Original article was published on Artificial Intelligence on Medium
The use of AI/Machine Learning in health care is the new dawn of Computer Science and Medical Sciences. AI is prepared for scanning MRI images, brain tumours, blood pressure variations and even Coronavirus. The increasing accuracy of AI systems is leading us to a new age where doctors are no more humans but programs running in silicon chips and bundles of wires.
Imagine a clinic where you need to feed in some data and you get medicines from it!
To what extent is this possible? Could ever AI replace human doctors? Come let’s explore.
The AI Physician
You go to the doctor. He/She will probably ask you about the symptoms you’re facing. Some physical checks will be taken. Based on his/her knowledge, he/she will give you medicines or request consultation from a specialist doctor. Where can an AI fit here?
You may provide your symptoms to software instead. A digital thermometer is attached to your body which feeds temperature reading to the software. Similar sensors could be used for blood pressure as well. Two cameras can check the eyes and the tongue just as human doctors do. Breathing patterns could be fed likewise. And we can always have the patient’s past history.
That’s All! The system has gathered various features regarding your body’s physical configuration. It now knows the state of your body. Now, we can use a tree-based algorithm that could detect simple diseases and provide medicines too! That’s what physicians do with their knowledge and experience.
Just imagine such a clinic! Maybe, AI could not be trustworthy as a human doctor but we can wait for developments.
Do current AI systems have accuracies that could match the human doctors?
To understand this, we need to take the example of Google AI’s Diabetic Retinopathy Detection Models. This model detects diabetic retinopathy in patients and has been working in field trials too. On the blog, they’ve mentioned,
We saw clear evidence that showing model predictions could help physicians catch pathology they otherwise might have missed. In the retinal image below, our adjudication panel found signs of vision-threatening DR. This was missed by 2 of 3 doctors who graded it without assistance; but caught by all 3 doctors who graded it when they saw the model predictions (which accurately detected the pathology).
So, doctors may miss the DR signs from a scan, but when they equipped with Google’s AI model there are lesser chances of doing so. A single brain could miss some scans but having two brains at a time would lessen that probability.
Another example comes from the National Cancer Institute, where an AI system can detect three types of brain tumors which over 94.6 % whereas conventional pathologist-based analysis had an overall accuracy rate of 93.9%.
This technology is especially encouraging for patients with newly detected tumors and patients with [recurrent tumors] who are undergoing second or third surgeries,” said Daniel Orringer, M.D., of NYU Langone Health, who helped lead the study.
Making medicines with AI
Making drugs for a particular disease requires testing and enormous research too. Why not automate the process using an AI? An AI can walk through millions of compounds and materials to find the best combination suited for a disease in a shorter time.
This is a particular advantage of machine learning. Because the algorithm itself does not have human biases it will search for the best way to meet criteria, and will search in places humans would not. Some of these will be absurd and humans will need to refine the models to redirect thinking. But — if other industries are anything to go by — some will hit on genuinely new and innovative solutions. While humans are put off by approaches outside their conventional thinking, AI can open up new routes to the desired result not previously considered.
AI has also been used in predicting drug side effects which could facilitate research to a great extent.
Any drug for this pandemic will be of great help. We’re using AI to predict new drug candidates for this disease too.
AI is continuously advancing in healthcare. One day, probably you’ll go to a family doctor for checkups. Wait, not a family doctor but a Robo-family doctor! Thanks for reading.