Key features include personalized notifications, storage of user history, and an interactive frontend that enhances user experience. The project aims to improve upon previous models by incorporating multi-input support, advanced prediction algorithms, and a comprehensive database for analytics and guidance.

Uploaded by

yashiyashwanth2
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
12 views7 pages

SmartCare: AI Medical Diagnosis System

The document outlines a healthcare web application project that utilizes machine learning and deep learning for accurate disease prediction based on user-input symptoms via text or voice. Key features include personalized notifications, storage of user history, and an interactive frontend that enhances user experience. The project aims to improve upon previous models by incorporating multi-input support, advanced prediction algorithms, and a comprehensive database for analytics and guidance.

Uploaded by

yashiyashwanth2
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

TITLE - NAME

Project Overview
Your project is a healthcare web application that allows users to enter symptoms (text or voice), predicts
possible diseases, and provides relevant precautions, medicine suggestions, and reminders. It also stores
user details and history for personalized follow-ups.

Key Goals
 Accurate disease prediction using multiple ML/DL models

 Multi-input support (text & voice) for user-friendly experience

 Interactive and visual output display

 Notification system for reminders (SMS & email)

 Storage of user info & prediction history for analytics and personalized guidance

Comparison with Previous Projects

Feature Previous Projects Your Project

Input Only text-based symptom entry Text + voice input (Web Speech
API)

Prediction Single ML model, basic outputs ML + DL models, NLP similarity,


weighted symptom scoring

Output Text-based, basic disease name Visual outputs with confidence


%, disease description,
precautions, recommended
medicines

User Info Not stored User info, prediction history,


timestamps stored in MongoDB

Notifications None SMS reminders via Twilio +


optional email report
Extra Features Minimal interactivity Multi-language support,
symptom severity weights, charts
for trends, interactive UI
Extra Features Added in Your Project
 Voice input support with real-time conversion

 Multi-ML/DL model predictions for higher accuracy

 Confidence percentages and ranking of possible diseases

 Precautions and medicine suggestions

 Personalized reminders and notifications

 Multi-language support for broader accessibility

 History tracking for repeat users

Datasets Used (Summary of 19 Datasets)


1. CAERS_cleaned.csv – Adverse events in patients from FDA reports (age, gender, symptoms).
2. clean_icd_chapters.csv – ICD chapter codes & descriptions.
3. clean_icd_codes.csv – ICD disease codes with detailed information.
4. clean_icd_groups.csv – ICD groups linking codes to disease categories.
5. Cleaned_Company_Name.csv – Pharmaceutical companies (for drug source reference).
6. Cleaned_Medicine_Description.csv – Medicine names, reasons, descriptions.
7. [Link] – Symptom-disease mapping for ML/DL training.
8. Disease_symptom_and_patient_profile_dataset_cleaned.csv – Disease symptoms with
patient info (age, gender, BP, cholesterol, outcomes).
9. [Link] – Patient demographics and drug prescriptions.
10. drugs_side_effects_cleaned.csv – Drug names, medical conditions, side effects.
11. indian_pharmaceutical_products_final_clean.csv – Drug products and therapeutic classes.
12. [Link] – Symptom, possible diseases, severity,
duration.
13. [Link] – Food & nutritional info (optional for preventive health suggestions).
14. Ratings_cleaned.csv – Drug ratings from reviews.
15. symptom_Description.csv – Disease and symptom description mapping.
16. symptom_precaution.csv – Disease-specific precautions.
17. [Link] – Symptoms and severity weights.
18. [Link] – Symptom-disease labeled data for model testing.
19. [Link] – Symptom-disease labeled data for model training.

Use: These datasets are used for ML/DL model training, disease description,
medicine recommendations, precautions, severity weighting, and confidence scoring.
Full Technical Flow
A) User Input
 Text input: Form or checklist selection

 Voice input: Web Speech API → converts voice → text automatically

 Additional info: Age, gender, optional health parameters

B) Backend (Python – Flask / FastAPI)


1. Receive Input: Symptom text/vector + patient info

2. Preprocess Input:

o Convert symptom names → binary vector (0/1) or weighted values from Symptom-
[Link]

o Handle missing values & normalize patient info

3. Prediction using ML/DL Models:

o ML Models: Random Forest, Decision Tree, XGBoost → quick tabular disease prediction

o DL Models: Keras/TensorFlow Neural Networks → learn complex patterns

o Optional NLP/Cosine Similarity: Match free-text symptoms to disease lists

4. Output:

o Top predicted diseases

o Confidence percentages

o Precautions, medicine suggestions, descriptions

C) Frontend (React)
 Display Predictions Visually:

o Disease name with confidence (%)

o Description & symptoms info (symptom_Description.csv)

o Precautions (symptom_precaution.csv)

o Medicine suggestions (Cleaned_Medicine_Description.csv)

 Interactive Elements:

o Tabs/cards for multiple predictions

o Progress bars/charts for confidence


o Multi-language support for users

D) Database (MongoDB)
 Store:

o User details (ID, name, age, contact)

o Symptoms provided

o Predicted diseases & confidence

o Timestamps & follow-ups

 Benefits:

o Track user history

o Personalized analytics & reminders

o Audit logs for repeated usage

E) Notifications
 SMS: Twilio API → reminder messages (next checkup, precautions)

 Email (optional): Detailed report including disease, confidence, precautions, medicines

F) API Usage
 Backend API: Receives symptom input → processes → returns JSON output

 Frontend API Calls: React fetches prediction results → displays visually

 External APIs:

o Twilio for SMS

o Optional email services for reports

G) Output & User Experience


 Visual output (React UI):

o Predicted diseases ranked by probability

o Confidence bars

o Symptom descriptions

o Precautions & medicine suggestions


 Remainders:

o SMS reminders for consultation

o Optional emails with detailed report

H) Extra Features / User-Friendly Enhancements


 Symptom severity weighting for accurate prediction

 Multi-language voice & text input

 History tracking for repeat users

 Interactive charts for improvement trends

 Medicine & precaution suggestions personalized per patient

In short: Your project combines ML/DL-based disease prediction, voice & text input, interactive
frontend, database storage, and notification system — all together in a Smart Medical Diagnosis System,
adding several features that previous projects didn’t have.

Recommended Project Titles For This Project

🧠 Professional & Research-Style Titles

1. "AI-Powered Smart Medical Diagnosis System using Machine and Deep Learning"

2. "Predictive Disease Diagnosis using Multi-Dataset Learning Models"

3. "Unified AI Model for Multi-Disease Prediction and Health Monitoring"

4. "A Comprehensive Medical Diagnosis System using Hybrid ML and DL Techniques"

5. "Advanced Disease Prediction Framework using Intelligent Data Analytics"

⚙️Tech & Innovation-Focused Titles

1. "Smart HealthCare Assistant: An AI-Based Disease Prediction Platform"

2. "MedMind: Intelligent Disease Detection and Health Tracking System"

3. "DocAI: AI-Driven Medical Prediction and Patient Support System"

4. "HealthLens: A Smart Medical Diagnosis Web Application"


5. "SmartCare: Voice-Enabled Medical Diagnosis using Python and AI"

💬 User-Friendly & Application-Oriented Titles

1. "Smart Medical Diagnosis and Reminder System using AI"

2. "Voice-Interactive Disease Prediction and Health Management System"

3. "AI Health Companion: Predict, Diagnose, and Remind"

4. "Patient-Centered Smart Diagnosis with Data Storage and Remainder Alerts"

5. "ML-Powered Web App for Smart Health Prediction and User Care"

Mava, nēnu suggest chesthuna top 3 catchy titles (reviewers ki instantly impressive ga untayi):

1. 🔹 "SmartCare: Voice-Enabled AI Medical Diagnosis System"

2. 🔹 "MedMind: An Intelligent Disease Prediction and Health Tracking System"

3. 🔹 "AI-Powered Smart Medical Diagnosis using Multi-Disease Datasets"

🏆 Final Shortlisted Unique Titles for Your Project

1. SmartCare: Voice-Enabled AI Medical Diagnosis System


→ Focus: Voice input + AI predictions + reminders + user-friendly web app
2. MedMind: An Intelligent Disease Prediction and Health Tracking System
→ Focus: Smart AI model + history tracking + remainders + multi-disease
3. AI-Powered Smart Medical Diagnosis using Multi-Disease Datasets
→ Focus: Research-style title (perfect for reports and presentations)
4. HealthLens: Deep Learning-Based Disease Prediction and Reminder System
→ Focus: DL + visualization + reminder & analytics-based health view
5. Smart HealthCare Assistant: A Hybrid ML-DL Disease Prediction Platform
→ Focus: Hybrid ML + DL model + integrated dataset + backend intelligence
✅ My Top Recommendation:

➡️"SmartCare: Voice-Enabled AI Medical Diagnosis System"

Why this one fits best:


 Easy to remember and sounds professional
 Clearly shows voice + AI + medical diagnosis
 Highlights your smart features (user-friendly + notifications + ML/DL integration)
 Suitable for report cover page, PPT title slide, and project registration

You might also like