Distributed Deep Learning for Smart IoMT Challenges in the Healthcare Domain

Distributed Deep Learning for Smart IoMT Challenges in the Healthcare Domain

Agila Harshini Thangavel
DOI: 10.4018/978-1-6684-9804-0.ch004
OnDemand:
(Individual Chapters)
Available
$33.75
List Price: $37.50
10% Discount:-$3.75
TOTAL SAVINGS: $3.75

Abstract

The Internet of Medical Things (IoMT) collects and transfers healthcare data over the network using sensors, software applications, and Edge devices. A greater number of Healthcare devices are being manufactured and there are various challenges like Interoperability, Security, Scalability, and privacy. IoMT devices are used to monitor and deliver treatments to patients remotely. For IoMt devices to reach their full potential the challenges need to be addressed. Healthcare devices when compromised can harm patients by disrupting personal data.
Chapter Preview
Top

Introduction

Digital healthcare products like fitness trackers, heart monitors, etc., which record real-time and biometric data are the clearest area of excitement now and in the future. Sustainable Digital healthcare shall transform society digitally by providing a faster response for emergency, and acute care and also reducing the amount of energy which in turn will reduce the carbon footprint of the server farm. Latency and Cyber security are the most dangerous challenges in IoMT devices. The classic security mechanisms in IoMT are provided by encryption, multifactor authentication, etc., which require more energy because of the longer bit length and are stored in a device that could be compromised. A lightweight technique that is both energy efficient and does not require the storing of encryption keys is ideally suited for a highly secure and energy efficient IoMT.

Figure 1.

Outline of latency from healthcare devices to server

978-1-6684-9804-0.ch004.f01
Top

Advancement In Healthcare Computing

This segment examines the fundamentals of computing in the healthcare domain as well as the catalyst which took computing towards a distributed architecture from a centralized cloud, which paved the way for the foundation for edge computing and fog computing. The specific point of reference for pioneering digital healthcare regarding price, use of less energy, and quality of experience are also examined.

Healthcare Application Types

Healthcare applications can be classified into various types by the type of device, data, or unique use case. The following are the primary healthcare classifications based on use cases:

  • Monitoring health in Real-Time

  • Emergency Management Systems

  • Information Dissemination in Healthcare

  • Health-aware Edge Devices

Multiple platforms can be used for real-time health monitoring at the same time. Observation of the body’s most basic functions like heartbeat, blood pressure, and temperature can be processed on Edge devices, and wearable sensors or can be combined and used. When a patient’s health falls below a specific value a real-time monitoring system will activate an alarm as patients these days have diagnostic tools in their hands because of the technological advancements in modern mobile gadgets. Many websites give information about healthcare, employing mobile devices, and customized applications to provide additional health information and assistance to individuals, particularly those suffering from certain chronic conditions.

Examples of modern health care for people are Wearable Sensors like smart watches, Mobile phone-based Sensors like heart rate sensors, and Ambient Sensors for example pedometers. Abnormal heart rates, blood pressure, glucose levels, and body temperature can be identified by Wearable sensors faster than finger prick glucose meters which are older technology. Sensor data is sent across long distances to a server from an edge computing application. For digital health-related purposes, built-in sensors like microphones, gyroscopes, and temperature sensors in smartphones can be used. Ambient sensors are placed around a room or series of rooms to gather data on the user's position without the patient wearing them. They provide better convenience, and this configuration is widely utilized in applications involving fall detection or, in dementia instances, for geriatric location tracking. Ambient sensors can be either interior or outside location sensors or both in certain specialist sensors.

Complete Chapter List

Search this Book:
Reset