Medical Research in the times of Blockchain & AI/ML

Introduction

Experimentation has been an essential form of human learning. Trying innovative ideas leads to either success or failure and then only development happens when we know what to do and what not to do. If not all, most of medical practice is evidence-based and what might appear empirical at one stage becomes evidential at some time in the future. From learning & teaching medicine to the discovery of newer diagnostic and therapeutic methods experimentation and research form the core thought process of scientific temper and are an existential part of healthcare as well as Medical Education.

The primary step of any experiment is to collect correct and validated data and then subject it to further statistical analysis and mental manipulation as may be required to reach an understanding of the problem being solved. The advent of computers has meant that large amounts of data can be collected and stored and further analysed in a truly short period to get previously unimaginable insights about many a medical mystery. This analysis and use of Big Data by Machine-Learning enabled Artificial Intelligence is poised to radically change the understanding that humans have about the human body both in diseased as well as in normal conditions. This understanding of Human Body and its processes will also not be limited to Humans but also will be inbuilt into machines which can deliver and optimise healthcare in the coming days.

At the core of all experimentation and research is data, Big or small. If computerised, electronic databases can be considered as Radical changes the advent of Blockchain as an established technology can be considered as a revolutionary change to those radical changes made by electronic databases. A piece of information or data is only as good as the authenticity and legitimacy it identifies itself with; Blockchain is the game-changing disruptive technology which brings absolute incorruptible trust into the system where all data is authenticated to be genuine and comprehensive.

We present to you a brief overview of how future Medical Research will possibly look like in the exciting times of Blockchain & AI/ML.

Data Collection

The advent of wearables and internet of things (IoT) means that huge data can be accumulated from point of care without the involvement of humans; thereby avoiding inevitable human error & tardiness. Machine to machine communication will ensure high sampling rates and speedy transfer of data to storage servers feeding Big Data to the ever-hungry Machine Learning Artificial Intelligence. Coupled with biometric identification and multiple levels of identity certification blockchain will ensure that the data is pure and true. There will be no chance of data being lost or even deficient in most cases.

Examples of such data would be all the information which can be collected by wearable devices like BioElectric parameters from the Heart & Brain as well as BioChemical markers like Blood Glucose & Cortisol. Other IoT devices can track a person’s activity, sleep, eating habits, body composition as well as general environmental parameters like air quality and pollution levels. When this is done on a global scale, we can see previously unimaginable insights into Human health and disease.

Data Storage

Devices Connected with each other using 5G and other upcoming technologies will ensure a continuous stream of data into the cloud. Since all of these petabyte levels of data comes in using Blockchain technology the purity and validity of such data are ensured. Security of this data against corruption and privacy-related issues will be things of the past as the data generating users can themselves control and allot which data can be shared with whom. The privilege of data control by the individual has never been attempted just because such technology did not exist before. Anonymity and privacy have been concerns which have prevented much progress due to lack of tech to solve it.

Such huge data, of course, would mean greater storage capacity and the exponential increase of server farms, but future semiconductor technologies will ensure that more can be stored in less with less heat being produced and less electricity being consumed. The enhancement of storage technology is one of the perceived hiccups to this effect.

Data Analysis

The Machine Learning of Artificial Intelligence feeds on Big Data. Such an ecosystem would flourish when the data flow is more than adequate as well as unadulterated. The advent of Quantum Computing technology is going to fundamentally change AI as well. These machines can not only produce sufficient and suitable statistical analysis as may be required by human operators in an instant but also create and conjure ML generated insights and analysis as the algorithms may deem fit.

Taking the example of AlphaGo, we can see that Machines can take the challenge of solving a problem in ways that were previously beyond the cognitive reasoning of the best of Human minds. When given the target of preventing disease using predictive analysis, AI is likely to flag out behaviours and patterns which could lead to future events especially in the most morbid conditions like Cardiovascular Diseases and Diabetes.

This preventive model of healthcare would be possible with the collection and analysis of seemingly insignificant but continuous data from a person over a longer period than previously considered. With quick modelling and simulation predictions can be generated much earlier and corrective & remedial measures can be put in place which can be trailed by diligent follow-up and testing and validating the effectiveness of such measures.

Results Presentation

Data analysis of Big Data can present with Big results which may not be possible for humans to easily ingest. While smaller & focussed conclusions can be easily related to Individuals or groups for timely action as in the case of predictive medicine, larger datasets will produce results which would be understood by humans only in part with the remainder left for the Machines to simplify and digest. This may seem an unsatisfactory result, but the partial solutions offered in itself will be much more than what has been previously endeavoured.

An example would be of the Electrocardiogram where data is collected primarily from Analogue signals coming from 3+6 (10) electrodes on the body. A machine enhanced EKG could easily take in data from 26 (64) electrodes & over a longer period and present a graph which we as humans might not even recognise leave alone interpret. But such a 3-dimensional electronic picture of the electrical activity of the heart will surely give exceptional predictive models for the machine to recommend enhanced preventive therapeutics.

Implementation

The first use of such a large application would be predictive analysis and recommendations of preventive measures to dispel the damage from much of the disease burden plaguing the world today. Most of the non-communicable diseases can be prevented if corrective measures are taken ahead of time. If not their morbidity and mortality can be considerably mitigated by early treatment options.

Secondly, the knowledge of this magnitude will form the research of tomorrow. The time cycle from data collection to publication after a peer-review currently makes most research redundant by the time it reaches the bed of the patient from the lab bench. Blockchain can render peer reviews & scientific journals redundant when the source and validity of the data are permanently verified to be genuine and immutable. Open Databases made public and free to use can exponentially increase the availability and affordability of medical knowledge not just to the healthcare professional but also to the common man.

A third and most important aspect of research is to convert the science to policy. Research updates Continuously & Freely available can enable not just individuals to plan and prevent diseases in their own life but larger governance and regulatory bodies can make dynamic and up-to-date decisions on-the-go to make substantial modifications in Healthcare policies. Whether it is an infectious pandemic or endemic NCDs a strong public voice can be generated when there is appropriate and definite data to work with.

Data Feedback

The application of the policies and preventive models will further create data which will give better predictive analytical power based on the feedback of what works and what does not. While a few decades ago this cycle would take years the use of technology can shorten the feedback times to a few days if not hours. This would keep the system learning and it would keep teaching us how and what to do with our healthcare.

Conclusion

The Three technologies of Artificial Intelligence, Blockchain & Quantum Computing developed independently but working synergistically will bring radical change in how we approach Healthcare and its related education and research and policy. Replacing human vulnerabilities and tardiness from a system will ensure that Human Beings do what they are best at, BE HUMAN. It will restore the Human aspect to healthcare and governance and keep more people more healthy. And make the world a better place to live in.