“Like a human being, a company has to have an internal communication mechanism, a “nervous system”, to coordinate its actions.”
Bill Gates, in Business at the Speed of Thought
Last week, we discussed the edge cloud as a key component of a future health learning system. This week, we will discuss the other two key components.
Swarm intelligence and collective artificial intelligence
Swarm Intelligence is the discipline of computational algorithms that is characterized by the collective intelligence of many individual entities. This domain is inspired by nature and groups such as insects and birds. Just like swarm intelligence, swarm learning is executed in a decentralized fashion without any leader so there is no need for a central custodian. A swarm network will have edge nodes that are used to exchange parameters for swarm learning enabled with blockchain technology. By using an application programming interface (API) to continually create an updated model, the swarm network continually “learn” and share knowledge but without sharing data. A swarm learning strategy will allow centers to share insights from the data (such as biometric data or echocardiograms) without sharing this data. While federated learning also keeps data at the edge, swarm learning will keep data as well as the parameters at the edge, thus obviating the need for a central custodian.
The artificial intelligence of medical things (AIoMT)
While the internet of things (IoT) is the interconnection of billions of physiologic devices to the Internet, internet of everything (IoE) will be essentially a “network of networks” to incorporate people, processes, and data to these devices to enable automation in data acquisition and analytics without human intervention. The internet of everything is basically an intelligent connection of people, process, data, and things; it builds on IoT by adding network intelligence and turning information into actions.
For a healthcare IoE to become AIoMT, embedded AI is the technological paradigm of integrating machine learning or other AI tools into the devices to attain a certain desired function. With biomedical devices, especially ones with continuous monitoring, the amount of physiological data received would be unmanageable by caretakers unless there is an “upstream” intelligent algorithm built in to the device to filter all the noise from the signal (for instance, filtering out normal sinus rhythm and sending an alert to the cardiologist only when there is an abnormal rhythm like atrial fibrillation or ventricular tachycardia). Recent work by MIT engineers of “tinyML” will enable these medical devices to have some built-in intelligence so that the clinicians are not overburdened with the surge of available data from these wearable devices. These AI-enabled devices not only perform inference but can also continuously adapt to new data for learning.
A health learning system
The convergence of the aforementioned edge computing, swarm network, and artificial intelligence of medical things can lead to the highest form of collaborative clinical learning and teaching in healthcare and clinical medicine. This convergence can create a real-time, real-world health learning system that can render a more equitable health care. With much more robust input of healthcare data from other sources such as social determinants of health, social media data, and pharmacy data, a swarm learning strategy can obviate most, if not all, the concerns about data privacy and security. This learning system is an AI equivalent of the biological peripheral and central nervous systems, to have a “brain” with continuous learning potential, as real world health data and analytics are made available. Future clinicians will learn mostly via this health learning system instead of solely from outdated textbooks and published reports.
While this artificial intelligence-enabled decision and operation support can be realized in the near future to care for any patient, anytime, and anywhere in the world, the human-to-human bond and friendship amongst cardiac caretakers in the world remains of paramount importance.
