Predictive Analytics, ML, AI – Do I Have to Choose?

Predictive Analytics, ML, AI – Do I Have to Choose?

There is plenty of confusion about which is better, predictive analytics, machine learning (ML), or artificial intelligence (AI). Which one should a company choose to use? The short answer is, you shouldn’t choose. Each of these aspects of data analysis may be useful to your company, and they have strong relationships with each other. The more detailed answer requires a deeper understanding of what these terms mean and how they relate to each other.

Defining the Options: Predictive analytics, ML, and AI

Predictive analytics is analyzing past events to forecast what might happen next. By understanding the past trends thoroughly and analyzing what factors contributed to those trends, we can predict how current factors will affect the future with some level of accuracy. While its never certain – predictive analytics is far more often correct about future predictions than simply relying on human experience or hunches. Predictive analytics provides probable answers to questions like Will demand for my product be higher next month? Will a part in this machine fail in the next year? Will next quarter’s profits be lower than last year?

Predictive analytics is usually contrasted with descriptive and prescriptive analytics. Descriptive analytics describes what is currently happening or what has happened. Whilst prescriptive analytics involves analyzing past events to recommend what decision should be made next.

Naturally, trying to predict the future is considerably
harder and often requires more data to be reasonably accurate a fair amount of
the time. All of these require careful analysis of data, but in general,
descriptive analytics is the simplest. Prescriptive analytics goes a step
further, is that much more complex, and often requires even more data and even
more intense forms of analysis.

Machine learning (ML) is a process that uses
mathematical models called algorithms to make sense of data in one way or
another. The math may determine things like which data should be grouped
together because it is similar, or what patterns exist in data that are not
easy for a human to find, or what data factors influence a change in another data
set. Machine learning gets its name because a single mathematical algorithm may
be used for a wide variety of purposes. It is the data that determines the
outcome, not a programmer who writes code instructions. Different data will indicate
to the algorithm different things, teaching the ML model about itself and data
like it. Hence, the machine learns from the data.

The more data that can be provided to train an ML algorithm,
the more accurate the ML model will become, the more it will learn about that
data. It often requires a LOT of data to make a good ML model, and it often
takes a powerful computational engine to train the model with all of that data.

Once trained, an ML model can understand a dataset to the
point where it may be used to predict where it will go from here (predictive
analytics), or it may find patterns in data that indicate what action would
affect the data in a desirable way (prescriptive analytics). So, machine learning
is a sophisticated method of data analysis that can be used for both predictive
and prescriptive analytics.

Artificial Intelligence (AI) is a general term that indicates any form of technology where a machine mimics the capabilities of a person. An example is natural language processing (NLP), where you speak to a machine in your own natural language, and the machine translates that into some form of instruction. It may execute that instruction, perhaps finding information for you, and then it may provide the answer back to you also in your own language (Natural Language Generation (NLG)).

Artificial intelligence also includes robotics and
automation, where a machine recognizes a certain condition, either via coded
instructions or from a pattern it recognizes due to ML, and does something
similar to something a human would do under those conditions.

Making a Decision

So, back to the original question: Which is better, AI, ML,
or predictive analytics, and what do I choose? You might as well ask which is
better – a seed, an apple, or a tree. They are all interrelated, and which you
use depends on your need. If I needed to grow something next spring, a seed
would work best. If you were hungry now, an apple would be the clear choice. If
you wanted shade, the tree would definitely be the way to go. But if you plant
the seed, it will grow into a tree that bears apples, so they are all part of
each other.

Based on your business requirement, you will choose what you
need.

For example, if your business is manufacturing, you might
use ML to analyze the sensor data from the components of the manufacturing
machines to determine if theres a pattern that predicts (predictive analytics)
when a component will break soon. Then, you might have that machine
automatically stop before it breaks, and send a signal recommending
(prescriptive analytics) that youll replace a certain part. This preventative
maintenance system would be a good example of artificial intelligence in
practical use.

Another example might be if you were an online retailer. You
might use descriptive analytics to determine that there are a large number of
blue sweaters in your inventory. Then you might use machine learning to find
the group of people who buy from you who tend to like blue clothing and live in
an area that will get chilly soon, so they are likely to need a sweater in the
near future (predictive analytics). You might then automatically recommend
(prescriptive analytics) a blue sweater to them the next time they are online,
especially if they purchase blue pants or maybe gloves. This is another good
example of artificial intelligence used in business.

Accuracy in Time to Act

As you come to understand that AI, ML, predictive and
prescriptive analytics arent just buzzwords invented to sound complex, you
will often begin to see what analytics you need in your organization and how
each type can help you. When you get ready to put advanced analytics into
practice, remember these three things that will help you make it work.

1. Accuracy Using all available data vastly
   improves the accuracy of ML models. Dont settle for taking a sample. Make sure
   that your technology and methods allow full analysis of all relevant available
   data. Also, over time ML models become less and less accurate as data and
   conditions change. Be sure to manage, measure accuracy, retrain, and update
   models as often as necessary to maintain the highest possible accuracy.
2. Action The best information, prediction, or
   recommendation in the world will be worthless if you do not take appropriate
   action in the moment when it will do the most good.
3. Automation Allow the machine to do what it
   does best. Automated monitoring, analytics, and sometimes even automatically
   triggered actions can take best advantage of rapidly changing business
   situations.

The more advanced forms of analytics can provide companies
with a huge competitive edge over their competitors who dont use them yet.
Understanding the terms and what they mean in the context of your business is
the first step to reaping the benefits.

Published at Tue, 15 Dec 2020 01:18:45 +0000

Top 10 AI and machine learning stories of 2020

Toward the tail end of pre-pandemic 2019, Mayo Clinic Chief Information Officer Cris Ross stood on a stage in California and declared, “This artificial intelligence stuff is real.”

Indeed, while some may argue that AI and machine learning might have been harnessed better during the early days of COVID-19, and while the risk of algorithmic bias is very real, there’s little question that artificial intelligence, evolving and maturing by the day for an array of use cases across healthcare.

Here are the most-read stories about AI during this most unusual year.

UK to use AI for COVID-19 vaccine side effects. On a day when vaccines, developed in record time, first begin to be administered in the U.S., it’s worth remembering AI’s crucial role in helping the world get to this hopefully pivotal moment.

AI algorithm IDs abnormal chest X-rays from COVID-19 patients. Machine learning has been a hugely valuable diagnostic tool as well, as illustrated by this story about a tool from cognitive computing vendor behold.ai that promises ‘instant triage” based on lung scans – offering faster diagnosis of COVID-19 patients and helping with resource allocation.

How AI use cases are evolving in the time of COVID-19. In a HIMSS20 Digital presentation, leaders from Google Cloud, Nuance and Health Data Analytics Institute shared perspective on how AI and automation were being deployed for pandemic response – from the hunt for therapeutics and vaccines to analytics to optimize revenue cycle strategies. 

Microsoft launches major $40M AI for Health initiative. The company said the the five-year AI for Health (part of its $165 million AI for Good initiative) will help healthcare organizations around the world deploy with leading edge technologies in the service of three key areas: accelerating medical research, improving worldwide understanding to protect against global health crises such as COVID-19 and reducing health inequity.

How AI and machine learning are transforming clinical decision support. “Today’s digital tools only scratch the surface,” said Mayo Clinic Platform President Dr. John Halamka. “Incorporating newly developed algorithms that take advantage of machine learning, neural networks, and a variety of other types of artificial intelligence can help address many of the shortcomings of human intelligence.” 

Clinical AI vendor Jvion unveils COVID Community Vulnerability Map. In the very early days of the pandemic, clinical AI company Jvion launched this intereactive map, which tracks the social determinants of health, helping identify populations down to the census-block level that are at risk for severe outcomes. 

AI bias may worsen COVID-19 health disparities for people of color. An article in the Journal of the American Medical Informatics Association asserts that biased data models could further the disproportionate impact the COVID-19 pandemic is already having on people of color. “If not properly addressed, propagating these biases under the mantle of AI has the potential to exaggerate the health disparities faced by minority populations already bearing the highest disease burden,” said researchers.

The origins of AI in healthcare, and where it can help the industry now. “The intersection of medicine and AI is really not a new concept,” said Dr. Taha Kass-Hout, director of machine learning and chief medical officer at Amazon Web Services. (There were limited chatbots and other clinical applications as far back as the mid-60s.) But over the past few years, it has become ubiquitous across the healthcare ecosystem. “Today, if you’re looking at PubMed, it cites over 12,000 publications with deep learning, over 50,000 machine learning,” he said.

AI, telehealth could help address hospital workforce challenges. “Labor is the largest single cost for most hospitals, and the workforce is essential to the critical mission of providing life-saving care,” noted a January American Hospital Association report on the administrative, financial, operational and clinical uses of artificial intelligence. “Although there are challenges, there also are opportunities to improve care, motivate and re-skill staff, and modernize processes and business models that reflect the shift toward providing the right care, at the right time, in the right setting.”

AI is helping reinvent CDS, unlock COVID-19 insights at Mayo Clinic. In a HIMSS20 presentation, JohnHalamka shared some of the most promising recent clinical decision support advances at the Minnesota health system – and described how they’re informing treatment decisions for an array of different specialties and helping shape its understanding of COVID-19. “Imagine the power [of] an AI algorithm if you could make available every pathology slide that has ever been created in the history of the Mayo Clinic,” he said. “That’s something we’re certainly working on.”

Twitter: @MikeMiliardHITN
Email the writer: mike.miliard@himssmedia.com
Healthcare IT News is a HIMSS publication.

Published at Mon, 14 Dec 2020 22:18:45 +0000