MediQue- AI based Medical ChatBot

Synopsis

Submitted for the partial fulfillment of the requirement for the award of Degree of

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE AND ENGINEERING

Submitted by: Guided By:

1. Shriya Pandey Prof. Praveen Kaithal
2. Atiya Akhtar
3. Sambodhi Dubey
4. Shubhanshi Bhandari

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

UNIVERSITY INSTITUTE OF TECHNOLOGY
RAJIV GANDHI PROUDYOGIKI VISHWAVIDYALAYA
BHOPAL- 462036
DEC 2021
 DECLARATION BY CANDIDATE

I hereby declare that the work which is being presented in the Synopsis
Major Project “MediQue- Artificial Intelligence based Chatbot” submitted
in partial fulfillment of the requirement for the award of Bachelor Degree
in Computer Science & Engineering. The work has been carried out at
University Institute of Technology, RGPV, Bhopal is an authentic record of
our work carried under the guidance of Prof. Praveen Kaithal, Department
of Computer Science and Engineering,UIT-RGPV, Bhopal.

The matter written in this synopsis has not been submitted by us for the
award of any other degree.

Shriya Pandey
Atiya Akhtar
Sambodhi Dubey
Shubhanshi Bhandari
 ACKNOWLEDGEMENT

It is our privilege to express our sincerest regards to our project guide,

Prof. Praveen Kaithal Sir, for his valuable inputs, able guidance,
encouragement, whole-hearted cooperation and constructive criticism
throughout the duration of our project. We deeply express our sincere
thanks to Prof. Sanjay Silakari Sir and Head of Department Dr. Roopam
Gupta Ma’am for encouraging and allowing us to present the project on
the topic “MEDIQUE-ARTIFICIAL INTELLIGENCE BASED CHATBOT” at
our department premises for the partial fulfilment of the requirements
leading to the award of Bachelor Degree in Computer Science &
Engineering.

We feel to acknowledge our indebtedness and deep sense of gratitude to

our parents and our friends for their constant support.
 ABSTRACT

Healthcare is very important to lead a good life. However, it is very

difficult to obtain a consultation with the doctor for every health problem.
The idea is to create a medical chatbot using Artificial Intelligence that
can diagnose the disease and provide basic details about the disease
before consulting a doctor. This will help to reduce healthcare costs and
improve accessibility to medical knowledge through medical chatbot.
Chatbots are computer programs that use natural language to interact
with users. The chatbot stores the data in the database to identify the
sentence keywords and to make a query decision and answer the
question.
 TABLE OF CONTENT

1. Introduction
2. Literature Survey and related work
3. Problem Description
4. Proposed Work
5. Design and Modeling of Work
6. Description of tools/hardware/software which will be used in the
project
7. Expected Outcome
8. Conclusion and Expected Future work
 INTRODUCTION

Medical advances have got to the point where we can treat a wider
number of diseases, there are new diagnostic and treatment methods
emerging, pharmacology produces an increasingly higher number of
drugs for different diseases, ages, body types, etc. This creates a huge
amount of reference information that doctors must either know or be able
to remember, which requires many years of practice and continuous
learning. At the very least, they need easy access to that information.
Here is what the AI-powered chatbots aim to do. As for patients, chatbots
help them to make decisions about whether they should see a specialist
and if so, which one, provides answers to basic questions and reminds
them to take their medication. Doctors may prescribe medications to help
patients manage various health conditions. But one of the biggest
challenges is getting patients to take their medications as recommended.
A study found that approximately 50% of patients don’t take their
medications properly or forget to do so altogether. Chatbots can not only
be used to dispense medical advice, but also to send daily reminders to
patients.
Medical chatbots are AI-powered conversational solutions that help
patients, insurance companies, and healthcare providers easily connect
with each other. These bots can also play a critical role in making relevant
healthcare information accessible to the right stakeholders, at the right
time. From enhancing patient experience and helping medical
professionals, to improving healthcare processes and unlocking
actionable insights, medical or healthcare chatbots can be used for
achieving various objectives. Medical chatbots are one of the most
matured and influential AI-powered healthcare solutions developed so far.
 Literature Survey and Related Work

There are countless cases where a digital personal assistant or a chatbot

could help physicians, nurses, patients or their families. Better
organization of patient pathways, medication management, help in
emergency situations or with first aid, offering a solution for simpler
medical issues. Health chatbots are also being used to address specific
issues in healthcare.

Some of the innovative Healthcare Chatbots currently serving-

● Your.MD
Designed to help patients find the healthcare information they need
to stay healthy, MD offers various features and functions. However,
what stands apart from all the other features is its symptom checker.
Available on the web and as a stand-alone app, it acts as a personal
health assistant allowing users to check symptoms, ask a question,
or take a health quiz.
Pros: The chatbot uses a vast repository of reliable healthcare
information to offer relevant answers to all user queries. And best of
all, it is available for free.
Cons: Language understanding capabilities are limited in this
chatbot, which negatively impacts the experience.

● Sensely
Sensely helps insurers, healthcare service providers, businesses,
and pharma companies build custom solutions to bridge the gap
between low fidelity chatbots and high-quality human
communications. From underwriting, claims handling, and symptom
 assessment to customer service, employee wellness, and clinical
trials, it offers a wide range of capabilities. It is important to
understand that Sensely is not a chatbot; instead, it offers the
platform and pre-built capabilities required to build different types of
medical chatbots.
Pros: It is possible to build a unique brand persona using various
characters with Sensely.
Cons: We are still a long way from truly embodying empathy with AI.
While the concept is good, its practical implications remain to be
seen.

● NINA (AskNestlé)
Built using Senseforth’s conversational AI platform A.ware, NINA is
one of India’s first AI-powered digital nutritionists. NINA helps young
parents create a daily meal plan for their children. Parents can
customize the meal plan to address specific nutritional requirements,
find innovative recipes to make the fussy eaters happy, set
reminders, and even keep a food journal. Parents can interact with
Nina on Google Assistant or the website. The bot handles around
50,000 queries each month.
Pros: Information provided by the chatbot is very interesting and
contextual.
Cons: Content has been curated keeping Indian audience in mind.
Hence, a lot more contextualization is required to cater to the needs
of a global audience.
● Ada
This medical chatbot brings together doctors, scientists, and industry
pioneers to improve the quality of personal healthcare. Instead of
banking only on a repository of information, Ada compares user
queries with thousands of similar cases. It then analyzes its findings
against the information available in the medical library to craft a
much more relevant and contextual response.
Pros: Ada puts access in the hands of the users.
Cons: Again, limited natural language understanding capabilities
negatively impact the user experience.

● SafedrugBot
This is another good example of how a simple solution like a chatbot
can make a huge difference. SafedrugBot offers guidance to medical
practitioners about the usage of drugs by breastfeeding mothers.
From understanding the side effects of various drugs to finding safer
alternatives, it helps doctors tap into a vast knowledge hub to
augment their capabilities. This chatbot is available on Telegram.
Pros: SafedrugBot helps doctors easily verify drug-related
information.
Cons: It’s available on only one messaging platform and fails to
deliver an omnichannel experience.
UML showing the use cases of MediQue Chatbot:
Data flow diagram of the medical chatbot:
LITERATURE REVIEW
Here the studies are based on recognizing emotions classification using
AI methods. The studies train emotions classification models from a lot of
labelled data based on recurrent neural network (RNN), deep learning,
and convolutional neural network. Linguistic interaction is most important
in counselling using Natural Language Processing (NLP) and
Natural-language generation (NLG) to understand dialogues of users.
Here the multi-modal approach is used for emotion-recognition. They
have collected corpuses to learn semantic information of words and
represent them as vectors using the word vector, synonym knowledge of
lexicals are collected.

A voice recognition chat-bot is developed in this paper. The questions

asked to the bot are not understood and are further processed using the
expert-system of third parties. The webbots are created as web-friends
based on text, a user entertainer. If the program is not only text-based,
but also voice-based equipped, they concentrated on the improved
system here. Here, a two-part process of capturing and analysing an
input signal is required for voice recognition. Recognition of data from the
server response and processing of information. The server used here is a
black box approach based on SOAP. Using an expert system makes it
possible to improve unlimited and autonomous intelligence.

This chatbot aims to make a conversation between human and machine.

Here the system stores the knowledge database to identify the sentence
and making a decision to answer the question. The input sentence will
get the similarity score of input sentences using bigram. The chatbot
knowledge is stored in RDBMS.
The chatbot implemented using pattern comparison in which the order of
the sentence is recognized and saved response pattern. Here the author
describes the implementation of the chatbot Operating system, software,
programming language, database and how results of input and output are
stored. Here the input is taken using text() function and other punctuation
is removed using trim() function and random() function is used to choose
a response from the database. The chatbot is used for an entertainment
purpose.

Here they use n-gram technique for extracting the words from the
sentences. Here n-gram is used for comparison and deduction of the
input with case data using Moro phonemes and phonemes as the
deciding parameter. Probability analysis for the closest match is
performed. The final expression is redirected through an expert system.

The chatbot developed here for healthcare purposes for the android
application. The user sends the text message or voice message using
Google API. Here the user gets only related answer from the chatbot.
SVM algorithm is used to classify the dataset. Here the Porter algorithm
is used to discard unwanted words like suffixes or prefixes.

The different documents served in web, the content is checked by tagging

the dataset using n-gram based low dimensional demonstration, TF-IDF
matrix that generates S, U, and V and finally multiplying the 3 matrices
cosine similarity is calculated.

Here the chatbot is created for the customer service that functions as
public health service. The application uses Ngram, TF-IDF and cosine
similarity. The knowledge base is created for storing the question and
answer. The application clearly shows extracted the keyword from the
question and by using unigram, bigram, and trigram which helps in fast
answering.

Authors have suggested a framework which is using R language to

communicate with Chatbot by using AIML. In this, authors have proposed
a model which is more suitable for the educational purpose, idea behind
this study is to provide more interactive way to student to get connect
university system.
 Problem Description

1) Waiting Time

In many medical situations, time is of essence. People don’t really have

the time to wait for an answer. There is a need to get immediate
response and solutions instead of having to pick up the phone, send an
email, or even wait in line in hospitals or clinics. Patients need to get
immediate responses with an appropriate answer in seconds.

2) Limited Customer Service

When there is a live agent handling customers on phone, chat, or in

person, they can handle only a limited number at the same time. This
means as the customers increase, the number of live agents will have to
be increased, which means the cost will also increase.

3) Delayed Medical Advice

4) Need for Daily Medication Reminders

Doctors may prescribe medications to help patients manage various

health conditions. But one of the biggest challenges is getting patients to
take their medications as recommended. A study found that
approximately 50% of patients don’t take their medications properly or
forget to do so altogether.

5) Appointment Scheduling
Scheduling an appointment over the phone can be extremely
time-consuming for all parties involved. A patient would need to phone
their local clinic and the operator would need to check their schedule to
see what slots are available. All of this back-and-forth can lead to higher
administrative costs.

6) Need to reduce expenses

7) Patient Feedback

The goal for any healthcare provider is to provide exceptional services to

their patients.

8) Invoicing and Payments

The medical billing process allows healthcare providers to receive

compensation for the treatment they provide patients. However, sending
invoices and keeping track of payments can be costly and
time-consuming.

9) Need for Improved Internal Communications

 PROPOSED WORK
The proposed method for developing the system consist of web
application. Firstly, chatbot is created which can help the users to get the
symptoms of their diseases. Then we will add the chatbot link over the
respective hospital website which will help the other people to gain the
information of medical reports.Database of the system helps to store the
records of the users.
We had train a chatbot using chatterbot library and also train the bot to
identify certain types of keywords in order to recognize the user's
intent.This information shall then be forwarded to the backend. The
chatbot can be trained to perform some logical reasoning and responses
without referring to the backend.The proposed method for developing the
system consist of healthcare application. It elaborates the business
aspects of the proposed system. Firstly, chatbot is created which can
help the users to get the symptoms of their diseases.Then hospital
website is integrated which will help the other people to gain the
information of hospital and their staff. Database of the system helps to
store the records of the user.
A sequence diagram simply illustrates interaction between objects in a
sequential order i.e. the order in which these interactions take place.
terms event diagrams or event scenarios to refer to a sequence diagram
used in a proper format. This diagram is widely used by entrepreneurs
and software engineers to formulate and understand requirements for
new and existing projects. The user will give the personal details such as
name, age, date of birth, contact number, etc. Chatbot will ask the
symptoms from the user and by using natural processing language
process it will respond to the disease afterwards it will ask to book the
appointment in the hospital. At last, the health care bot will provide the
appointment details and allow user to exit the portal by ending up the
conversation.
v
 Hardware Requirement:

● i3 Processor Based Computer or higher

● Memory: 1 GB
● Hard Drive: 50 GB
● Monitor
● Internet Connection

Software Requirement:

● Windows 7 or higher
● WAMP Server
● Notepad++
● My SQL 5.6
● Google Chrome Browser
 EXPECTED OUTCOME

● Reduced Waiting Times

Chatbots enable users to get immediate response and

solutions instead of having to pick up the phone, send an email,
or even wait in line. In many medical situations, time is of
essence. People don’t really have the time to wait for an
answer.
A recent report from Ubisend found that 69% of consumers use
chatbots to get an instant answer to a question. Customers can
simply type in a question and the chatbot responds with an
appropriate answer in seconds. This eventually means higher
customer satisfaction and better brand image for your
business.

● Scalable Customer Service

When there is a live agent handling customer on phone, chat,

or in person, they can handle only a limited number at the
same time. This means as the customers increase, the number
of live agents will have to be increased, which means the cost
will also increase
Chatbots don’t have this limitation. They’re built to handle
thousands of enquiries simultaneously. Implementing a chatbot
helps any service provider in the healthcare industry scale their
customer service without compromising on quality and at a
fraction of the cost.

● Timely Medical Advice

A well-implemented chatbot can provide timely medical advice.

NHS 111 is a number that people in the UK can call for urgent
medical problems. An example might be if someone misread a
 prescription and they’re not sure what to do next. People can
also call this number if they don’t know who to contact for
medical help.
The service was launched in 2013, but the system was
becoming overwhelmed. The NHS developed an app with
Babylon Health to reduce the burden on the NHS 111 helpline.
Instead of calling the helpline, people could now simply chat
with an AI-powered chatbot and get medical advice on what
steps to take next.

This is just one example of how chatbots are being used in the
healthcare industry. Another benefit is that chatbots are
available 24/7 to help people in need.
Chatbots can also link to relevant pages. For example, a user
looking for medical alert systems can be directed to the right
page or connect with a representative if they still have
additional questions.

● Daily Medication Reminders

Doctors may prescribe medications to help patients manage

various health conditions. But one of the biggest challenges is
getting patients to take their medications as recommended. A
study found that approximately 50% of patients don’t take their
medications properly or forget to do so altogether.
Chatbots can not only be used to dispense medical advice, but
also to send daily reminders to patients. Florence is a popular
chatbot in healthcare, which acts as a personal medication
reminder and health tracker for those that use it.

All patients need to do is simply write down the medicine they

need to take, how many times a day they need to take it, and
 the time. Florence then automatically sends a message to its
users each time they need to take it.
Patients can interact with Florence inside popular messaging
platforms, like Facebook Messenger, Skype, and Kik. The app
can even send relevant information about different diseases
and locate a doctor or pharmacy near their location.

● Appointment Scheduling

Scheduling an appointment over the phone can be extremely

time-consuming for all parties involved. A patient would need to
phone their local clinic and the operator would need to check
their schedule to see what slots are available. All of this
back-and-forth can lead to higher administrative costs.
Some chatbots, like Hello Tars, enable healthcare facilities to
help their patients book appointments with the right doctor.

Patients can launch the scheduling bot from a website or even

access it from their mobile device while on the go. Some
chatbots can even send reminders to patients to let them know
that they have an appointment as well as what specific actions
they need to take afterwards.
Chatbots have practical applications for businesses as well.
They can be integrated with automation tools to generate leads
and follow up with them.

● Massive Cost Savings

Chatbots, as we’ve seen from the examples above, are

well-suited to handle many healthcare tasks from providing
informational content to streamlining scheduling — all of which
translates to huge cost savings.
A study from Juniper Research found that chatbots can help
 save businesses in the retail, eCommerce, banking, and
healthcare industries an estimated $8 billion a year by 2022.
Businesses can save on having to hire additional customer
support representatives and providing the necessary training.

● Patient Feedback

The goal for any healthcare provider is to provide exceptional

services to their patients. Chatbots are a great way to collect
feedback. Healthcare providers can use that feedback to make
their practice even better.
Sending out customer surveys may work. But patients may not
always have the time to fill them out or they might simply not be
willing to.
Joy Organics uses chatbots in its business to measure
customer satisfaction. They created a chatbot to collect
feedback. Customers can chat with the bot and rate their
overall shopping experience.

This kind of insight has helped Joy Organics improve their

customer services. Likewise, healthcare providers can use
chatbots in a similar manner to gather feedback from their
patients and see where they can improve.
Businesses can also use chatbots to collect feedback. For
example, a company like Drink LMNT can send a quick
message to customers and ask them to evaluate a recent
purchase of electrolyte drinks:

Collecting customer feedback works because it shows you

value their opinions. Of course, it’s important to actively listen
to what customers are saying and make efforts to improve
operations.
● Automated Invoicing and Payments

The medical billing process allows healthcare providers to

receive compensation for the treatment they provide patients.
However, sending invoices and keeping track of payments can
be costly and time-consuming.
In addition to facilitating appointments, chatbots can also be
used to automate invoicing. For example, a chatbot can be
programmed to send an invoice and collect payment after a
patient meets with their doctor. This can help reduce
administrative costs:

● Improved Internal Communications

Chatbots aren’t limited to providing support for patients. They

can also be used to facilitate internal communications for
employees.
For example, employees that have questions about a particular
process can visit the company’s site and interact with a chatbot
to get help. Healthcare providers can even set up chatbots to
link to training videos and other relevant materials.
 EXPECTED FUTURE WORK

● Serve as 24/7 companions, monitor health status in real-time, and

automatically call for assistance in case of an emergency.
● Help manage chronic conditions, mental health issues, and
behavioral and psychological disorders.
● Proactively identify symptoms, crosscheck them against medical
history, suggest the next steps, and improve the treatment success
rate in cases where early diagnosis can play a critical role.
● Make self-care easier by acting as a virtual assistant and providing
timely medical advice.
 CONCLUSION
A Chatbot is a great tool for conversation. Here the application is
developed to provide quality of answers in a short period of time. It
removes the burden from the answer provider by directly delivering the
answer to the user using an expert system. The project is developed for
the user to save the user their time in consulting the doctors or experts for
the healthcare solution. Here we developed the application using the
N-gram, TF-IDF for extracting the keyword from the user query. Each
keyword is weighed down to obtain the proper answer for the query. The
Web- interface is developed for the users, to the input query. The
application is improved with the security and effectiveness upgrades by
ensuring user protection and characters and retrieving answers
consequently for the questions.
 REFERENCES
1. Agrawal, M., Cheng, J., Tran, C.: What’s up, doc? a medical diagnosis bot

2. Amato, F., Marrone, S., Moscato, V., Piantadosi, G., Picariello, A., Sansone, C.:

Chatbots meet ehealth: Automatizing healthcare. In: Workshop on Artificial Intel-

ligence with Application in Health, Bari, Ital, pp. 40–49 (2017)

3. Aswini, D.: Clinical medical knowledge extraction using crowdsourcing

techniques.

Int. Res. J. Eng. Technol. 6(2019)

4. Bali, M., Mohanty, S., Chatterjee, S., Sarma, M., Puravankara, R.: Diabot: A

predictive medical chatbot using ensemble learning

5. Bao, Q., Ni, L., Liu, J.: Hhh: an online medical chatbot system based on
knowledge

graph and hierarchical bi-directional attention. In: Proceedings of the Australasian

Computer Science Week Multiconference, pp. 1–10 (2020)