AI-Based Internet of Things (AIoT): Applications of AI With IoT

Introduction

Artificial intelligence (AI) and the IoT are connecting physical items and devices by encouraging physical objects and devices to see, hear, and think. Physical objects or devices can “speak” to one another and communicate their decisions through the exchange of information. Products that were previously unintelligent are becoming intelligent as a result of technologies such as the IoTs, which connect previously unintelligent objects to the internet through a variety of embedded devices, communication protocols, sensor networks, internet protocols, and applications. There are various types of AI-based IoT, i.e., AIoT-enabled healthcare services used in the medical industry: electronic health and telecare networks; diagnostic, preventive, and other associated technologies. There are also plans to install rehabilitation and monitoring equipment. Wi-Fi body area networks and radio frequency recognition systems, for example, contribute to the system’s operation but are not required for its overall operation. While research in related fields has shown that remote health tracking is feasible, the potential benefits in a variety of situations may be even more significant to consider in light of the findings, which should be investigated further. When remote health surveillance is used to track non-critical patients at home rather than in the hospital, it can help to reduce the demand for hospital resources such as physicians and beds. It could be used to improve access to healthcare for remote communities or to allow elderly people to stay in their homes for longer periods of time. The prospect of increased access to healthcare services while reducing the strain on healthcare institutions is a positive development. This is critical given the global shortage of healthcare workers and emerging infectious diseases. Furthermore, it has the potential to empower individuals by giving them greater control over their own well-being at all times. As a result, populations are shifting to self-care options, which have the potential to make health care more sustainable by reducing reliance on government interventions throughout the healthcare ecosystem. Figure 1 depicts a graphical representation of current IoT-based healthcare equipment in use in the healthcare sector.

Figure 1.

IoT-based Healthcare Devices (Al-Dhief et al., 2020)

At the moment, healthcare systems are made up of a patchwork of different, always-on monitoring devices. These devices use single physiological waveform data or discretized vital information to send out alerts when major events happen (Cvach, 2012). But these simple ways to make and use alarm systems aren’t very reliable, and the sheer number of them may cause both caregivers and patients to get “alarm fatigue” (Drew et al., 2014; Patel et al., 2022). In this situation, the ability to find new medical knowledge is limited by the fact that past research has often failed to fully use high-dimensional time series data. Jphan et al., proposed in (Rothschild et al., 2005) that the reason these alarm mechanisms frequently fail is that they rely on isolated sources of information and lack context regarding the patients’ true physiological conditions from a broader and more comprehensive standpoint. As a result, more comprehensive and improved approaches to studying interactions and correlations between multimodal clinical time series data are required. This is important because research shows over and over that people can’t think about changes that affect more than two signals (Carayon, 2006; Carayon & Gürses, 2005).