Assignment

Table of Contents List of Figures 3

Executive Summary 4

Problem Statement 5 Need 5 Objectives 6 Background and Technology Survey 7

Requirements 8 Marketing Requirements 8 Engineering Requirements 8 Constraints and Tradeoffs 10 Applicable Standards 10

Ethical and Societal Considerations 10

Project Management Plan 11 Schedule and Gantt Chart 11

Appendices 11 Appendix A: Resumes of Team Members 11

List of Figures Figure 1: Visual Field Defects Diagram Figure 2: Gantt Chart

Executive Summary

The goal of this project is to build and test a headset that helps doctors and

neuroscientists assess potential stroke victims remotely via a telemedicine platform. Our device

facilitates a basic peripheral vision test to assess the visual fields of a stroke patient. Visual field

loss is a symptom of a stroke in 10-15% of stroke patients (amounting to 200,000 cases per

year). Using the device, a doctor can remotely select from a series of figures or pictures to be

shown in any quadrant of the peripheral line of sight while receiving a real-time video output of

the patient's eyes to ensure the test is being done to the standards of practice. Based on the results

of the test, the doctor will be able to make a diagnosis with a near-100% success rate.

This technology is needed in the medical industry because there is no standardized visual

field evaluation for stroke victims. Each doctor applies a visual field test differently. This causes

irregularity and inaccuracy in diagnosing the stroke among victims. Not to mention, 40% of

these stroke cases occur in rural areas where there are few neurological specialists present to

make a proper diagnosis. Telemedicine has improved the rural medical community’s access to

subject-matter specialists and our device will integrate with existing telestroke platforms that are

already used in these settings. This will improve the success and ease that neuroscientists using

these platforms are able to achieve in diagnosing strokes.

To determine the need of this device, a customer interview was conducted with Dr.

Mouhammad Jumaa. Dr. Jumaa is an assistant professor in the department of Neurology at the

University of Toledo. His primary practice and area of interest is in the diagnosis and treatment

of stroke patients.

Problem Statement Need Stroke Detection Goggles will assist doctors in assessing potential stroke victims. Detecting a

stroke early can help save lives and improve a recovering victim’s quality of life by limiting the

long term effects of the stroke. Currently, there is a shortage of qualified personnel to detect

these early symptoms of strokes in all areas of the country, specifically in rural areas. To

improve the rural community’s access to medical specialists, telemedicine platforms have been

developed. These are basically tools, cameras, and monitors that allow a patient and a doctor to

interact remotely. This means that a patient in Los Angeles can be assessed by a doctor in New

York in real-time. The rural community has come to rely heavily on telemedicine for stroke

diagnosis resulting in approximately 40% of strokes being diagnosed via telemedicine at this

point in time. However, the methods currently used in the industry are not standardized and can

result in missed diagnosis and subsequently negative long-term effects for victims.

The focus of our device is on one symptom of stroke: visual field loss. This condition describes

the event where a portion of a victims peripheral vision is lost. The shape and degree of the

visual field loss indicates severity and locality of the stroke within a victim’s brain. Targeting the

area of the brain the stroke is in is ideal when administering treatment. Doctors use a series of

hand gestures within the visual fields of the patient and rely on their feedback to make

judgements about where the stroke is taking place. For instance, the figure on the following page

shows the variety of types of vision loss a patient could be facing corresponding to the region of

the brain a stroke is occurring. The current tests are very subjective and a step towards

standardization will help the community in successful diagnoses as well as rapid improvement of

the process. The goggles will fill these needs, it will integrate well with current telemedicine

platforms like Vydio and Intouch while offering these doctors a standardized method for visual

field testing and improve its ease and accuracy. There will no longer be a need for a specialist at

the bedside and the inconveniences that come from 2-D screens and cameras unable to detect

subtle eye movements that show a patient is not following the procedure they are supposed to.

Figure 1: ​ Visual Field Loss Based on Locality of Stroke

Source: https://frcemsuccess.com/visual-field-defects-2/

With our device, doctors will be able to perform a visual field test by flashing shapes or

numbers in a selected quadrant of the visual field. This action coupled with a patient's responses

will help determine if loss of vision has occurred. By testing these fields a doctor can more

effectively detect and diagnose a stroke.

Objectives The objective of Stroke Detection Goggles is to provide neurologists with a medical

device capable of performing visual field testing via a remote connection. It should allow the

doctor to be able to view the patient’s eyes using an infrared camera with minimal delay to

ensure the test is being conducted accurately. The following are a variety of quality and

performance metrics that need to be hit in order for the device to be considered a viable tool for

this application.

1. The goggles must assist the neurosurgeons so that a near 100% correct diagnosis rate is

achieved.

2. The goggles must follow all HIPAA standards and regulations relating to patient data.

3. The goggles-to-face interface should be adjustable to accommodate any user as well as be

comfortable enough to allow for a test length of approximately 5 minutes.

4. The device must easily interface with currently-used telemedicine platforms.

5. The doctor must have full control of functionality and receipt of outputs from a remote

connection.

Background and Technology Survey Currently there is no standardized way to test for an acute deficit in the visual fields in

patients. This is especially difficult in areas where no neurologist can be reached to conduct the

test in person. Our product would allow for doctors to accurately and repeatedly test for these

acute ocular deficits over any distance, provided they have a broadband connection. This

standardization is what sets our product apart from any current method. If it could become a

usable, reliable, and recognized way to conduct this test it could lead to a significant increase in

accurate stroke diagnosis and a decrease in the time to diagnosis, which is crucial for faster and

better treatment.

Our product uses both hardware and software technologies to construct a device capable

of performing a visual test controlled by a doctor remotely and displaying a video of the patient's

eyes for the doctor to view. The device should utilize a microcontroller and either 7- segment

displays or LCD screens to output figures for the patient to identify. The microcontroller is a

Raspberry Pi 2B, which is compatible with a video module that greatly simplifies video

integration. These connections will be housed in a headset akin to virtual reality goggles (i.e.

Oculus Rift, Sony Vive) and wirelessly connected to the host facility secure wireless network.

Through the local network, the headset will transmit video output to a secure server. The server

will provide an external portal for doctors to log into. Using a mobile or desktop application

doctors will be able to select the headset much as they would any other telemedicine device. The

server will connect the doctor to the required headset and rout the video feed and device

commands. All data from the headset will be transmitted point to point, and not be stored so as to

minimize any risk of data leakage.

Requirements Marketing Requirements

There are several marketing requirements for the stroke detection goggles. The first is to

allow a doctor to use an app or remote desktop to control lights or pictures in the headset. This

allows the doctor to choose what is displayed and where. It is important that this feature be

included in the device so that doctors have greater control over the test and better orient the test

to a patient's individual needs. The second requirement would be to create a headset that is

comfortable for the patient to wear for the duration of the test as well as using straps and padding

that can be disinfected in between uses. Creating and maintaining a device that can be cleaned to

hospital standards is important to eliminating the potential for disease to spread. Thirdly creating

a device and user interface that is intuitive and easy to use. A first impression can make or break

a product, that is why ease of use is so important. This leads into the fourth requirement which is

to integrate our product with existing telemedicine services such as InTouch or Vydio. InTouch

and Vydio are the two largest telemedicine companies used in the United States. Getting our

product into one of these platforms increases the incentive for companies to adopt our product as

a standard practice. Finally creating a product that is cost effective for health care facilities to use

and maintain. The device should be a durable and reliable unit with a reasonable expected life

span.

Engineering Requirements From a hardware perspective, the device must run off of a microcontroller (Raspberry Pi

2B or later) that has wifi connectivity as well as camera integration. The camera must be infrared

capable because we need to provide the doctor with a video feed of the eyes in a dark

environment. We must also have a minimum of 12, seven-segment LED displays placed in the

designated visual field locations. Each screen must be controlled individually, demanding a high

pin count or digital logic control scheme. All of the hardware must be powered by an integrated

medical grade Lithium-ion rechargeable battery with a minimum operating time of 1 hour per

charge. No external wires, other than the battery charging wire, are to be required for operation.

The “housing” or goggles themselves must be a custom design that encapsulates the

visual field spans adequately in a compact package. The housing must also allow for mounting of

the electronics. This must be designed in SolidWorks and 3-D printed. Injection molding may be

an option to bring costs down when manufacturing for the device would ramp up. The housing

must also have elastic banding that allows the nurse to fix the device on the patient’s head and

support itself while also maintain patient comfort with hypoallergenic plastic padding on the

contact points with the face.

The software requirements of the device are as follows. For ease of use and comfort of

the patient, the physical unit will be wireless save for the charging via USB. Doctors will be able

to use the device from any mobile or desktop device with an internet connection. The user

interface for the doctor will be easy to use, intuitive, and responsive. He or she will be able to see

the patient’s eyes at all times during the test, while easily manipulating the displays inside the

headset. Since time is of the essence for diagnosis, the UI must be responsive and resilient

against latency. Log in times for the doctor should be minimally slower than the headset can

allow. To ensure the privacy of patients, data must be handled in a specific manner. All avenues

of data transmission must be encrypted and not recoverable without the private encryption key of

the host server. Further, storage of data (specifically video) will be as minimal as an

uninterrupted test environment will allow. Should any data be “at rest” it will be encrypted with

a key specific to the test in progress and be securely purged at the test’s conclusion. The

telemedicine market is already largely cornered by companies such as Intouch. Rather than

attempt to compete with existing market leaders, we intend to make our product readily

integrable with the existing frameworks. By doing so, we open up the ability for the testing unit

to be deployed much more widely. Additionally, this course of action allows a revenue stream

through licensing of the product.

Constraints and Tradeoffs There a few constraints and tradeoffs that occur when designing this device. The first

being choosing a microcontroller. When considering a microcontroller one must look at all

functions that will be needed. We need a controller that is lightweight, has wifi, video, and GPIO

capability. It must also have enough processing power to stream and encrypt video and receive

inputs from a server while also being cost effective. Secondly is the choosing a material for the

headset. We want a headset that will be comfortable for the user, easy to clean and maintain, and

also low cost to build. The headset displays are a primary concern for cost and complexity of the

product. Custom LCDs (Liquid Crystal Displays) would provide the most flexibility and control

by the doctor operating the display. However, forming an LCD to the form of the headset visor

would require specially designed models from a manufacturer, causing a significant price

increase. Also, an LCD would significantly increase the required development time of the

command and control system. However, if a single LCD screen is used, it could reduce the

complexity of the headset wiring. Another constraint on the design of this device is patient

comfortability and ease of wearing the device. It’s important for patients to be comfortable while

wearing the device so as not to have an impact on the tests performed using the device.

Applicable Standards There are a few standards that need to be followed. Safety standards need to be followed

when designing and building our device due to the use of live circuitry. Due to the data collected

by the headset being of a medical nature, HIPAA standards for the transmission and storage of

any kind will be satisfied. Additionally, cleanliness standards need to be established and

considered when designing our device due to the reuseable nature of the device and the medical

setting in which it is to be used.

Ethical and Societal Considerations There are a few Ethical issues and considerations that need to be noted in this report. The

first being ensuring that a patient's data is secured and not recorded without proper consent and security. This is one of the major reasons HIPAA regulations need to be met. Without following these rules and regulations, not only would the medical community not buy this product, but the possibility of confidential information being leaked could be high.

Project Management Plan Schedule

The decision to pursue this venture was made in late March. Since then, we have constructed a basic proof of concept to test and tweak the visual field displays. We plan to continue the design and initial build process through May of 2019 in order to have a minimum viable product ready and functional. Thereafter will be a Mk2 build which will be more refined and include more long term features desired by the customer. By the beginning of the Fall 2019 semester, the project will be in the testing phase, wherein we will be looking for correlation between positive test results and stroke diagnosis. During this time, we will also be producing the final report due at the conclusion of the semester. In the last few weeks of the semester we will be working to ensure that all documentation is correct and the project is either ready for deployment or ready to be handed off to an incoming team that will continue the project as far as they can.

Gantt Chart

Appendices

Appendix A: Resumes of Team Members