SANATANA VAIDY :

(Balancing of mental and physical health)

A Main project thesis submitted in partial fulfilment of requirements for the award of degree for
VIII semester

BACHELOR OF
TECHNOLOGY IN
COMPUTER SCIENCE & ENGINEERING
BY

S. Parimala Sri (21131A05N7)

T. Devika (21131A05P6)
V. Sri Krishna Vamsi (21131A05R2)
T. Sri Sai Chaitanya (21131A05Q4)

Under the esteemed guidance of

Name of the Project Mentor
Dr.CH SITA KUMARI
(Associate Professor)
Department of CSE
GVPCE(A)

Department of Computer Science & Engineering

Gayatri Vidya Parishad College of Engineering –
(A) (Affiliated to JNTU – K, Kakinada)
Visakhapatnam
Academic Year: 2024 – 2025
 CERTIFICATE

This is to certify that the project entitled “Santana Vaidy (balancing of mental and physical health)”
provides personalized health recommendations and natural remedies based on Ayurvedic principles in users'
local languages submitted by

S. Parimala Sri (21131A05N7)

T. Devika (21131A05P6)

V. Sri Krishna Vamsi (21131A05R2)

T. Sri Sai Chaitanya (21131A05Q4)

In their VIII semester in partial fulfilment of the

requirements for the Award of Degree of
Bachelor of Technology
IN
COMPUTER SCIENCE & ENGINEERING During the academic year 2024-2025

Guide
Dr. CH. SITA KUMARI Head of the Department
(Associate Professor) Dr. C. V. Guru Rao
Department of CSE (Professor & HOD)
GVPCE(A) Department of CSE
GVPCE(A)

------------------------

Signature of External
 DECLARATION

We hereby declare that this project entitled “SANATANA VAIDY (balancing of mental and
physical health)” is a bonafide work done by us and submitted to “Department of Computer Science and
Engineering, G. V. P College of Engineering (Autonomous) Visakhapatnam, in partial fulfilment for the
award of the degree of B. Tech is of our own and it is not submitted to any other university or has been
published any time before.

PLACE: VISAKHAPATNAM S. Parimala Sri (21131A05N7)

DATE: T. Devika (21131A05P6)

V. Sri Krishna Vamsi (21131A05R2)

T. Sri Sai Chaitanya (21131A05Q4)

 ACKNOWLEDGEMENT

We would like to express our deep sense of gratitude to our esteemed institute Gayatri
Vidya Parishad College of Engineering - (A), which has provided us with an opportunity to
fulfill our cherished desire.

We express our profound gratitude and our deep indebtedness to our guide and coordinator
Dr. CH. SITA KUMARI Associate Professor, Department of Computer Science and Engineering,
whose valuable suggestions, guidance and comprehensive assessments helped us a lot in realizing
our project successful completion.

We are very thankful to Dr. C. V. Guru Rao, Head of the Department of Computer
Science and Engineering, Gayatri Vidya Parishad College of Engineering (Autonomous), for
giving us an opportunity to do the project in college.

We express our sincere thanks to our Principal Dr. A.B. Koteswara Rao, Gayatri Vidya
Parishad College of Engineering (Autonomous) for his encouragement to us during this project,
giving us a chance to explore and learn new technologies in the form of internship.

Finally, I am indebted to the teaching and non-teaching staff of the Computer Science and
Engineering Department for all their support in completion of the project.

Project Members
S. Parimala Sri (21131A05N7)

T. Devika (21131A05P6)

T. Sri Sai Chaitanya (21131A05Q4)

V. Sri Krishna Vamsi (21131A05R2)

 ABSTRACT

This project presents an AI-driven chatbot that engages users in their native language, providing
health-related guidance rooted in Ayurvedic principles. By leveraging advanced natural language
processing and localized AI technologies, the system ensures culturally relevant and easily accessible
healthcare information. Focused on holistic well-being and traditional wisdom, this solution empowers
users with practical, natural options, reaffirming the importance of Ayurveda in modern healthcare.

AI-powered chatbots equipped with Natural Language Processing (NLP) can bridge the gap
between ancient Ayurvedic knowledge and modern accessibility challenges. These systems interact with
users in their preferred languages, analyse user queries, interpret symptoms, and provide personalized
advice rooted in Ayurvedic doctrines. By tailoring guidance to individual needs and using localized AI,
this technology ensures inclusivity and relevance for diverse populations.

A unique feature of this platform is the integration of Bhagavad Gita slokas, which add a spiritual
and philosophical dimension to wellness. The teachings emphasize mental health, stress management, and
balance—values closely aligned with Ayurveda. By offering slokas with practical interpretations in
multiple languages, the platform enhances accessibility and promotes holistic healing that combines
physical, mental, and spiritual well-being.

Additionally, the platform includes a section addressing myths and facts about Ayurveda,
debunking misconceptions with scientifically validated insights. This feature bridges the gap between
ancient wisdom and contemporary science, helping younger generations appreciate the rationale behind
traditional practices. By combining AI, Ayurveda, and philosophical teachings, this platform
democratizes Ayurvedic knowledge and empowers users with accurate, natural, and practical approaches
to well-being.

KEYWORDS: LLMs, AI, ML, Rasa, Healthcare, APIs, NLP, Ayurvedic Medicine, Conversational AI,
Homeopathy, Natural Remedies, Symptom Analysis, Rural Healthcare, Database Management,
Traditional Medicine, Multilingual Support.
 INDEX

Chapter No Title Page No

1 Introduction…………………..…….… 7
1.1Objective
1.2Purpose
1.3Scope
2 Literature Survey…………….……..… 9
2.1 Previous papers
2.2 Objectives
3 Existing Systems …………….………. 11
3.1 Methodology used
3.2 Drawbacks
4 Proposed System …………….………. 12
4.1 Flow Diagram
4.2 Modules
4.3 Block Diagram
4.4 Methodology
4.5 Performance Measure
4.6 Advantage
5 Dataset and Algorithm…………….….. 20
6 Sample Coding………………….…….. 22
7 Outputs…………….………………….. 27
8 Testing…………….……………..……. 33
8.1 Introduction
8.2 Unit Testing
9 Conclusion…………….…………….... 35
10 Future Scope…………….………….… 36
11 User Manual…………….………….…. 37
12 References…………….…………….… 38

6
 1. INTRODUCTION

Ensuring access to healthcare remains a significant challenge, particularly for individuals in rural
areas who often face limitations in both awareness and financial means. Many low-income families
struggle to obtain timely medical care due to the high cost of treatments and the scarcity of healthcare
facilities. Even for minor illnesses, visiting a hospital may be both inconvenient and unaffordable.
Consequently, such issues are often neglected, potentially leading to more severe complications.
Additionally, synthetic medications, while effective, may cause side effects, further deterring individuals
from seeking medical attention. To address these barriers, we propose an AI-powered chatbot designed to
assist users with common health concerns and improve healthcare accessibility. The chatbot
communicates in the user's local language, eliminating language barriers and making healthcare advice
more inclusive. Incorporating the ancient knowledge of Ayurveda—a trusted system of natural medicine
—the chatbot offers safe and effective remedies. Ayurveda promotes holistic well-being, offering simple
solutions for minor ailments through home remedies, diet, and lifestyle adjustments.

1.1 Objective

The primary objective of this project is to develop an AI-powered chatbot that enhances healthcare
accessibility by providing Ayurvedic-based guidance. The chatbot aims to:

 Offer personalized health recommendations based on user queries and health history.
 Promote Ayurveda as a cost-effective and natural alternative for managing minor ailments.
 Minimize hospital visits by providing preliminary healthcare advice.
 Educate users on preventive healthcare practices to improve overall well-being.

1.2 Purpose

The purpose of this project is to create an inclusive healthcare solution that integrates AI and Ayurveda to
offer accessible, effective, and personalized health guidance. By providing users with AI-driven insights
into natural remedies, lifestyle modifications, and holistic wellness, the chatbot aims to:

 Bridge the gap between traditional Ayurvedic knowledge and modern healthcare needs.
 Ensure culturally relevant healthcare support through NLP and localized AI models.
 Empower individuals, especially in underserved communities, with reliable health information.

7
1.3 Scope

This project focuses on:

 Developing an AI chatbot that understands and processes natural language queries related to
health concerns.
 Implementing Ayurvedic principles for personalized health recommendations.
 Ensuring accessibility through multilingual support and user-friendly design.
 Providing a knowledge base that includes Ayurvedic remedies, diet suggestions, and preventive
care practices.
 Integrating Bhagavad Gita teachings to support mental and emotional well-being.

By combining artificial intelligence with Ayurvedic expertise, this project seeks to create an inclusive and
sustainable healthcare platform. The chatbot not only addresses immediate health issues but also educates
users on maintaining long-term wellness. This integration of traditional medicine with modern AI
technology provides a practical and affordable alternative for addressing healthcare needs, especially in
underserved communities.

8
 2. Literature Survey

Healthcare chatbots have gained significant attention in recent years, integrating AI, NLP, and
machine learning to enhance healthcare accessibility. Various studies highlight the effectiveness of
chatbots in diagnosing diseases, providing preliminary healthcare advice, and improving patient
engagement.

2.1 Previous Papers

R Jegadeesan, Dava Srinivas, N Umapathi, G Karthick, and N Venkateswaran authored the paper
"Personal Healthcare Chatbot for Medical Suggestions Using Artificial Intelligence and Machine
Learning." This chatbot relies on Natural Language Processing (NLP) to interpret user queries, which
can sometimes lead to misinterpretations, especially for complex medical conditions. It cannot replace
professional medical consultation, and its effectiveness is limited by the quality and completeness of the
training dataset.

Sagar Badlani, Tanvi Aditya, Meet Dave, and Sheetal Chaudhari wrote the paper "Multilingual
Healthcare Chatbot Using Machine Learning." While the chatbot's multilingual capability is
beneficial, it may introduce semantic inaccuracies in medical translations. The system relies on a
predefined dataset, making it difficult to address unique or rare medical queries accurately. Furthermore,
it cannot replace professional medical advice, as overlapping symptoms may lead to incorrect predictions,
and the dependence on the Random Forest classifier could introduce biases.

Lekha Athota, Vinod Kumar Shukla, Nitin Pandey, and Ajay Rana authored the paper "Chatbot for
Healthcare System Using Artificial Intelligence." This chatbot primarily functions as a question-
answer system and lacks complex diagnostic reasoning. It relies on N-gram, TF-IDF, and cosine
similarity for responses, which may not fully capture contextual nuances in medical conversations. The
system is limited to its predefined knowledge base and does not dynamically learn from new medical
advancements. Additionally, it lacks real-time consultation capabilities and emergency handling, reducing
its effectiveness for critical medical situations.

Shraddha A. S, Shreepada Bhat, Shubhashri V. K, Sinchana Karnik, and Narender M authored the
paper "Disease Prediction Chatbot." This chatbot leverages Natural Language Processing (NLP) and
machine learning to predict diseases based on user symptoms, but its accuracy is highly dependent on the
quality of the dataset it is trained on. It struggles with diagnosing rare or complex diseases that require
expert medical evaluation and lacks real-time interaction with healthcare professionals, limiting its
9
reliability for critical cases. Additionally, NLP-based systems may misinterpret user input, leading to
incorrect predictions that could cause unnecessary concern or delays in seeking professional medical
care.

Aishwarya Gupta, Divya Hathwar, and Anupama Vijayakumar wrote the paper "Introduction to AI
Chatbots." This paper provides a broad theoretical overview of AI chatbots, their types, and applications
but lacks detailed real-world implementation examples or case studies. While it categorizes chatbots
based on market trends and usability, it does not thoroughly assess their effectiveness or provide
performance metrics. Furthermore, it does not extensively discuss ethical and privacy concerns,
particularly in sensitive applications such as healthcare and customer service. The reliance on predefined
responses in certain chatbot types also limits their ability to handle complex or dynamic conversations
effectively.

2.2 Objectives

This project aims to bridge the gap between traditional Ayurvedic wisdom and modern AI-driven
healthcare. The key objectives include:

 Developing an AI-powered chatbot that provides Ayurvedic health guidance.

 Integrating Bhagavad Gita teachings for mental well-being and stress management.
 Enhancing accessibility through multilingual NLP support.
 Educating users on Ayurveda-based preventive healthcare practices.
 Promoting cost-effective and natural healthcare solutions for minor ailments.

By leveraging AI, NLP, and Ayurvedic principles, this chatbot ensures accessible and holistic healthcare
guidance, particularly benefiting underserved communities. The fusion of modern technology with
ancient wisdom enhances the effectiveness and reach of Ayurveda in contemporary healthcare settings.

10
 3. Existing System

3.1 Methodology

Personalized Ayurveda Analysis:

 AI-driven tools analyse symptoms and user Prakriti (body constitution) to provide
customized Ayurvedic remedies.
 Uses NLP and machine learning models to interpret health-related queries.

AI-Powered Ayurvedic Chatbot:

 Chatbot interacts in multiple regional languages, ensuring accessibility.

 Provides dosha analysis and suggests Ayurvedic treatments.

Spiritual and Mental Well-being Integration:

 Incorporates Bhagavad Gita teachings to promote mental health and ethical well-being.
 Guides users on lifestyle adjustments based on Ayurvedic principles.

Misinformation Control and Knowledge Enhancement:

 AI-driven fact-checking mechanisms ensure accurate Ayurvedic information.

 Digitized texts like Charaka Samhita provide AI-driven healthcare solutions.

3.2 Drawbacks

 Limited Personalization: Generalized recommendations may not fully account for mental state,
seasonal changes, and lifestyle variations.
 Accuracy Concerns: Symptom checkers may not capture the full complexity of Ayurvedic
diagnostics.
 Cultural Relevance Challenges: Insufficient support for regional languages and cultural nuances
may limit accessibility.
 Potential Misinformation: If not properly curated, AI-driven suggestions could perpetuate myths
or misinterpret Ayurvedic concepts.

11
 4.Proposed System

Our project integrates traditional Ayurvedic medicine with modern technology to provide
individuals with easy access to Ayurvedic treatments while promoting mental well-being through
historical texts such as the Bhagavad Gita. The core component of this system is an AI-powered
Santana chatbot, developed using machine learning (ML), natural language processing (NLP), and
the Rasa framework. This chatbot acts as a virtual health assistant, enabling seamless communication in
the user’s local language and delivering personalized Ayurvedic recommendations based on
symptoms and preferences.

The chatbot is designed to analyse user inputs and offer dietary advice, herbal remedies, and
lifestyle modifications aligned with Ayurvedic principles. By leveraging NLP techniques and ML
algorithms, it ensures accurate responses, making holistic healthcare accessible to a wider audience. The
chatbot continuously learns from interactions, refining its ability to understand complex queries and
provide contextually relevant suggestions.

Beyond physical health, the platform emphasizes mental well-being by integrating teachings
from the Bhagavad Gita. These insights offer stress management techniques, mindfulness practices,
and ethical guidance, helping users achieve emotional stability and holistic wellness. By merging
spiritual wisdom with Ayurvedic medicine, the system fosters a balanced approach to health that
considers both the mind and body.

Additionally, the platform features a Myths Page, which educates users about common
misconceptions surrounding Ayurveda and provides scientific explanations to debunk false beliefs. This
ensures that users gain a clear and evidence-based understanding of Ayurvedic practices, helping to
combat misinformation. By integrating AI with traditional health knowledge, our system aims to
provide comprehensive, accessible, and reliable healthcare guidance, empowering users to make well-
informed health decisions.

12
 4.1FlowDiagram:

Fig: 4.1 Flow Diagram

4.2 Modules

Generalized Ayurvedic Chatbot

This chatbot provides basic Ayurvedic recommendations based on common symptoms and
general health queries. It is trained using the Rasa framework, which enables natural language
processing for intent recognition and response generation. The chatbot does not store user data and serves
as a quick-access tool for users seeking immediate Ayurvedic guidance without requiring login.

Personalized Ayurvedic Chatbot

This advanced chatbot offers customized health recommendations based on individual health
history and user preferences. It uses Neural Language Models (NLMs) to process user inputs and refine
its responses. Users must log in using their Aadhar number, allowing the system to store and retrieve
their health data for personalized treatment plans, lifestyle recommendations, and follow-up
guidance.

Slokas Page

This module provides users with Ayurvedic wisdom and mental well-being insights through
Bhagavad Gita slokas. It integrates Google Translator and APIs to generate slokas dynamically and
13
provide translations in the user’s preferred language. This feature enhances accessibility, ensuring users
from different linguistic backgrounds can benefit from spiritual and ethical guidance alongside
Ayurvedic health practices.

4.3BlockDiagram:

Fig: 4.2 Block Diagram

4.4 Methodology:

Our website is designed to provide mental and physical health improvement using traditional methods
combined with modern AI-based technology. The methodology follows a systematic approach that
ensures user-friendly interactions, accurate responses, and holistic well-being recommendations. The
process is divided into key phases:
14
1. User Authentication and Personalization
 Users are required to enter their Aadhaar number to verify their identity.
 The system connects to the Aadhaar API, which checks whether the user exists in the UIDAI
database.
 If the user is found, their medical history is retrieved from a secure patient database, ensuring
personalized health insights.
 A personalized chatbot is initialized based on the retrieved data, allowing for tailored health
advice.
Why this step?
Ensures user identity verification for secure access.
Allows personalized health recommendations based on past medical history.

2. AI-Powered Personalized Health Chatbot

 The chatbot is built using Rasa, an advanced AI-powered conversational platform.
 Users can interact with the chatbot via speech-to-text conversion, where voice input is converted
into text using Google Speech API.
 The chatbot provides personalized recommendations based on user inputs such as:
 Dosha imbalances (as per Ayurveda).
 Nutritional deficiencies (Vitamin D, Iron, etc.).
 Lifestyle habits (sleep, exercise, stress levels).
 Traditional healing techniques (Ayurvedic herbs, home remedies).
 Responses are dynamically generated based on real-time user queries, ensuring an engaging and
interactive conversation.
Why this step?
Offers customized solutions rather than generic health advice.
Uses natural language processing (NLP) for smooth interactions.
Supports multi-language translation, ensuring accessibility for diverse users.

3. Integration of Bhagavad Gita Slokas for Mental Wellness

 Users can enter a chapter number and sloka number to access specific Bhagavad Gita verses.
 The website connects to the Geetha API, which fetches the requested sloka.

15
  The sloka is displayed in original Sanskrit and is translated into the user’s preferred language
using Google Translator API.
 The translated sloka is presented with an explanation, helping users apply its wisdom in daily life.
Why this step?
Enhances mental well-being through spiritual guidance.
Encourages positive thinking, stress reduction, and emotional resilience.
4. Multilingual Speech & Text Support
 The chatbot and the Bhagavad Gita module both support speech-to-text and text-to-speech
functionalities.
 Users can interact using voice commands in their preferred language.
 Responses are translated using Google Translate API, ensuring language flexibility.
Why this step?
Improves user accessibility, even for non-English speakers.
Makes the platform more inclusive and user-friendly.

5. Health Recommendations & Holistic Wellness Guidance

 The chatbot provides comprehensive health solutions, integrating:
 Dietary recommendations (balanced nutrition, Ayurvedic diet plans).
 Exercise tips (yoga, meditation, breathing techniques).
 Herbal remedies (Ashwagandha, Triphala, Bhringaraj, Shatavari).
 Lifestyle modifications (daily routines, stress management).
 The system learns from user feedback, improving recommendations over time.
Why this step?
Provides a complete health improvement plan.
Combines scientific and traditional healing approaches.

6. Continuous Learning & System Enhancement

 User interactions are analysed to improve the chatbot’s response accuracy.
 Expert feedback is incorporated to enhance medical credibility.
 The system is updated regularly to introduce new features and optimize performance.

16
Why this step?
Ensures the chatbot remains accurate, reliable, and up-to-date.
Helps in continuous refinement based on user needs.

4.5 Performance Measure:

To evaluate the effectiveness and efficiency of our health improvement platform, we establish key
performance indicators (KPIs) focusing on user engagement, response accuracy, system efficiency, and
health outcome improvements. These measures ensure that our chatbot-driven system delivers a seamless,
reliable, and personalized experience.
1. User Engagement & Interaction
To assess how effectively users interact with the chatbot and website:
 Active User Retention: Targeting a 70-80% retention rate of users returning for multiple sessions.
 Average Session Duration: Users should engage for at least 5 minutes per session.
 Response Time: Ensuring chatbot replies within 1 second to maintain a smooth experience.
 User Satisfaction Score: Collecting feedback to maintain a ≥8/10 satisfaction rating.
2. Chatbot Accuracy & Personalization
To measure the chatbot’s ability to provide relevant and accurate responses:
 Response Accuracy: Achieving at least 90% correctness in chatbot responses.
 Personalization Score: Evaluating how well responses align with user-specific needs (≥85%
relevance).
 Misunderstood Queries: Keeping the percentage of failed or misunderstood queries below 5%.
 Multilingual Accuracy: Ensuring translations are at least 90% accurate across supported
languages.
3. System Performance & Scalability
To maintain a high-performing and stable system:
 System Uptime: Maintaining 99.9% availability to ensure uninterrupted service.
 Latency: Keeping response processing time below 1 second.
 Database Query Speed: Retrieving user medical history and health data within 500ms.
 API Response Time: Ensuring all external APIs (Aadhaar, Geetha API, etc.) respond within 1
second.
4. Effectiveness of Health Recommendations
17
To ensure that the chatbot's health suggestions provide real value:
 Health Improvement Feedback: More than 80% of users report a positive impact.
 Success Rate of Recommendations: At least 75% adherence to suggested remedies.
 Expert Validation Score: Maintaining an approval rating of 4/5 or higher from
Ayurvedic and medical professionals.
 Repeat Consultations: At least 60% of users returning for further health advice.

4.6 Advantages of Our Health Improvement Platform

Our platform leverages AI-powered chatbots, traditional Ayurvedic principles, and modern medical
insights to provide a personalized, accessible, and effective approach to health and wellness. The key
advantages of our system are:
1. Personalized Health Assistance
 Uses user-specific data (age, weight, health history) to provide tailored recommendations.
 Adapts responses based on dosha analysis, nutritional deficiencies, and lifestyle habits.
 Supports multi-language communication, ensuring accessibility for a diverse user base.
2. Integration of Traditional and Modern Medicine
 Combines Ayurvedic remedies (e.g., herbs like Ashwagandha, Triphala) with modern dietary
recommendations.
 Provides a holistic approach by suggesting home remedies, diet plans, and lifestyle modifications.
 Validates recommendations with expert-reviewed medical research, ensuring credibility.
3. Intelligent Chatbot with High Engagement
 Built using Rasa AI with speech-to-text and translation APIs, making it highly interactive.
 Engages users with empathetic responses and asks clarifying questions to enhance the experience.
 Delivers accurate responses with a 90%+ correctness score, improving reliability.
4. Efficient and Scalable System
 Utilizes fast API integrations (Aadhaar API, Geetha API) for seamless authentication and query
processing.
 Ensures 99.9% uptime, reducing the chances of system downtime.
 Can scale to handle a large number of users without performance degradation.
5. Quick and Reliable Health Insights
 Provides instant feedback on health concerns, reducing dependency on physical consultations.
18
  Offers scientifically backed explanations for common health myths, enhancing awareness.
 Encourages preventive healthcare, reducing the risk of chronic diseases.
6. User-Friendly and Accessible
 Simple and intuitive interface, accessible via voice and text interactions.
 Breaks complex health data into easy-to-understand suggestions.
 Supports users with limited medical knowledge by offering clear, structured guidance.

7. Data Security and Privacy

 Uses Aadhaar-based authentication for secure access to personalized data.
 Follows strict data privacy policies to protect user health records.
 Stores data in a secure, encrypted database, ensuring confidentiality.
8. Positive Impact on Mental and Physical Well-being
 Includes stress management techniques inspired by Bhagavad Gita teachings.
 Recommends yoga, meditation, and breathing exercises for mental and emotional balance.
 Helps users make sustainable lifestyle changes, promoting long-term health benefits.

19
 5. Dataset and Algorithm
5.1 Dataset
Our platform utilizes two primary datasets:
1. Health Knowledge Dataset
This dataset contains information on diseases, symptoms, precautions, and Ayurvedic remedies. It helps
the chatbot provide accurate responses to user queries.
 Diseases and Symptoms: Covers a wide range of illnesses, their causes, and possible treatments.
 Precautionary Measures: Lifestyle and dietary recommendations for disease prevention.
 Ayurvedic Remedies: Traditional herbal solutions and their scientific explanations.
 Medical Research Integration: Modern medical insights supporting traditional practices.
2. User Information Dataset
This dataset is used to personalize health recommendations.
 User Profile: Name, age, gender, weight.
 Health History: Past illnesses, ongoing medications, allergies, chronic conditions.
 Lifestyle Factors: Dietary habits, activity levels, stress indicators.
 Engagement Data: Chatbot interaction patterns for improved recommendations.

5.2 Algorithm
Our system is built on a combination of AI techniques for natural language processing, machine
learning, and personalized response generation. The core algorithms include:
1. Chatbot Framework: Rasa
 Intent Recognition: Identifies the user’s intent (e.g., symptom inquiry, diet advice).
 Entity Extraction: Extracts relevant health-related details (e.g., "headache" → symptom).
 Context Handling: Maintains conversation history to provide contextual responses.

20
2. AI Model: LLaMA 3.3
 Deep Learning-based Language Model: Enhances chatbot response accuracy.
 Medical Knowledge Embedding: Trained on medical research and Ayurvedic texts.
 Context-Aware Response Generation: Ensures meaningful, user-specific replies.

3. Personalization Module
 User Profile Matching: Maps symptoms and conditions to past health records.
 Customized Suggestions: Generates diet plans, lifestyle tips, and precautions.
 Adaptive Learning: Improves responses based on user feedback and preferences.
4. Speech & Translation APIs
 Speech-to-Text API: Converts voice inputs into text for better accessibility.
 Text Translation API: Translates responses into the user’s preferred language.
5. Secure Data Handling
 Aadhaar API Authentication: Ensures secure and verified user access.
 Encrypted Data Storage: Protects sensitive health information.

21
 6. Code Documentation

6.1 React Application Structure

a. Main Page

function App() {

return (

<Router>

<NavBar />

<Routes>

{/* Default route for Home page */}

<Route path="/" element={<Home />} />

<Route path="/home" element={<Home />} />

<Route path="/myths" element={<Myths />} />

<Route path="/geetha" element={<Geetha />} />

<Route path="/about" element={<About />} />

<Route path="/language-selection" element={<LanguageSelection />} />

<Route path="/chat" element={<ChatPage />} />

<Route path="/chat_dynamic" element={<Chat />} />

<Route path="/communication" element={<CommunicationPage />} />

22
 </Routes>

<Footer />

</Router>

);

b. Fetching Data from API

const fetchSloka = async () => {

try {

const response = await fetch(`https://spiritual-path.onrender.com/api/sloka/${chapter}/${sloka}`);

const data = await response.json();

setSlokaText(data.verse || "Sloka not available.");

setBhavamText(data.translation || "Bhavam not available.");

setAudioSrc(data.audio_link || "");

} catch (error) {

console.error("Error fetching sloka:", error);

setSlokaText("Error fetching sloka.");

setBhavamText("Error fetching bhavam.");

};

6.2 Flask Application

a. Install Dependencies
pip install flask requests google-auth google-auth-oauthlib google-auth-http

b. Google API
23
 # Set the path to the Google Cloud credentials (ensure this file exists)

os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "sa_speechtotext.json"

client = texttospeech_v1.TextToSpeechClient()

# Prepare the text for conversion to speech

synthesis_input = texttospeech_v1.SynthesisInput(text=text)

# Define voice selection parameters

voice = texttospeech_v1.VoiceSelectionParams(

language_code=language_code,

ssml_gender=texttospeech_v1.SsmlVoiceGender.FEMALE

# Define audio configuration (MP3 encoding)

audio_config = texttospeech_v1.AudioConfig(

audio_encoding=texttospeech_v1.AudioEncoding.MP3

c. Model Training

# Retrieve conversation history for the user

if aadhaar_number not in user_conversations:

user_conversations[aadhaar_number] = [

{"role": "system", "content": "You are an Ayurvedic health advisor who provides context-aware
responses."},

{"role": "user", "content": f"User Details: {user_details}"},

data = request.json

messages = data.get("messages", [])

if messages:

user_input = messages[-1]["content"]
24
 user_conversations[aadhaar_number].append({"role": "user", "content": user_input})

response = requests.post(

GROQ_API_URL,

headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"},

json={"model": "llama-3.3-70b-versatile", "messages": user_conversations[aadhaar_number],

"temperature": 0.7},

6.3 Rasa
a. Stories

- story: thyroid(hyper) known dite-yes senario-1

steps:

- intent: thyroid_condition_known

- action: utter_ask_thyroid_type

- intent: thyroid_type_hyper

- action: utter_ayurvedic_medicine_tyroid_hyper

- action: utter_ask_for_diet_plan

- intent: affirm

- action: utter_home_remedies_tyroid_hyper

- action: utter_food_diet_tyriod_hyper

- action: utter_take_care

- story: thyroid(hyper) known dite-no senario-2

steps:

- intent: thyroid_condition_known

- action: utter_ask_thyroid_type

- intent: thyroid_type_hyper

- action: utter_ayurvedic_medicine_tyroid_hyper

25
 - action: utter_ask_for_diet_plan

- intent: deny

- action: utter_take_care

b. Domine File

utter_head_whole_response:

- text: "This often indicates Stress, Daily Routine Change, or Acidity.\n\ Ayurvedic remedies:\n\
Stress – 1. Practice Pranayama (breathing exercises) 2. Use Brahmi oil for head massage 3. Drink
Ashwagandha tea\n\ Daily Routine Change – 1. Drink warm water with honey 2. Take turmeric milk
before bedtime\n\ Acidity – 1. Drink Aloe Vera juice 2. Have a spoonful of ghee on an empty
stomach"

utter_head_sides_response:

- text: "This often indicates Migraine or Sinus.\n\ Ayurvedic remedies:\n\ Migraine – 1. Apply cold
compress with peppermint oil 2. Drink ginger tea 3. Take Triphala powder\n\ Sinus – 1. Use
eucalyptus oil for steam inhalation 2. Drink turmeric milk 3. Gargle with salt water"

utter_head_crown_response:

- text: "This often indicates Lack of Sleep, Brain Freeze, or Tension.\n\ Ayurvedic remedies:\n\ Lack
of Sleep – 1. Drink warm milk with nutmeg powder before sleep 2. Practice meditation 3. Consume
Ashwagandha\n\ Brain Freeze – 1. Drink ginger tea 2. Cover your head with a warm cloth\n\ Tension
– 1. Massage your scalp with coconut oil 2. Drink chamomile tea"

utter_head_eyes_response:

- text: "This often indicates Change in Eye number or Stress.\n\ Ayurvedic remedies:\n\ Change in
Eye Number – 1. Use Triphala eyewash 2. Drink carrot juice with a pinch of black salt\n\ Stress – 1.
Drink Brahmi tea 2. Massage the eyes with sesame oil"

utter_flu_response:

- text: "This often is caused by high fever, headache, muscle aches, cough, runny nose, or vomiting
sensation.\n\ Ayurvedic remedies:\n\ 1. Drink hot ginger tea with honey 2. Gargle with turmeric water
3. Inhale steam with eucalyptus oil 4. Take a spoonful of Tulsi leaf juice"

utter_vision_types:

- text: "1. Did you use contact lenses for a long time?\n\ 2. Had lack of sleep?\n\ 3. Have conditions
like arthritis, diabetes or pregnancy? Yes | No"
26
 utter_dry_eyes_response:

- text: "You have symptoms of Dry Eyes.\n\ Ayurvedic remedies:\n\ 1. Apply almond oil around your
eyes 2. Use Triphala eyewash 3. Drink coconut water to hydrate your body"

7. Outputs:
Home page:

Fig 7.1 This is

the home page of Sanatana Vaidy from which we can navigate to other pages of website

Slokas page:

27
Fig:7.2.1 On clicking on sloka we navigate to this page in which we have to enter chapter number and
sloka number which we want to learn or study

Fig:7.2.2 After clicking on submit we can see the sloka and its meaning

Myths Page:

Fig:7.3 On clicking Myths we can navigate to myths page in which we can see some
myths and their scientific reas
28
Chat Page:

Fig:7.3 Clicking on chat we go to authentication page by using patient aadhar number patient can log-in
to the chat section.

Fig: 7.4.1

29
Fig: 7.4.2

Fig: 7.4.3

30
 Fig: 7.4.4

In figure 7.4.1, 7.4.2, 7.4.3, 7.4.4 we can tell the symptoms to the bot and it will give the suggestion
according to the symptoms which are told by the user.

Fig: 7.4.5

Fig: 7.4.6
31
 Fig: 7.4.7
In figure 7.4.5, 7.4.6, 7.4.7 This conversation is about the existing user where bot already knows the
medical condition of the patient and some personal details.

8. Testing

8.1 INTRODUCTION TO TESTING

Software Testing is defined as an activity to check whether the actual results match the
expected results and to ensure that the software system is Defect free. It involves the execution of a
software component or system component to evaluate one or more properties of interest. It is required
for evaluating the system. This phase is the critical phase of software quality assurance and presents
the ultimate view of coding.
Importance of Testing
The importance of software testing is imperative. A lot of times this process is skipped, therefore, the
product and business might suffer. To understand the importance of testing, here are some key points
to explain
32
  Software Testing saves money
 Provides Security
 Improves Product Quality
 Customer satisfaction
Testing is of different ways the main idea behind the testing is to reduce the errors and doit with a
minimum time and effort.
Benefits of Testing
 Cost-Effective: It is one of the important advantages of software testing. Testing any IT
project on time helps you to save your money for the long term. In case if the bugs caught in
the earlier stage of software testing, it costs less to fix.
 Security: It is the most vulnerable and sensitive benefit of software testing. People are
looking for trusted products. It helps in removing risks and problems earlier.
 Product quality: It is an essential requirement of any software product. Testing ensures a
quality product is delivered to customers.
 Customer Satisfaction: The main aim of any product is to give satisfaction to their
customers. UI/UX Testing ensures the best user experience.

8.2 Unit testing

During the testing phase, we consulted Ayurvedic doctors to evaluate the accuracy of our
chatbot’s suggestions based on user-provided symptoms. The doctors provided positive feedback,
confirming that the recommendations were relevant and aligned with Ayurvedic principles.

Additionally, we incorporated suggestions from an online herbal website to enhance the chatbot’s
responses. The recommendations generated by our system were cross-referenced with the website’s
information, ensuring consistency and accuracy.

To validate the accuracy of our scripture-based feature, we randomly selected chapter and sloka
numbers. The chatbot successfully retrieved the correct slokas and translated them into different
languages. This process was repeated 15 to 20 times, consistently yielding accurate results.

33
 Lastly, we tested the myths page to ensure proper functionality. The interactive cards flipped
correctly as intended, and after conducting 20 trials, we confirmed that the feature was functioning
smoothly without any errors.

9. Conclusion

Sanatana vaidy is designed to provide accessible and personalized healthcare by integrating

multilingual support and holistic health guidance. By combining modern medical science with the ethical
principles of Ayurveda, it offers a balanced approach to health and well-being. Users can also listen to
Bhagavad Gita slokas, promoting mental peace and relaxation alongside physical wellness.
Through fact-based education and an interactive user experience, Sanatana vaidy ensures that individuals
receive accurate and reliable health information in an engaging way. By empowering users with
knowledge, the system enables them to make informed health decisions, ultimately improving their
overall quality of life.

34
 10. Future scope

In the future, we aim to develop a 3D model of the human body to enhance user interaction with
our chatbot. This feature will allow individuals, especially those unfamiliar with medical terminology, to
visually identify painful areas by selecting them on the model. By interacting with a lifelike
representation of the human body, users can convey their symptoms more effectively, enabling the
chatbot to understand their concerns with greater accuracy. This will improve the precision of symptom
analysis and ensure that the chatbot provides more relevant Ayurvedic recommendations.

Additionally, we plan to integrate this feature into a real-time model, making the interaction more
dynamic and responsive. Users will be able to rotate, zoom, and explore different sections of the body,

35
helping them pinpoint their issues more intuitively. This real-time integration will also open opportunities
for future advancements, such as augmented reality (AR) and virtual reality (VR) applications, which
could further enhance user engagement and understanding.

By combining advanced visualization with AI-powered analysis, we aim to make healthcare

interactions more accessible, especially for those who may struggle with textual descriptions of their
symptoms. This innovation has the potential to bridge the gap between traditional Ayurvedic knowledge
and modern technology, ensuring that users receive a seamless and user-friendly diagnostic experience.

11.User Manual

Using the Chatbot

Logged-in Users:

 Users who have logged into our website with their Aadhaar number can access a personalized
chatbot experience.
 The chatbot saves patient health data, allowing it to provide suggestions based on previous health
history.
 After logging in, users can see the chatbot interface and start describing their symptoms.
 The chatbot responds accordingly, offering suggestions based on Ayurvedic principles.
36
Guest Users (Without Login):

 Users can access the chatbot without logging in by selecting their preferred language on the
website.
 After selecting the language, users can begin a conversation with the chatbot.
 The chatbot provides health suggestions based on the symptoms shared.
 A voice-over feature, represented by a speaker symbol, is available for audio responses.

Using the Slokas Page

 The Slokas page can be accessed through the "Sloka" option on the home page header.
 There are two text boxes:
 One for entering the chapter number
 One for entering the sloka number in that specific chapter
 After entering the values, the sloka and its meaning are displayed on the screen.
 By clicking on the speaker symbol, users can listen to the sloka.
 A dropdown menu allows users to change the display language of the sloka and its meaning.

Using the Myths Page

 The Myths page can be accessed by clicking on "Myths" in the home page header.
 On this page, users can see various myths presented in an interactive card format.
 By simply moving the cursor over a card, the scientific reason behind the myth is displayed.

12. References

[1] https://www.researchgate.net/profile/Umapathi-Nagappan/publication/372658029_Section_A-
Research_paper_Personal_Healthcare_Chatbot_for_Medical_Suggestions_Using_Artificial_Intelligence_
and_Machine_Learning_Eur/links/65279ef33fa934104b1911bd/Section-A-Research-paper-Personal-
Healthcare-Chatbot-for-Medical-Suggestions-Using-Artificial-Intelligence-and-Machine-Learning-
Eur.pdf

[2] S. A. Ahmed, G. Gahlawat, F. Imam and G. Saranya, "AyurChat: Empowering Health through AI-
Driven Ayurvedic Guidance," 2024 15th International Conference on Computing Communication and
Networking Technologies (ICCCNT), Kamand, India, 2024, pp. 1-6, doi:
37
10.1109/ICCCNT61001.2024.10724133.

[3]https://www.researchgate.net/publication/
344983576_Chatbot_for_Healthcare_System_Using_Artificial_Intelligence/link/
5fd99fcc92851c13fe8c25ae/download?
_tp=eyJjb250ZXh0Ijp7ImZpcnN0UGFnZSI6InB1YmxpY2F0aW9uIiwicGFnZSI6InB1YmxpY2F0aW9u
In19

[4] https://medium.com/dair-ai/papers-explained-mistral-7b-b9632dedf580

[4] https://academic.oup.com/bioinformatics/article/36/4/1234/5566506

[5] https://www.ayurai.io/?utm_source=chatgpt.com#features

[6] https://towardsdatascience.com/blender-bot-part-3-the-many-architectures-a6ebff0d75a6

[7] https://huggingface.co/docs/transformers/en/model_doc/blenderbot

38