DocTalk

(a healthcare assistant)
VII SEMESTER SYNOPSIS REPORT

Submitted in partial fulfillment of the requirement of the degree of

BACHELORS OF TECHNOLOGY
to

The NorthCap University

by

Harshit 21CSU036

Under the supervision of

Dr. Vandana

Department of Computer Science and Engineering

Department of Computer Science and Engineering

School of Engineering and Technology

The NorthCap University, Gurugram- 122001, India

Session 2024-2025

CERTIFICATE
This is to certify that the Project Synopsis entitled, “DocTalk” submitted by “Harshit” to The

NorthCap University, Gurugram, India, is a record of bonafide synopsis work carried out by

them under my supervision and guidance and is worthy of consideration for the partial fulfilment

of the degree of Bachelor of Technology in Computer Science and Engineering of the

University.

Dr. Vandana

Date: ………………

INDEX

1. Abstract
In recent years, the integration of Artificial Intelligence (AI) into healthcare has significantly
reshaped how initial medical consultations are conducted. Among the most promising
 innovations in this space is the healthcare assistant chatbot, designed to provide immediate
responses to basic health inquiries and guide users through early symptom assessment. This
project, titled *DocTalk*, centers on developing an intelligent, user-friendly chatbot that uses
Natural Language Processing (NLP) and machine learning—specifically decision tree algorithms
—to analyze user-described symptoms and suggest potential health conditions.

The chatbot aims to bridge the gap between patients and professional medical help by offering
initial guidance, recommending whether a visit to a healthcare provider is necessary, and even
advising on the type of specialist to consult. As a result, *DocTalk* not only improves
accessibility for users who might otherwise delay medical consultation due to cost or distance,
but also supports healthcare institutions in triaging non-critical cases efficiently. Ultimately, the
goal is to create a scalable, affordable, and intelligent digital assistant that hospitals can
integrate into their systems, or that can operate independently as a web-based health advisory
tool.

2. Introduction (description of broad topic)

a. Background
b. Feasibility Study

3. Study of Existing Solution (implementation-based project)/Literature Survey (research

based)
4. Comparison with existing software solutions (implementation based)/Literature (research
based)
5. Gap Analysis
6. Problem Statement
7. Objectives
8. Outcomes
9. Gantt Chart
10. Responsibility Chart
11. References
12. Annexure I: Front page of plagiarism report by guide
13. Annexure II: Screenshots of Faculty mail / comments from guide

Important Notes:
*Overall and AI Plagiarism of the report should be equal to or less than 10%: provided by guide
*Font style: Times New Roman, Font size: 12, 14 (heading)
*Synopsis report size should be of at least 10-12 pages
 ABSTRACT
In recent years, the integration of Artificial Intelligence (AI) into healthcare has significantly reshaped
how initial medical consultations are conducted. Among the most promising innovations in this space is
the healthcare assistant chatbot, designed to provide immediate responses to basic health inquiries and
guide users through early symptom assessment. This project, titled *DocTalk*, centers on developing an
intelligent, user-friendly chatbot that uses Natural Language Processing (NLP) and machine learning—
specifically decision tree algorithms—to analyze user-described symptoms and suggest potential health
conditions.

The chatbot aims to bridge the gap between patients and professional medical help by offering initial
guidance, recommending whether a visit to a healthcare provider is necessary, and even advising on the
type of specialist to consult. As a result, *DocTalk* not only improves accessibility for users who might
otherwise delay medical consultation due to cost or distance, but also supports healthcare institutions in
triaging non-critical cases efficiently. Ultimately, the goal is to create a scalable, affordable, and
intelligent digital assistant that hospitals can integrate into their systems, or that can operate
independently as a web-based health advisory tool.

Recently, healthcare assistant chatbots emerged as the newest service tool to aid in patient
care through the automation of routine medical answers. The objective of this project is to
design a chatbot that can engage with a user and prescribe remedies for his or her symptoms
based on pre-existing conditions. Leverage NLP and machine learning to make a chatbot an
accessible healthcare tool for simple diagnosis, providing information related to symptoms,
drugs, and when to consult with professional healthcare.
The objective of this project is to prepare an initial stage of medical consultation which can be
used by hospitals or as a standalone web-based tool.
 INTRODUCTION
The rapid evolution of Artificial Intelligence (AI) technologies has given rise to numerous applications
that are transforming industries globally, with healthcare being one of the most impactful sectors. One
such development is the emergence of AI-driven chatbots designed to mimic human conversation and
provide instant support to users. In the medical domain, these chatbots serve as virtual health
assistants, delivering preliminary guidance, answering frequently asked health-related questions, and
helping users understand their symptoms—all while reducing the burden on healthcare professionals.

Healthcare systems, especially in densely populated or rural areas, often struggle with a high patient
load, limited staffing, and restricted access to timely consultations. With the advent of chatbots like
*DocTalk*, users can receive reliable and prompt information, enabling them to make informed
decisions about their health. These systems can advise whether symptoms indicate a minor illness that
can be managed at home or if they require urgent medical attention, thereby acting as an intelligent
filter before a professional consultation.

The recent COVID-19 pandemic further highlighted the importance of remote health tools. During times
when physical visits to hospitals posed risks or were limited, AI-based chatbots became instrumental in
providing round-the-clock support. Not only did they assist in symptom evaluation, but they also played
a critical role in reducing panic by disseminating accurate information.

*DocTalk* is designed with this vision—to serve as a first line of digital healthcare support by harnessing
the capabilities of machine learning and NLP. Its core objective is to democratize access to basic
healthcare information by offering real-time, reliable, and contextually aware health assessments in a
way that is simple and free to use.

With the continuously advancing advances in Artificial Intelligence (AI), chatbots have come
as solutions to the problems healthcare systems face around the world and in managing
volume in taking care of patients, providing timely assistance, and lessening the workload
for healthcare providers. With this regard, health assistant chatbots give users real-time
health information regarding symptoms and recommends direction according to user input.
These chatbots are to mimic a conversation between the user and a medical professional to
assist and guide people on non-critical health issues and also to help people make a decision
about whether to go to a doctor's office or not.
In the last months of the COVID-19 pandemic, there has been widespread use of healthcare
chatbots as governments and healthcare providers have directed their attention towards AI-
based solutions in support of unprecedented medical consultations demand. Chatbots could
potentially enhance healthcare accessibility by helping the patient in far-flung areas and
also support the health care practitioners with automated responses. Concerns of this
project are the Health Assistant Chatbot, which may be used as a diagnostic tool for sending
recommendations based on symptoms provided for ease of use and reliability.

a. Background
Constantly increasing patient populations, complex clinical presentations of medical
conditions, and limited health care resources intensify pressures in the global healthcare
sector. Against this background, AI-driven solutions are implemented to help healthcare
professionals manage the system more efficiently. The momentum, however, has already

been gained in other sectors, such as customer service and finance, and now, healthcare
has also begun to take off.

These healthcare assistant bots perform lots of jobs-from answering simple questions of
health to symptom analysis and linking patients to health experts. These chatbots use NLP
for patient questions and return responses that basically mimic health professionals' advice.
Being an automated tool, the cost-effectiveness of healthcare chatbots scales easily to
advocate the solution of health gaps in resource-restricted settings. On the other hand, for
this project, the Healthcare Assistant Chatbot will more focus on common medical
conditions and will advise using responses it has already predetermined in its database.
Besides, it may enlighten one on when a consultation from a professional doctor is
required.

b. Feasibility study

● Technical Feasibility:

1. Technology Used:
a. Python: Backend logic and machine learning model implementation.
b. Tkinter: For creating the graphical user interface (GUI).
c. Scikit-learn: For building and training the decision tree model.
d. Pandas and NumPy: For handling and analyzing the data.
e. Matplotlib: For visualizing results, if applicable.
 2. Key Issues:
a. Optimizing the decision tree classifier for accurate predictions.
b. Ensuring the GUI is user-friendly and responsive.
c. Integrating datasets (Training.csv and Testing.csv) effectively.

3. Project Timeline: Estimated at 3-5 months for development, testing, and

deployment.

● Financial Feasibility

1. Overall expenses: Minimal costs, as it primarily involves open-source

tools and frameworks.
2. Financing: External funding is not required.

3. Revenue: Could be monetized through partnerships with healthcare

providers or advertisements once deployed.

● Market Feasibility

Demand:
 Increasing demand for AI-driven healthcare solutions, particularly chatbots capable of
providing initial diagnoses and recommendations.
 A growing focus on accessible healthcare through digital platforms.
Differentiation:
 The chatbot’s ability to suggest diagnoses and recommend specialists distinguishes it
from generic healthcare chatbots.
Competition:
 While some existing solutions like WebMD or symptom-checking apps exist, they are
often not specialized for localized or easily accessible recommendations.
Potential Users:
 Individuals seeking quick and accessible medical advice.
 Medical practitioners interested in integrating chatbots for triaging patients.
 STUDY OF EXISTING SOLUTION

Several existing tools and platforms aim to provide healthcare assistance through AI-driven
chatbots or symptom-checkers. However, each has its limitations, particularly in terms of
accuracy, accessibility, and affordability. Below are some notable examples:

Existing Platforms
1. WebMD Symptom Checker
o Features: Offers basic symptom checking and suggests possible conditions.
o Limitations: Primarily web-based, lacks personalized doctor recommendations,
and does not integrate machine learning for advanced diagnostics.
2. Ada Health
o Features: AI-powered symptom assessment with a focus on medical accuracy.
o Limitations: Limited free functionality and no specialist recommendations.
3. Healthily
o Features: Self-care advice with symptom analysis and a user-friendly interface.
o Limitations: No real-time decision-making or doctor integration.
 COMPARISON WITH EXISTING SOLUTIONS

Accuracy: Many existing solutions rely on rule-based systems, which may not adapt well to
complex datasets. The decision tree classifier in our project enables improved accuracy by
learning patterns directly from the data.

Specialist Recommendations: Unlike other tools, your chatbot provides actionable

recommendations, connecting users to specific doctors.

Affordability: Our project is free to use, making it accessible to a larger audience compared to
premium services like WebMD and Ada Health.

Customization: By training the model on datasets tailored to common local conditions, your
chatbot can deliver highly relevant insights.
 The table below compares and contrasts “Vani Sanket” with the top existing solutions:

Tool Accuracy Languages Specialist Pricing Special Features

Supported Recomme
ndations

WebMD medium English No Free Basic symptom

checker, generic
advice

Ada Health High English, No Freemiu AI-based

limited m symptom
others analysis, paid
advanced features

Healthily high English No Freemiu Self-care advice,

m symptom analysis

DocTalk high English Yes Free Symptom-based

diagnosis,
decision tree
model, doctor
recommendations
 GAP ANALYSIS

After evaluating existing healthcare chatbot solutions, the following gaps were identified:

1. Limited Accuracy of Existing Solutions

o Most existing platforms rely on static rule-based algorithms, resulting in generic
or inaccurate diagnoses. They lack dynamic machine learning models like
decision trees to analyze complex symptom combinations.
2. No Specialist Recommendations
o Current tools only suggest potential conditions without recommending
appropriate specialists, leaving users without actionable next steps for their health
concerns.
3. Accessibility Constraints
o Many platforms operate on a freemium model, restricting full functionality to paid
users. This limits accessibility for individuals seeking free, comprehensive
healthcare assistance.
4. Poor User Experience
o Some solutions lack a user-friendly interface, making them difficult for non-tech-
savvy users to navigate.
o Multilingual support is often absent, narrowing their reach.
5. Lack of Integration with Local Healthcare Needs
o Platforms often do not consider localized health conditions or tailor
recommendations for specific regions, reducing their effectiveness for diverse
populations.

How DocTalk Addresses These Gaps

 Implements a decision tree classifier to ensure accurate and tailored diagnoses.

  Provides specialist recommendations, making healthcare advice actionable.
 Offers a completely free-to-use platform, improving accessibility for all.
 Features a simple and intuitive Tkinter GUI, enhancing user experience.
 Focuses on training datasets relevant to local health concerns for better relevance.

PROBLEM STATEMENT

The problem statement can be defined as:

The current state of healthcare assistance through AI-driven platforms highlights several
challenges:
 1. Accuracy Issues:
Existing platforms for symptom checking rely heavily on static, rule-based
algorithms, leading to generic and inaccurate diagnoses.
2. Absence of Specialist Guidance:
Most tools fail to provide actionable recommendations, such as suggesting
appropriate medical specialists based on the diagnosed condition. This leaves users
uncertain about the next steps for their health concerns.
3. Accessibility Limitations:
The freemium models employed by many solutions restrict functionality, making
comprehensive healthcare assistance unavailable to those unable to afford
premium services.
4. Relevance to Local Healthcare Needs:
Platforms do not address localized health concerns or provide tailored solutions for
specific regions, reducing their effectiveness in diverse populations.
5. Complex User Interfaces:
Many tools suffer from poor user interface designs, making them challenging for
non-technical users to navigate efficiently.

OBJECTIVES

In the process of creating this project, our main objectives for this project would
broadly be:-
  Develop an accurate symptom diagnosis chatbot using a decision tree
classifier.
 Provide specialist recommendations based on diagnosed conditions.
 Create a user-friendly GUI with Tkinter for seamless interaction.
 Ensure free and accessible healthcare assistance.
 Optimize real-time responses for efficient diagnosis.

TOOLS/PLATFORM USED

While developing this project, multiple technologies and tools were employed to ensure the
proper functioning of the Health Care Chat-Bot.
Programming Languages
 Python: Used for backend logic and implementation of the decision tree classifier.
Frameworks and Libraries
 Tkinter: For creating a graphical user interface (GUI).
 Scikit-learn: For building and training the decision tree model.
 NumPy and Pandas: For data manipulation and analysis.
  Matplotlib: For visualizing the results.
Development and Deployment Tools
 Jupyter Notebook: For testing and debugging the model.
 VS Code: Integrated development environment (IDE) used.
 Git: Version control system for project management.
External Resources
 Training and Testing Datasets: CSV files containing symptom and disease data.

DESIGN METHODOLOGY

1. Symptom Analysis and Diagnosis

The primary component of DocTalk is the accurate analysis of symptoms provided

by the user. This involves two main steps:
 Symptom Input and Processing: Users input their symptoms via a simple
graphical interface built using Tkinter. The input is then preprocessed to
remove redundancies, normalize the data, and prepare it for machine
learning analysis.
 Decision Tree Classification: Using Scikit-learn, a decision tree classifier
processes the symptoms to predict potential health conditions. This model is
trained on labeled datasets (Training.csv and Testing.csv) to ensure
reliability and accuracy.

2. Specialist Recommendations

Based on the diagnosis results, the system recommends the appropriate type of
medical specialist to the user.
 A mapping module is implemented to link diagnoses with medical
specialties. For instance, respiratory conditions are mapped to
pulmonologists, and skin-related issues to dermatologists.
3. User Interface Development

The project incorporates a user-friendly interface with the following features:

 A text box for symptom input.
 Dropdown menus for optional selections like additional symptoms or
preferences.
 A display area for showing diagnosis results and specialist
recommendations.

4. Data Integration and Validation

The datasets used (Training.csv and Testing.csv) are preprocessed for:

 Feature Selection: Ensuring only the most relevant symptoms are used for
training to improve model efficiency.
 Validation Split: Dividing the data into training and validation sets for
unbiased model performance evaluation.

5. Workflow Optimization

The system is designed to provide real-time responses with minimal computational

overhead:
 Parallel processing techniques are applied to handle simultaneous user
requests.
 Model predictions are cached for frequently queried conditions to enhance
response time.
 OUTCOME

Upon the successful completion of our seventh-semester project, the outcome is

DocTalk, a reliable and efficient healthcare chatbot. DocTalk aims to bridge the
gap between accessible healthcare and advanced diagnostics by:

1. Providing accurate symptom-based health condition analysis.

2. Recommending appropriate specialists based on the diagnosis.
 3. Offering a user-friendly platform for individuals to seek quick and free
healthcare assistance.
This tool has the potential to improve healthcare accessibility by delivering quick,
actionable, and personalized medical advice to users.

GANTT CHART
TASKS WEEK WEEK WEEK WEEK 4 WEEK WEEK WEEK 7
1 2 3 5 6

defining
problem
stateme
nt

assignin
g tasks

Data
collectio
n and
preproc
essing

Building
decision
tree
model

designi
ng GUI

Model
integrati
on with
GUI

Testing
and
validatio
n
 RESPONSIBILITY CHART

TASK Harshit(tea Mentor

m
member)

defining R I
problem
statement

assigning R I
tasks

Data collection R I
and
preprocessing

Building R I
decision tree
model

designing GUI R I
and
Integrating
model with
GUI

Testing and R I
validation

REFERENCES

1. Scikit-learn Documentation: https://scikit-learn.org/

2. Pandas Documentation: https://pandas.pydata.org/
3. NumPy Documentation: https://numpy.org/
4. Tkinter Official Guide:
https://docs.python.org/3/library/tkinter.html
5. Matplotlib Guide: https://matplotlib.org/
6. Python Official Documentation: https://docs.python.org/3/
7. Training and Testing Dataset: Provided as CSV files
(Training.csv and Testing.csv).
 Annexure I:

Front page of plagiarism report by guide

 Annexure II:

Screenshots of Faculty mail / comments from guide