SeizSmart A mobile application for detecting, tracking, and reporting seizures in real-time. Feasibility Presentation Version 2.0 CS 410 Spring 2019 Team.

Slides:



Advertisements
Similar presentations
Care Devices for Epilepsy NYU Epilepsy Center NYU school of Medicine NYU.
Advertisements

Security Management IACT 918 July 2004 Gene Awyzio SITACS University of Wollongong.
A Smart Sensor to Detect the Falls of the Elderly.
Android ADK based app for offer making based on location and activity reporting John Waters.
PERSONAL MEDICAL MONITOR Jason Ewing, Morgan Hinchcliffe, Dina Irgebayeva and Aida Kulmambetova.
Home Health Care and Assisted Living Professor John A. Stankovic Department of Computer Science University of Virginia.
CONTINUES COGNITIVE AND VITAL SIGNS MONITOR Design Review.
Service Transition & Planning Service Validation & Testing
Event Management & ITIL V3
WEEK TWO, Session 2 Information Gathering. Helpdesk metrics must be reprioritized from measuring internal efficiencies to evaluating customer retention.
JUNIOR RASHID SHAFEER CS 7 – 7509 YCET CONTENTS o EVOLUTION o OVERVIEW o PROMISES AND POTENTIAL o REALITY o ANALYSIS o CONCLUSION o REFERENCES.
1. Test Information Distribution Engine (TIDE) 2.
ENGINEERING INNOVATION AWARDS NATIONAL ENGINEERING INNOVATION PRIZE (NIEP) CARDIOCAST.
HPHC - PERFORMANCE TESTING Dec 15, 2015 Natarajan Mahalingam.
Network and Server Basics. Learning Objectives After viewing this presentation, you will be able to: Understand the benefits of a client/server network.
Rationale M easurement of wrist acceleration (ACM) by means of wearable devices has been exploited to automatically detect ongoing motor seizures in patients.
IntroOH-1 CSE 5810 Remote Health Care Monitoring by Wearable Sensors and Mobile Devices Kanchan Jha Computer Science & Engineering Department The University.
Windows App Development Company India
Personal Home Healthcare System for the Cardiac Patient of Smart City Using Fuzzy Logic Shijia Liu.
KasPer Pro HRMS with Self Service Brings a Fully Featured Human Resources Management Solution to the Office 365/SharePoint Online Environment OFFICE 365.
People Inc. from P&A Software
Systems Analysis and Design in a Changing World, Fifth Edition
SharePoint 101 – An Overview of SharePoint 2010, 2013 and Office 365
Vital Signs Monitoring system
Sensors Journal, IEEE, Issue Date: May 2013,
IOT – Firefighting Example
Cost Effective Mobile Base Health Monitoring System Under Cloud Environment. To interpret health from their mobiles under cloud environment and creating.
EI Architecture Overview/Current Assessment/Technical Architecture
Team: Michael Alexander, Samantha Bergh, Mollie Lange and Peter Ma
Getting Started: Heart Safe School Accreditation.
Intelligent IVI with AI
Cisco Data Virtualization
Outline Introduction Standards Project General Idea
F Onorati1, G Regalia1, C Caborni1, R Picard1,2. 1
SmartHOTEL Planner Add-In for Outlook: Office 365 Integration Enhances Room Planning, Booking, and Guest Management for Small Hotels and B&Bs OFFICE 365.
Boomerang Adds Smart Calendar Assistant and Reminders to Office 365 That Increase Productivity and Simplify Meeting Scheduling OFFICE 365 APP BUILDER.
Uniting Office 365 and PRINCE2, UPrince and Project Online Make Managing Structured Projects More Efficient Without Increased Overhead Costs OFFICE 365.
Webparts360: A Low-Code App Development Tool That Enables Non-Programmers to Build Business Solutions for Microsoft Office 365 Quickly, Easily OFFICE 365.
Letsignit, an Automated Signature Solution for Microsoft Office 365 and Microsoft Exchange, Provides Efficiency in Branding and Customization OFFICE.
in All Office 365 Apps for Enterprise Companies
IWRITER 365 Offers Seamless, Easy-to-Use Solution for Using, Designing, Managing, and Sharing All Your Company Templates in Microsoft Office 365 OFFICE.
Design and Implementation
Smart Org Charts in Microsoft Office 365: Securely Create, Collaborate, Edit, and Share Org Charts in PowerPoint and Online with OrgWeaver Software OFFICE.
Balanced Scorecard Designer
Top Features of Smart Watch That Can Save Your Life.
9/17/2018 Future TIME Kevin R Banning, R&D Director Sept 17, 2014.
How it works: Step 1 99% of CAD systems can do this without modification is sent to a unique address for your agency Example:
Software Project Planning &
SocialBoards Self-Service, Multichannel Support Ticket Notifications in Microsoft Office 365 Groups Help Customer Care Teams to Provide Better Care OFFICE.
It’s About Time – ScheduleMe Outlook Add-In for Office 365 Enables Users to Schedule Meetings Easily with People Outside of Your Organization Partner Logo.
Epilepsy in Diagnostic Imaging
+Vonus: An Intuitive, Cloud-Based Point-of-Sale Solution That’s Powered by Microsoft Office 365 with Tools to Increase Sales Using Social Media OFFICE.
Internet of Things (IoT)
HL7 International January Working Group Meeting Health Care Device WG
Model-View-Controller Patterns and Frameworks
Microsoft Windows 10 Provides Power Behind Mobile Health Monitoring Solution for Patients “Diabetics or heart patients receive a new level of health surveillance.
People Inc. from P&A Software
ASSESSMENT IN COUNSELLING PREPAIRED BY: DR.MUNA ABDEEN ABDELRAHMAN.
eCareTaker: Context Aware Web Services
Chapter 2 Marketing Plan. Chapter 2 Marketing Plan.
COSO I COSO II. Meycor COSO, a Comprehensive Solution for Enterprise Risk Management (ERM)
Letsignit, an Automated Signature Solution for Microsoft Office 365 and Microsoft Exchange, Provides Efficiency in Branding and Customization OFFICE.
Reportin Integrates with Microsoft Office 365 to Provide an End-to-End Platform for Financial Teams That Simplifies Report Creation and Management OFFICE.
Smart Go Training Presentation 4/4/2019 3:22 PM
CS 410 Professional Workforce Development I Spring 2019
How will the NHS Long Term Plan work in our community?
Everyone is talking about
Modern benefits administration and HR software, supported by us.
Epilepsy Awareness for Schools & Early Years Settings
Contract Management Software 100% Cloud-Based ContraxAware provides you with a deep set of easy to use contract management features.
Presentation transcript:

SeizSmart A mobile application for detecting, tracking, and reporting seizures in real-time. Feasibility Presentation Version 2.0 CS 410 Spring 2019 Team Silver Abel Weldergay, Kevin Sokol Alpha Din Gabisi, Jeffrey McAteer Danielle Luckraft, Peter Scheible Hi everyone, my name is Abel, and today my team and i will be presenting our feasibility assessment of a product we have proposed, that will be able to detect, track and report epileptic seizures in real time, using current technology. Let me tell you a little bit about epilepsy, epilepsy is a neurological disorder that occurs in a particular part of the brain, as you can see in the illustration, in which the main symptom is epileptic seizures throughout the individual's lifetime

Table of Contents Team …………………………………………………....…….3 Background ………………………………………………..4-5 Problem ……………………………………………………..6-10 Solution ……………………………………………....…….11-15 Competition .……………………………………….…….16 Customers ………………………………………………...17 Conclusion ……………………………………………......18-20 References ……………………………………………….. 22-27 Here is what we are going to talk to you about, We will first discuss the problem, and the characteristics associated with the problem, then we will discuss our proposed solution, and how we realize that solution, both in hardware and software Then we will discuss the competing systems, both direct and indirect, Then we will close out with the customer base, which includes not only people directly affected by epilepsy, but also their loved ones, friends, and the medical community that can benefit as well

The Team Here is our team, Abel Weldaregay Team Lead / Back-End Developer Kevin Sokol UI/UX Developer - Smartphone Peter Scheible UI/UX Developer - Smartwatch Danielle Luckraft Webmaster / Developer Jeffrey McAteer Infrastructure & ML Engineer Alpha Din Gabisi Database Engineer Here is our team, I will be discussing the background and the problem, and i will be graduating in the spring of 2020 And we have jeffery, who will be graduating this december, and will be presenting the solution And we have alpha, who will be graduating in the fall of 2020, and will be presenting our competition and customers And we have danielle, who is a senior and is our webmaster, And we have kevin and peter who are both seniors and have been doing research and developing for us

Background - Epilepsy Now let me tell you some facts about epilepsy, Epilepsy is the 4th most common neurological disease in the world[12]. Cases of epilepsy in the US have increased over the past five years[18]. Cases in the US are predicted to increase further by 2020[18]. Now let me tell you some facts about epilepsy, Epilepsy has affected more than 50 million people worldwide, making it one of the most common neurological disease in the world, People with epilepsy are 3-6 times greater at risk of premature death And 80 % live in low or middle income countries And 75% do not get treatment Cases in epilepsy have also consistently increased in the US from 2014-2019, and are predicted to increase in 2020 The ability to detect and track epileptic seizures is essential to minimizing bodily injuries or death

Characteristics of Generalized Seizures Rapid change in heart rate[2] Rapid convulsions in limbs and face[2] Loss of consciousness[2] Here are the characteristics of generalized seizures There is an increase in heart rate, repetitive body movements, and loss of consciousness Now let me tell you about the timeline of a generalized seizure

Problem Statement -They only measure one metric Epileptic seizures are difficult to detect in a timely and accurate fashion and undetected seizures can result in injury or even death. Current smartwatch detection technology does not provide an ability to automatically detect the onset of a seizure based on a combination of heart rate behavior and repetitive body movements. Available devices do not provide capabilities that tune detection variables to match individual patient seizure characteristics. Existing solutions to detect seizures use smartwatch technology which must be in the proximity of a smartphone in order to notify emergency contacts. -They only measure one metric -Don't provide the capability to make the detection process unique for each patient -require the patient to be in close proximity of the smartphone to send a notification to emergency contacts

Who is Affected -It can affect anyone from young children to adults Epilepsy affects those all ages; from young children to seniors[3]. Approximately 25% of individuals diagnosed with Epilepsy have generalized tonic-clonic seizures[3]. Epilepsy is more likely to affect those who: are autistic, have experienced a stroke, or have suffered a significant infection or head trauma[4]. -It can affect anyone from young children to adults -and about 25% of patients with epilepsy suffer from generalized seizures, which is the most common and most observable - it is more likely to affect people who Are autistic Have Experienced a stroke Or have suffered a significant infection or head trauma

Problem Characteristics Existing technology relies on the detection of an increase in heart rate or repetitive body movements (but not both) to confirm the onset of a seizure. Concurrent recognition of repetitive body movements and a rapid change in heart rate is essential for achieving high accuracy of results and low false positive risk of a seizure detecting device. Current solutions do not offer direct notification of emergency contacts from a wearable detection device. They instead rely on a “relay” (such as a smartphone) which must be in proximity of the wearable in order to notify emergency contacts. Available solutions capable of detecting, tracking, and reporting seizures require subscription services, prescriptions, or both. Current solutions rely on an increase in heart rate or repetitive body movements in order to detect the onset of a seizure, instead of a combination of both, which impacts the accuracy of the detection process restricts the smartwatch to be in close proximity of a smartphone to send a notification to emergency personnel, which impacts the mobility of the patient Available solutions that have the potential of tracking and recording seizures require subscription, prescription, or in some cases a combination of both.

Current Process Flow Most existing solutions detect seizures based on body motion. Some detect seizures based on users heart rate. The process flow for both are identical. No existing system detects based on a combination of both metrics.

Current Process Flow Wearables may access more data than HR/Motion. Not all patients respond to seizures in the same way. Current processes only begins recording seizure data after detection time. Simpler notification capability is needed.

Solution Statement How do we fix these problems? Our proposed solution, SeizSmart, implements an advanced, wearable seizure detection capability using off-the-shelf smartwatch technology that is able to: automatically detect epileptic seizures from combined heart rate and motion metrics, Detect seizures using an algorithm which matches individual patient seizure characteristics, track and record all information surrounding seizure events, and provide automatic notification to emergency contacts without requiring a relay. How do we fix these problems? We propose SeizSmart, software which runs on any off-the-shelf smartwatch. The system detects seizures using both HR and Motion, tunes detection to individual patients, records valuable metadata surrounding seizure events, and provides notifications without a relay device.

Solution Characteristics Smartwatch technology is used for detection, tracking, and recording of generalized seizures. Machine learning technology is used to evaluate heart rate and body motion characteristics to establish a seizure profile for each patient. Heart rate performance and body motion are continuously monitored. Both heart rate and body motion information is used to indicate a detection. Available data about the environment during the onset of a seizure is collected. Automatic notification to emergency contacts or first responders is available when appropriate. Off-the-shelf smartwatch hardware is used, ML takes in both HR and motion and creates a unique seizure profile for each patient This training data is recorded continuously and uploaded in batches for training to the user’s learning server. Notifications are tiered - first the patient and their EC, finally a 911 alert if neither respond within a time window. (Learning server may not be in cloud, see https://cloud.google.com/edge-tpu/)

Solution Process Flow Detection is based on a combination of heart rate and body motion characteristics. Detection performance is enhanced using a trained machine learning approach. Emergency notification is issued directly from the user’s smartwatch. Our solution process flow begins by measuring HR and motion, which is fed into the trained NN. If the biometrics match seizure characteristics, We alert the patient, their emergency contact, and queue a 911 call. If the patient responds, we clear the 911 call and tell their EC they are OK. If the EC responds, we clear the 911 call. If neither respond the 911 is dispatched directly from the smart watch.

Process Flow Comparison ML Detection Technique Tiered Notifications Fewer components Measures >1 Data Point Records all event data Let’s recap the differences in process flow: We collect more than 1 type of biometric reading, We use modern ML instead of a simple threshold, And our notifications alert both caregivers and first responders without requiring a relay.

Major Functional Component Diagram SeizSmart has three cycles of data flow: A passive biometric collection and batch upload to train a neural network, An active biometric testing against a known seizure profile using a trained neural network, And an on-demand configuration for emergency contacts and reporting preferences through a web browser (possibly your phone -> ) Notice how the smartphone is on the periphery and it not necessary, the passive collection and training cycle needs only be intermittent, and the critical flow happens entirely within the watch and needs only a cellular network to issue notifications.

Epilepsy Health Storylines Competition Matrix Direct Indirect SeizSmart SmartMonitor empatica embrace 2 SeizAlarm Epilepsy Journal Epilepsy Health Storylines Detect, record and track generalized seizures in real time ✔ ❌ Monitor repetitive shaking motion Continuously monitor the user's heart rate Only checks for elevated heart rate Alert emergency contact when the user does not respond Report data about the environment at the onset of a seizure being detected Function fully without dependence on a smartphone or external device Use machine learning to detect generalized seizures Require a subscription or prescription

Benefits to Customer Base Detection Performance and Hardware Flexibility Each user’s individual seizure profile provides more accurate and customized seizure detection. The user may configure emergency response notifications as desired. SeizSmart is compatible with both android and iOS smartwatch technology without the need for specialized hardware. SeizSmart will be available without a subscription and a prescription will not be required. Peace of Mind A smartphone does not need to be in close proximity to the smartwatch for detection and notification of emergency contacts. SeizSmart is capable of notifying emergency personnel in extreme situations. We offer two main benefits to our users Detection Performance and Hardware flexibility. Users have a seizure profile tailored to their lifestyle. So they can tweak emergency response notification as desired. IOS and Android smartwatches are compatible no need for specialized hardware. No prescription or subscription is required. Peace of mind is also offered to the user. So they do need to always be in proximity to their smartphone for notification of emergency contacts. Seizsmart can contact emergency contact as required

What SeizSmart Will Not Do It will not predict seizures in advance of known symptoms It detects all types of generalized seizures except for absence seizures It is not a medical application and is not intended to be used in the diagnosis, monitoring, prevention, or treatment of epileptic seizures. We do not predict seizures We can detect every type of generalized seizures except for absence seizures. These seizures are seizures in which an individual blanks out or stares into space We are not a medical application so there is no need for FDA approval

Key Points Summary SeizSmart is a mobile application based on smartwatch technology that is designed to improve the detection, tracking, and reporting of generalized seizures. The Problem Current applications only check for an increase in heart rate or rapid body movements. Current applications require a prescription or subscription plan in order to detect and track seizures. Current applications require the smartwatch to be in close proximity to the relay device to transmit alerts and notifications. The Solution Continuously monitor the end-user’s heart rate and body movements. Apply machine learning to the collected data about the end-user’s seizures to build a unique, personalized, more accurate seizure profile. Execute within the smartwatch itself to enable independent operation without requiring proximity to a relay device. As we all know our competitors check for an increase in heart rate or rapid body movement. Some of their devices require a prescription or a subscription to work and a smartphone must be next to the user We aim to provide maximum benefit to the user by monitoring both heart rates and body movement we use this data to build a unique profile for each individual user. We dont require the smartwatch to be in proximity to the user

Who Benefits/Why Important/Why Feasible Anyone who suffers from generalized seizures. Medical/research teams looking for data about epilepsy. Why important? Provides end-users with the ability to detect, track, and record seizures using a seizure profile uniquely crafted for them. Why feasible? Seizmart leverages advancements in existing smartwatch and machine learning technology to detect seizures in real time. The people who suffer from generalized seizure and the people who support them Our application provide users with ability to record and detect their seizure accurately using a seizure profile tailored to them We use existing technology to track and record seizures by storing collected data about the end-user into a database and providing a web interface for users to view the collected data

References - Epileptic Seizure Detection Tzallas, A. T., Tsipouras, M. G., Tsalikakis, D. G., Karvounis, E. C., Astrakas, L., Konitsiotis, S., & Tzaphlidou, M. (2012, February 29). Automated Epileptic Seizure Detection Methods: A Review Study. Retrieved from https://www.intechopen.com/books/epilepsy-histological-electroencephalographic-and-psychological-aspects/automated-epileptic-seizure-detection-methods-a-review-study Giannakakis, G., Sakkalis, V., Pediaditis, M., & Tsiknakis, M. (1970, January 01). Methods for Seizure Detection and Prediction: An Overview. Retrieved from https://link.springer.com/protocol/10.1007/7657_2014_68 Devices & Technology. (n.d.). Retrieved from https://www.dannydid.org/epilepsy-sudep/devices-technology/ February;25(2):28-29, N. R., Publish date: December 6, 2., & Publish date: December 18, 2. (2019, January 07). Mobile Devices May Provide Accurate Seizure Detection and Help Prevent SUDEP. Retrieved from https://www.mdedge.com/neurology/epilepsyresourcecenter/article/130162/epilepsy-seizures/mobile-devices-may-provide

References - Epileptic Seizure Detection Continued van Elmpt, Wouter J C, et al. “A Model of Heart Rate Changes to Detect Seizures in Severe Epilepsy.” Seizure, U.S. National Library of Medicine, Sept. 2006, www.ncbi.nlm.nih.gov/pubmed/16828317. Borujeny, Golshan Taheri, et al. “Detection of Epileptic Seizure Using Wireless Sensor Networks.” Journal of Medical Signals and Sensors, Medknow Publications & Media Pvt Ltd, 2013, www.ncbi.nlm.nih.gov/pmc/articles/PMC3788195/. Velez, Mariel, et al. “Tracking Generalized Tonic-Clonic Seizures with a Wrist Accelerometer Linked to an Online Database.” Seizure, U.S. National Library of Medicine, July 2016, www.ncbi.nlm.nih.gov/pubmed/27205871.

References - Heart rate and Epileptic Seizures Kołodziej, M., Majkowski, A., Rak, R. J., Świderski, B., & Rysz, A. (2017, September). System for automatic heart rate calculation in epileptic seizures. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/28523469 Nei, M. (2019). Cardiac Effects of Seizures. American Epilepsy Society. Zijlmans, Maeike, et al. “Heart Rate Changes and ECG Abnormalities during Epileptic Seizures: Prevalence and Definition of an Objective Clinical Sign.” Epilepsia, U.S. National Library of Medicine, Aug. 2002, www.ncbi.nlm.nih.gov/pubmed/12181003.

References - Epilepsy “Demystifying Epilepsy and Increasing Awareness.” Mayo Clinic, Mayo Foundation for Medical Education and Research, https://newsnetwork.mayoclinic.org/discussion/epilepsy-demystify-disease-and-increase-awareness/. “Epilepsy Foundation.” Epilepsy Foundation, 13 Mar. 2019, www.epilepsy.com/.

References - Direct Competitors “About SmartWatch Inspyre™ by Smart Monitor – Smart-Monitor.” Smart, smart-monitor.com/about-smartwatch-inspyre-by-smart-monitor/. “Embrace2 Seizure Monitoring | Smarter Epilepsy Management | Embrace Watch.” Empatica, www.empatica.com/embrace2/. “SeizAlarm Epilepsy Seizure Detection.” SeizAlarm Epilepsy Seizure Detection, seizalarm.com/.

References - Indirect Competitors “Epilepsy Journal App | OllyTree Applications.” Epilepsy Journal, www.epilepsy-journal.com/. “Health Storylines™.” Health Storylines™, www.healthstorylines.com/. Jahmunah. “An Encounter with Epilepsy...” A Bucket Full of Science, 12 Feb. 2017, abucketfullofscience.wordpress.com/2017/02/12/an-encounter-with-epilepsy/.