Skip main navigation

£199.99 £139.99 for one year of Unlimited learning. Offer ends on 28 February 2023 at 23:59 (UTC). T&Cs apply

Find out more

Introduction to AI for Healthcare and key challenges

Hi, Ladies and gentlemen, welcome back to this first course of this MOOCs on the basic introduction of artificial intelligence for health. So in this I will be just taking back you to our how innovations in healthcare industries, and what had happened in last few say in last hundred years. So if you remember, it was maybe in 1885 or 1890s when first x-ray was discovered. So all great inventions are actually discoveries in medicine. So if you see, remember about, what to say, x-rays, you know it was discovery and it has changed the way we were practicing the medicine.
Because we were not way only able to see what’s happening on our body, but also we were seeing inside our body using x-rays. And then comes discovery actually of antibiotics that is Penicillin by electron Alexander family. Now what he has done was he was not so accurate guy he displaced he’s all, what to say, research things, and he went back for vacations, and when he came back to his lab and he found in his disk some fungus which are killing Streptococcus bacteria. And that’s how he started developing antibiotic Penicillin from this thing.
And then our third invention is about using of information technology computers in healthcare, CT scanning, then came MRI and various different sort of diagnostic tools or information technology tools were applied in healthcare. However the fourth revolution or innovations is because of information, technology, computers, internet and big data analytics. And if you see artificial intelligence, internet of things, so all these are forming as a fourth innovation in health industries. So this class will be more about how we are using these artificial intelligence for our healthcare sector, in our healthcare sector. And also I would like to emphasize on the key issues of current healthcare practices. So what are the big issues is most of us just ignore participatory health.
I will explain in details what do I mean by participatory health. Because as it sounds, we want patients to participate in their own healthcare management. Then, we are still focused on curative care and I think that is not the way we should more focus on preventive care, not curative care. And I will show an example how we can shift this practice from curative care to preventive care. And obviously, medical errors in hospitals, and the fourth and the most biggest problem is one-size-fits-all care. What do I mean by it is, if you take any lab values, you take any, what to say, a protocol of treatment, we use similar for all population.
But although we are humans, we are not same entities. We have our own genetics, we have our own, what to say, information accordingly we should be treated. So one disease for me is not same for other people. So we need to personalize, we need to go towards precision medicine. So that is what we will be talking about. So to begin with, let me ask, I want you to reflect “what is natural intelligence ? ” Then we go back to artificial intelligence, and then why we need artificial intelligence?
So if you know, human is the most complex creature of species in whole universe so far, and brain is the one which is the complex things I think even after advancement in our science and technology. Still we are unable to know exactly how brain is functioning. And if you see the slide, so this is it took some million years for us to evolve from ape to the last guy who is using computer So what is different between these people ? One is the time. Maybe after some thousands of years we went one stage to another stage That is okay. Then what is the second most important thing which is changing them from that from Apes to human using computers ?
That is nothing but our natural intelligence. The way we started thinking analyzing thing, analyzing the information, passing the information, working with the information. That is what made us change. But the good thing is, or I don’t know good or bad, but the thing is that now we went into the computer already. So that is comes the artificial intelligence. So we want to work with we want to use computer as our brain. But I want to emphasize, it’s not as simple as the natural intelligence has evolved in millions of years, and we want to develop artificial intelligence in few decades.
So I think most probably in last one or two decade, these artificial intelligence has evolved because of high power of computation power and digital, availability of digital data. So definitely, it may take a lot of time till computers or machines to come at least to the level of human intelligence. It may not take it may take some more of time. Then if you see what is this ? This is our neurons. Clusters of neurons. And I want to make it very clear, our brain never works with information. Our brain only analyzed the electricity signals. And computers again, they also don’t work with the information, but with the binary zero, ones.
So after these analysis, we need to interpret what we need out of this analysis. And definitely you might be knowing this is a single neuron in our brain, and there are billions of them. And what this neuron does is just it collects signals from surrounding cells, surrounding neurons. And if you see there is an axon, and it come, what to say, analyze these signals and pass one result of these signals to the neighboring axon, to the neighboring, what to say, neuron. So this is how a brain collect information, analyze information in one way. And hope you know this gentlelady, who is she ? She’s Sophia. And who is she ? She’s robot artificial intelligence.
She can see us, she can listen to us, she can make a dialogue with us. And you might be knowing she is the first robot who got citizenship of Saudi Arabia. And now I come back to our main topic artificial intelligence. How artificial intelligence work in computers. So what we are trying to do is that we are trying to imitate our natural intelligence model. Meaning cell axon neuron pausing to another neuron, other neuron pausing to another. So at the end we get some results out of that. Exactly in the same way, we put notes behind notes, notes behind notes, which we call them layers or hidden layers. And we do some input variable and video output variable.
And what we see that we really don’t care what is happening in the between input and output. Our interest is if I put these fighting’s and I want these two things diagnose yes diagnose no. So these are like our two outputs and they can be find to how many variables. You have 25 or 200 as many number of variables you can put in, and you can get as many variables as an output. That is exactly if you see how the artificial intelligence or CNN, there are many many models in artificial intelligence. So this, the basic idea is this, you put some variable in, and you take out your interesting variable of output.
And what really this computer does with image processing, so we are not talking about text, we are not talking about numbers, we are talking about image. The image, each image has its own shape ,its own form. So the system will analyze this pattern. If you see the rise how it should be. if you see the plane how it should look. So same way, if you put like hundreds and thousands of images of cats, dogs any animal, then it should learn. Okay, if this is the shape, it is attack as one given animal may be cat, so we should provide them because cat is not one shape. There are like a hundred different shape different angle.
The cat can be sitting, cat can be sleeping, cat can be playing. So we need to show all these images to the system so that whenever we want a new our testing image, so we even don’t know what will be our testing image. The cat will be sitting, cat will be running, or what cat will be doing. So we need to have all this information in the computer, then when you show this cat again, maybe sleeping, maybe playing, then the system has to some sort of correlation with the patterns, with the futures extraction they have with the previous cats. That’s how it works.

This video will introduce what is AI and its application in healthcare. With a wide range of benefits, there are also several challenges faced in the application of AI in healthcare domain.

After watching the video, could you explain the model how artificial intelligence imitate natural intelligence? Although there are debates on AI has more advantages over natural intelligence, why does the AI still need to imitate patterns from natural intelligence? Leave some comments before you move to the next activity.

This article is from the free online

Artificial Intelligence for Healthcare: Opportunities and Challenges

Created by
FutureLearn - Learning For Life

Our purpose is to transform access to education.

We offer a diverse selection of courses from leading universities and cultural institutions from around the world. These are delivered one step at a time, and are accessible on mobile, tablet and desktop, so you can fit learning around your life.

We believe learning should be an enjoyable, social experience, so our courses offer the opportunity to discuss what you’re learning with others as you go, helping you make fresh discoveries and form new ideas.
You can unlock new opportunities with unlimited access to hundreds of online short courses for a year by subscribing to our Unlimited package. Build your knowledge with top universities and organisations.

Learn more about how FutureLearn is transforming access to education