Using Artificial Intelligence to Make Sense of IoT Data
There is a coherent overlap between the Internet of Things and Artificial Intelligence. IoT is basically an exchange of data or information in a connected or interconnected environment. AI is about simulating intelligent behavior in machines that carry out tasks ‘smartly’. AI tries to imitate natural human intelligence or the cognitive functions that humans perform using their mind such as learning and problem-solving.
As IoT devices generate large volumes of data, AI is functionally necessary to make sense of this data. Data is only useful when it is actionable for which it needs to be supplemented with context and creativity. IoT and AI together make this context, i.e. ‘connected intelligence’ from connected devices.
Traditional methods of analyzing structured data are not designed to efficiently process these large amounts of real-time data that is collected from IoT devices. This is where AI-based analysis and response play a critical role in extracting optimal value from the data.
Bringing the power of AI to IoT
When we say AI, we really mean machine learning because ML provides the ability to detect patterns in data presented. It learns from these patterns in data to adjust the ways in which it then analyzes that data or triggers actions.
With ML embedded in an IoT environment you get more ‘connected intelligence’. We are now seeing significant investment in the convergence of IoT and AI. Microsoft announced in May its vision for intelligent cloud / Intelligent Edge. Azure IoT Edge will enable low-power devices to run containers and perform AI locally but retain a connection to the cloud for management and modelling. Similarly, in April, Amazon Web Services (AWS) updated its edge computing platform, Greengrass, to incorporate ML.
An industry that is a great example of IoT & AI working hand in hand is healthcare. There are monitoring devices you can attach to monitor important parameters of human health such as cardiac monitoring, blood pressure monitoring, sugar monitoring, etc., and constantly report this data to backend. This is basically IoT, sensing the conditions to various devices attached and communicating with the backend using communication networks.
At the backend, based on the data collected, data is stored in data lakes. Such data is collected from hundreds, thousands and millions of users. Then AI/ML algorithms are run on this collected data. With the help of these algorithms, the system predicts the disease and suggests preventive measures automatically. It is also capable of alerting the doctors when a threshold is crossed and doctor’s intervention is necessary. This is AI/ML aspect of the work.
Evolution of Internet of Things
As IoT continues to be one of the most popular technology buzzwords of the year, the discussion has evolved from what it is, to how to drive value from it, to the tactical: how to make it work.
IoT produces a treasure trove of big data. This data can help cities predict accidents and crimes, give doctors real-time insights into information from pacemakers or biochips, enable optimized productivity across industries through predictive maintenance on equipment and machinery, create truly smart homes with connected appliances and provide critical communication between self-driving cars. The possibilities that IoT brings to the table are endless.
There is a growing need to improve the speed and accuracy of big data analysis in order for IoT to live up to its promise. If not, the consequences could be catastrophic. They could range from being annoying ones like home appliances that don’t function together as advertised to the life-threatening ones where pacemakers malfunction or hundreds of car pileups. The only way to keep up with this IoT generated data and gain the insights it holds within is using ML programs. In an IoT situation, ML can help companies take the billions of data points they have and narrow them down to what’s really insightful.
For example, wearable devices that track your health are already a burgeoning industry but soon these will evolve to become devices that are both inter-connected and connected to the internet, tracking your health and providing real-time updates to a health service.
The goal is that the doctor receives notification when a certain condition is met like say the heart rate has increased to an unsafe level, or even stopped, for example. To be able to call out potential problems, the data has to be analyzed in terms of what’s normal and what’s not. Similarities, correlations and anomalies need to be quickly identified based on the real-time streams of data.
In order to analyze the data immediately as it is collected to accurately identify previously known and never-before seen new patterns, machines that are capable of generating and aggregating this big data must also be used to learn normal behaviors for each patient and track, uncover and flag anything outside the norm that could indicate a critical health issue.
IoT, Artificial Intelligence and Healthcare
The rise of the IoT has been vital in the digital transformation of modern healthcare, particularly in healthcare delivery and monitoring. An additional source of rich data, healthcare IoT devices also allow for more connected, remotely managed healthcare equipment that can directly feed data not only into individual treatment plans and patient records, but into larger AI-driven healthcare analytics systems.
There has been a rapid growth in wearable healthcare devices, from fitness trackers to portable blood pressure and insulin monitors. The demand for IoT devices for wellness management has seen an upsurge in the recent years. In fact, according to the data from Global Industry Analysts, the market is set to be worth $4.5 billion by the year 2020, driven by growing need for more automated management and monitoring of chronic health conditions and boosted by the growing popularity of healthy living.
The growth in the demand for wearables is fueled by the change in perception of these devices that are for much more than just step counting and simple individual health monitoring. These devices allow for remote and at-home monitoring and management of serious health conditions, allowing doctors, gym instructors, nutritionists etc. to make better decisions and risk assessments. Faster change to treatments based on faster diagnosis of changing conditions can ultimately lead to reduced costs, enhanced quality and improvement in patient engagement.
Future of IoT is AI
The Internet of Things is only getting smarter. Companies are incorporating artificial intelligence, in particular machine learning into their IoT applications. The key is to find meaningful insights in data that can be actionable. Companies formulating an IoT strategy, evaluating a potential new IoT project, or seeking to derive more value from an existing IoT deployment may be inclined to explore a role for AI.
Gartner predicts that by the year 2022, more than 80 percent of enterprise IoT projects will include an AI component, up from only 10 percent today. As AI & IoT become more convergent, it is both useful and imperative to understand how these two important trends work in tandem to benefit specialists and the average person alike.
Author: Rakesh Rajalwal Chief Architect – BizAcuity