DOCTOR-PATIENT REAL TIME CHAT APPLICATION

University of Texas at Arlington University of Texas at Arlington

MavMatrix MavMatrix

2022 Spring Honors Capstone Projects Honors College

5-1-2022

DOCTOR-PATIENT REAL TIME CHAT APPLICATION DOCTOR-PATIENT REAL TIME CHAT APPLICATION

Sri Subhash Pathuri

Follow this and additional works at: https://mavmatrix.uta.edu/honors_spring2022

Recommended Citation Recommended Citation

Pathuri, Sri Subhash, "DOCTOR-PATIENT REAL TIME CHAT APPLICATION" (2022).

2022 Spring Honors

Capstone Projects

. 34.

https://mavmatrix.uta.edu/honors_spring2022/34

This Honors Thesis is brought to you for free and open access by the Honors College at MavMatrix. It has been

accepted for inclusion in 2022 Spring Honors Capstone Projects by an authorized administrator of MavMatrix. For

more information, please contact [email protected], [email protected], [email protected].

DOCTOR-PATIENT REALTIME

CHAT APPLICATION

SRI SUBHASH PATHURI

Presented to the Faculty of the Honors College of

The University of Texas at Arlington in Partial Fulfillment

of the Requirements for the Degree of

HONORS BACHELOR OF SCIENCE IN COMPUTER SCIENCE

THE UNIVERSITY OF TEXAS AT ARLINGTON

May 2023

iii

ACKNOWLEDGMENTS

I convey my heartful gratitude to Dr. Shawn Gieser for revising my initial topic and

coming up with a concrete plan for this project. In addition, I am thankful to him for

effectively guiding my team and me patiently throughout the senior design project period.

With all his support, I was able to brainstorm effectively and develop a dynamic feature

that impacted my senior design project to a new level.

The project could not have been completed without the support of my teammates,

Rithik Kapoor and Dhruva Malik. We went through mainly all-nighters and conflicts and

sacrificed other aspects to transform the initial pitch into reality. Even though we could not

meet regularly for a couple of months due to the pandemic, we were able to keep with the

expectations and have detailed documentation with the help of GitHub for tracking our

progress and being compatible with a plethora of android devices.

May 1, 2022

ABSTRACT

DOCTOR-PATIENT REALTIME

CHAT APPLICATION

Sri Subhash Pathuri, B. S. Computer Science

The University of Texas at Arlington, 2022

Faculty Mentor: Shawn Gieser

During the pandemic, the majority of the working population spent most of their

time doing sedentary desk jobs. As a result, they may have developed inadequate spine

posture, which can cause profound implications such as back pain if not corrected at an

early stage. Implementation of machine learning and computer vision has allowed us to

capture and analyze video input from the user’s cellphone to remind them whenever they

are not in the correct position. To further support users, a chat feature has been

implemented through which the users can quickly get in touch with orthopedic experts for

evaluation and feedback. With the help of strategies from kinesiology, a comprehensive

analysis has been conducted to find the root cause of abnormality in the spine. As a result,

anyone can instantly monitor their posture and make minute corrections to recover swiftly

after consultation or a medical operation.

TABLE OF CONTENTS

ACKNOWLEDGMENTS ......................................................................................... iii

ABSTRACT ............................................................................................................... iv

LIST OF ILLUSTRATIONS ..................................................................................... vii

Chapter

1. INTRODUCTION ......................................................................................... 1

1.1 Background .............................................................................................. 1

1.1.1 Honors Project Responsibilities ...................................................... 2

1.1.2 Honors Project Tools ...................................................................... 2

1.2 Value Proposition..................................................................................... 3

2. TECH STACK ............................................................................................... 4

2.1 Android Studio & Java............................................................................. 4

2.2 Firebase .................................................................................................... 5

3. METHODOLOGY ........................................................................................ 7

3.1 Authentication .......................................................................................... 7

3.2 Real Time Database ................................................................................. 9

3.2 Chat Feature Implementation ................................................................... 10

4. CONCLUSION .............................................................................................. 13

4.1 Android Studio Conclusion...................................................................... 13

4.2 Firebase Conclusion ................................................................................. 13

Appendix

A. ANDROID STUDIO CODE MANAGEMENT PRACTICES ..................... 15

REFERENCES .......................................................................................................... 17

BIOGRAPHICAL INFORMATION ......................................................................... 18

vii

LIST OF ILLUSTRATIONS

Figure Page

2.1 Firebases removes the need for special servers ............................................. 6

3.1 Login Screen .................................................................................................. 8

3.2 SignUp Screen ............................................................................................... 9

3.3 Functionality of the real time database .......................................................... 10

3.4 Chat feature where a patient can ask about any problem ............................... 11

3.5 A snapshot of how the user list activity is designed ...................................... 12

4.1 Users would populate this way in Firebase when they sign up ..................... 14

CHAPTER 1

INTRODUCTION

1.1 Background

Throughout the pandemic, most of the working population started working from

home remotely. After understanding multiple surveys, it has been proven that people are

more alert and have better composure in the first half of the day, and as the day passes,

these levels begin to drop until they stop working. Many people spent an increasing amount

of time doing desk jobs and have a poor seated posture, which could lead to many health

issues like back pain, headaches, and spine curvature. Bad posture leads to not having

confidence, which further leads to low self-esteem. This decreases the productivity of the

employee, which is a great loss for the organization.

We aim to develop a unique application that will support people in improving their

posture by using the cellphone’s camera to monitor the user while they are seated and notify

them whenever they get into a bad posture. Although there are various other posture

correction applications on the Google play store or iOS application store, almost all of them

involve some expensive tracker placed on the body and do not seem to use the vast

capabilities of machine learning and computer vision. The spine is an essential part of the

body. Without it, we cannot keep ourselves upright or stand up. It provides us the body

structure and support. With the aid of the spinal cord help, we are able to move freely, and

it is designed to protect the central nervous system. Without that, we cannot properly move;

hence keeping the spine healthy is vital to living a happy life. This is the main idea behind

why we wanted to make a profound impact on everyone’s lives by developing a product

that is immensely beneficial.

1.1.1 Honors Project Responsibilities

As a part of the capstone project, additional components that are essential to

provide the users with the best overall experience were worked on. Worked on connecting

the doctors available in the local network to users seeking advice on specific therapy

practices or post-surgery cool-down procedures through chat. Adding medical support with

the help of online physicians can boost the morale of users and serve as an added incentive

to use the platform for maintaining a healthy lifestyle. The general tips provided on the

internet may not be suitable for all of us. Hence, its essential to provide users with the

ability to connect with external resources such as doctors to guide them with respect to

their specific health conditions. The real benefit of implementing this feature is that in our

application, doctors can readily view the patient’s history and view all the analyzed metrics

swiftly.

1.1.2 Honors Project Tools

This Honors project utilizes the same tools as the rest of the application. Java was

to develop the application in android studio. Firebase is used as the database for the

backend. Google’s ML Kit Vision Application was used, and it is the backbone of our

application. In Java, I used to recycler view for instantly populating the messages in the

conversation in real-time based on the internet traffic and flow of data through packets. All

the messages would be stored as unique components with their identification tagged with

the user in the cloud platform. Accessing real-time data was efficient by incorporating the

Firebase SDK. Once all the instances of firebase applications were set up, reading and

updating the database was a seamless process.

1.2 Value Proposition

There are various telehealth services available. However, the benefit of

incorporating that feature into our application is that physicians can readily view the patient

history and view the analyzed metrics of his back posture from any period of time, based

on the patient’s consent. This vastly reduces the amount of documentation and evidence

needed for physicians to decide on the treatment right away and solve it in the early stages

itself. Several times in society, the presence of spine and posture problems in younger

people might have incorrect interpretations from their friends and family circle as to not

exercising properly. With the help of this feature, the users can share all their feelings

confidentially. At-home, post-operation therapy has not only proven to be clinically

efficacious, but patients feel more comfortable communicating through the computer

screen and smartphone than in person. Another important aspect is that it is crucial for the

patient to feel at ease in their environment while performing the necessary exercises. And

for that, there is nothing more comfortable than being in the place where you want to be.

Even the expenses for gasoline are reduced with stress from traffic, making it a viable

option to pursue.

CHAPTER 2

TECH STACK

2.1 Android Studio & Java

Although many excellent tools are available for mobile application development,

Android Studio and Java stood out for several reasons. Android Studio was used to take

advantage of certain performance advantages immediately. Java is considered one of the

best programming languages for mobile applications. It is an object-oriented language

developed by Sun Microsystem in 1995. In all other coding systems, the code is first

translated by a compiler into instructions. In contrast, Java, on the other hand, turns code

into bytecode, which is then interpreted by Java runtime environment, and is more efficient

than the former. This is an easy-to-learn language with fluent English-like syntax and less

complex jargon. It has a rich API, XML (Extensible markup language) parsing, database

connection, networking, and utilities and provides almost everything that a developer can

expect. This is open source and is also freely available.

Android has succeeded in keeping Java at the forefront in the last couple of years.

For Android applications, you can use android APIs as well as Java to write code. These

applications are called native apps, developed by utilizing native tools. Android Studio is

the official integrated development environment for Android application development. It

is based on IntelliJ IDEA, a Java integrated development for software, and incorporates its

code editing and developer tools. This is much easier compared to cross-platform

frameworks such as React and Angular, where state management becomes extremely

complex. IntelliJ’s company initially developed Android Studio, it offers a diverse set of

features that enhance productivity when building apps, such as the flexible Gradle-based

build system, that is a fast and feature-rich emulator and a unified environment where you

can develop for all varieties of android devices. It has built-in support for the Google Cloud

platform, making it easy to integrate Firebase and Google Cloud Messaging and App

Engine.

2.2 Firebase

Firebase is one of the optimal services to utilizes in the backend domain due to its

seamless integration and user-friendly interface. This platform provides services at

multiple tiers to support all types of customers. First of all, it has its own proprietary

backend platform to securely store all the user data, and secondly it has multiple services

such authentication to effectively synchronize the backend platform and the application.

This platform supports everything from databases and analytics to crashing reports;

therefore, it is easy to debug applications multiple times to solve all errors. Throughout this

app, Firebase was incorporated into this application, it is being used for secure

authentication, storing chat in a real-time database and storing images in the Cloud Storage

part. The best part is that one can instantly implement all the features in real time by

connecting all data points together with a single unique UserID for each separate entry in

the database.

Figure 2.1: Firebase removes the need for special servers

CHAPTER 3

METHODOLOGY

In addition to utilizing the machine learning models for posture detection, deep

learning and neural networks were used to formulate the various back problems with the

help of Kinesiology and other practices. A variety of situations were researched were where

issues can occur and collect information also to provide the users during the consultation

so that doctors can provide treatments by looking to the resources present from our side,

without any further research and mainly using their profound experience.

Once all the information is collected, a comprehensive analysis was performed and

came up with a well-defined flow as to which problem can lead to other serious issues if

not taken care of and the respective side effects for each circumstance. In this way, we

initially plan to learn and become well-versed with the basics of the various medical

practices so that we can truly understand and build systems suited to those needs.

3.1 Authentication

The authentication part of the application consists of 3 parts, which is the SignIn

page, SignUp page, and the Forget password page. All these pages contain the basic

instances of all the design components, such as buttons from the aesthetic point of view.

The functionality in Android studio works through the way of listening, each component

waits to listen to the user interaction, and we created a step-by-step process informing the

application what to do based on the type of interaction with the user. Hence, once the user

types in all credentials, the application will observe their interaction.

When the submit button is pressed, the Listener function for that button will be

activated. Once the button is activated, it will send a snapshot of the entered credentials

and verify it with already stored details in the cloud and confirms if the user is legitimate

or not. The passwords are not stored, as it is in the application. Every time a new user

creates a password, the application uses an internally modified version of script to hash

account passwords. Even when an account is uploaded with a password using a different

algorithm, Firebase Auth will rehash the password the first time that the account logs

successfully in. Accounts downloaded from Firebase Authentication will only ever contain

a password hash if one for this version of the script is available or contain an empty

password hash otherwise. They have mainly incorporated this to make sure the personal

data of each user is in safe hands. In addition, they have created a progress bar that displays

the loading symbol when the authentication process is going on. This process gets activated

when a user clicks the submit button, and they have programmed it so that it only stops

showing once the entire procedure with Firebase is complete as a way to be transparent

with the user.

Figure: 3.1: Login Screen

Figure: 3.2: SignUp Screen

3.2 Real Time Database

Once the authentication is complete, Firebase stores a copy of the credentials in the

auth section of the platform, however, the central database where all the user and chat data

is stored is the real-time database. This database was chosen because in any chat

application, there is a need for bi-directional transfer of data. Firebase Realtime database

allows applications to approach cross-platform data within real-time after joining NoSQL

cloud-storage. Even piece is data is still being cached in the memory of a device when

offline and start synchronizing after getting into a local café that has a Wireless internet

connection. They can set customized data permissions at different levels so that all the data

points are sealed, to prevent any malicious attacks online. The data in the cloud is arranged

in a JavaScript Object notation (JSON) fashion. The system goes through multiple layers

and obtain the required data by chaining the commands as shown below:

Figure 3.3: Functionality of the real time database

3.3 Chat Feature Implementation

It took an entire week to create a logical flow as to create an initial design of the

software architecture. Once, the flowchart was ready, a development mechanism was

chosen where any two users can stay online in the application at the same time

concurrently. In addition, a couple of design components were created of each populated

message icon and box so that each aligns to the respective edge based on each user’s

perspective. For example, in the first user’s screen, their messages should appear on the

right side of the application and the second user’s messages should appear on the left side

of the screen to differentiate.

Figure 3.4: Chat feature where a patient can ask about any problem

This application used a recycler view in the application, as it is a flexible layout

whenever a user opens the application, all the previous messages need to be re-populated

every time. And this application has a loop such that the text extracted from the cloud will

be enclosed in the message box components, and will be packaged together, to be displayed

sorted by time from the bottom in the descending order. A special library called Picasso is

utilized, to store the images and for downloading and caching library for android. Further,

to update the database each time, a Database Reference instance was created to generate

the most updated snapshot on demand to analyze where to update in the database based on

the local timestamp of the device. To facilitate, the process of each message being properly

arranged, a User class that consists of all the parameters, so that during each cycle, all

essential details such as name, email, age, and all messages are updated together every

time.

To decide which person to chat with, an addition screen was developed which was

called the UserListActivity. This page consists of a Linear Layout which displays all user

profiles and their images as a list of User Cards. In this way, we are able to understand

which all users are online and offline. These cards are designed to appear dynamically if

the number of users does not fit the screen at once, the user will be able to scroll infinitely.

Figure 3.5: A snapshot of how the user list activity is designed

CHAPTER 4

CONCLUSION

This project allowed me to realize my great potential which was untapped till now.

Honors College faculty played a big role in encouraging me throughout the process.

Android Studio and Firebase are both incredible tools that made this honors project

manageable. The members of our team were quite familiar with the workload required to

achieve all the required goals. We were surprised by our efficient co-ordination. Although,

there was an initial learning curve with syntax and the logic of these tools, this quickly

outweighed its benefits.

4.1 Android Studio Conclusion

The mechanisms of Android Studio are slightly different than the plain Java

programming language. Each visible page of the application is divided into the two

components, which are the functional and design components. This gave us additional

flexibility to create custom connections between components to make the best user

interface for the customers. The Picasso library was useful in easily displaying profile

photos for each message in the chat screen for easy differentiation.

4.2 Firebase Conclusion

Firebase developed by Google is a marvelous product, which was extremely useful for our

team. With other backend platforms, the developer needs to setup the location of the servers

and their maintenance costs, and also mange load balancers. These all are taken care of

automatically in Firebase. Hence, we were able to merge the Firebase Database,

Authentication, and Storage platforms to facilitate in organizing all the messages based on

each pair of users. Firebase authentication empowered our team to develop a unique

project that was not only functional but was also secure and safe. Based on experience,

everyone should kickstart their Cloud journey by taking full advantage of Firebase.

Figure 4.1: Users would populate this way in Firebase when they sign up

APPENDIX A

ANDROID STUDIO CODE MANAGEMENT PRACTICES

1. XML layouts and AppActivity Screens: Creating of mobile applications involves two

steps. The first part is the creation of screen filled with basic components with no

functionality present, such as adding scroll bars, buttons, text views, image views, etc.

The second part, which is the App Activity screens involve the Java code that dictates

all the behavior of the activity screens based on the haptic feedback of the user.

2. OnClickListeners and User Adapter: All applications function based on our touch

response on the screen with respect to the needs on demand. To facilitate the instant

response when the user clicks anything, each widget in Android Studio has a system

similar to a timer, which keeps on listening (waiting) for a specific way of touch. Once

the user touches the button, and if his gesture suits the requirements, then that specific

widget will direct you to the next task. In order to display a list of users, a card-based

system was created whereby all data from Firebase gets downloaded and packaged in

the form of independently combined widgets, which can be stacked side by side in real

time. This flexible card-based system is known as the User Adapter.

REFERENCES

Kim, DeokJu et al. “Effect of an exercise program for posture correction on

musculoskeletal pain.” Journal of physical therapy science vol. 27,6 (2015): 1791-4.

doi:10.1589/jpts.27.1791

Jonaitis, Jenna. “12 Benefits of Good Posture - and How to Maintain It.” Healthline,

Healthline Media, 14 Apr. 2020, https://www.healthline.com/health/fitness-

exercise/posture-benefits.

“Why Posture Matters.” Harvard Health, Harvard Medical School, 24 Jan. 2017,

https://www.health.harvard.edu/staying-healthy/why-good-posture-matters.

“Why Android Developers Should Prefer Android Studio over Eclipse?” Digital

Transformation Agency, https://www.brainvire.com/android-developers-prefer-

android-studio-eclipse/.

BIOGRAPHICAL INFORMATION

Sri Subhash Pathuri is an undergraduate student studying Computer Science at the

University of Texas at Arlington. He had participated in two internships throughout his

college career; one at Animal Cloud, which was a startup company, and the second one at

Humana, as a software reliability intern. These internships enabled Sri Subhash to gain

valuable experience and knowledge as how to truly contribute to a company in a healthy

manner. Sri Subhash has also worked as an undergraduate research assistant at the

Innovative Data Intelligence Research Laboratory. Working on multiple projects in the

fields of machine learning, block chain, and computer vision, were useful in improving the

skillsets and helped in winning the HackTX (UT Austin hackathon) in 2021. Sri Subhash

plans to work for a couple of years to explore all areas and find the optimal field of interest.

Further, Sri Subhash would return to obtain a master’s degree in that field and try to

advance further and contribute significantly.