Deciphering interfaces for AI and ML during the Trial Phases of Clinical Research

Unravelling the human genetic code had its share of challenges but then we possibly did not have Artificial Intelligence (AI) or Machine Learning (ML) to share a helping hand back then.

As most of my fellow brethren dabbling with AI and ML would know, the fundamentals of Data Science are an integral part of AI and Blockchain technologies. These intricate and complex technologies that constitute myriad networks between healthy volunteer trials to proof of concept/ dose range finding studies stretch the limits of human innovation.

If we have to distinguish between neuronal networks and scientific intellect – it’s akin to choosing between the Devil and the Deep Blue Sea since it requires many years of expertise to unleash the power of Big Data to formulate theories and enable speedy decision making for Real World Evidence (RWE) trials or explore aspects pertaining to retrospective analysis for past clinical trial data without having to recruit patients all over again.

In 1956, the term Artificial Intelligence was defined by John McCarthy as the science and engineering of making intelligent machines.

Since Artificial intelligence (AI) also known as machine intelligence, is intelligence demonstrated by machines, unlike the natural intelligence displayed by humans and animals. Leading AI textbooks define the field as the study of "intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.

A few ways of implementing AI and ML for Phase III studies could be to understand the following:

·       Building a linear regression model to predict the recruitment rate on an upcoming study so that a multicentric study could be planned in a better way

·       Linear Regression – ML would enable the creation of statistical power for a study and patient number distribution between innovator brand, test product and placebo

·       One could also create AI predict on what percentage of sites in different countries would be high on data quality and speed of patient accruals

A predicament that even the most seasoned clinical researchers would need to overcome would be around which type of AI to deploy where and when, the “how” aspect would automatically be answered effectively when we review the 7 AI types.

Seven Types Of Artificial Intelligence

1.    Reactive Machines

2.    Limited Memory

3.    Theory of Mind

4.    Self-aware

5.    Artificial Narrow Intelligence (ANI)

6.    Artificial General Intelligence (AGI)

7.    Artificial Superintelligence (ASI)

If experienced healthcare personnel were to recall any technology that has completely revolutionized medicine, it would be Artificial Intelligence. AI is an integral part of our everyday life and has different stages, types and branches.

Reviewing the different stages or the types of learning in Artificial Intelligence, we see parallels between phases of clinical research and the Stages Of Artificial Intelligence

One can review articles that state that Artificial General Intelligence, Artificial Narrow Intelligence, and Artificial Super Intelligence are the different types of AI. To be more accurate, Artificial Intelligence has three stages.

Types Of Learning In Artificial Intelligence

1. Artificial Narrow Intelligence
2. Artificial General Intelligence
3. Artificial Super Intelligence

These are the three stages through which AI can evolve, rather than the 3 types of Artificial Intelligence but implementation in clinical research is yet to be evaluated as working models.

Artificial Narrow Intelligence (ANI)

ANI is the stage of Artificial Intelligence (aka Weak AI) involving machines that can perform only a narrowly defined set of specific tasks. At this stage, the machine does not possess any thinking ability, it just performs a set of predefined functions.

Examples of Weak AI include Alexa, Self-driving cars and Siri.

Artificial General Intelligence (AGI)

AGI (aka Strong AI) is the stage in the evolution of Artificial Intelligence wherein machines possess the ability to think and make decisions just like us humans.

There are currently no existing examples of Strong AI, however, it is believed that we will soon be able to create machines that are as smart as humans.

Strong AI is considered a threat to human existence by scientists like Stephen Hawking who stated that the development of full artificial intelligence could spell the end of the human race. It would take off on its own, and re-design itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete and would be superseded.

Think of Terminator 3 – Rise of the Machines where the human race would be teetering on the brink of extinction.

Artificial Super Intelligence (ASI)

Artificial Super Intelligence is the stage of Artificial Intelligence when the capability of computers will surpass human beings. This is depicted in Sci-Fi movies where machines have taken over the world.

Terminator Series – the last movie comes in this category.