By: Azhar Azam
Artificial Intelligence (AI) is a computer science in which machine simulates human intelligence through robotic processes or robots to perform tasks which are normally attributed to human brains such as learning, reasoning, visual perception, speech recognition, problem-solving, decision-making, and language translation.
Types of Artificial Intelligence
Intelligent virtual personal assistants such as Google Assistant, Apple’s Siri, Microsoft’s Cortana, Amazon’s Alexa, and Samsung’s Bixby are the some forms of artificial intelligence that help to find useful information on the web.
AI plays a key role in strategy games like Chess where computer can think of large numbers of possible moves. It will soon enable smart cars to self-drive themselves on the roads and take decisions based on what they see ahead.
Social networking, online shopping, online customer support, ride-hailing services such as Uber, auto-piloted flights, mobile banking, fraud prevention like unwarranted transaction, image search, chatbot, and instant translation machines are all sorts of artificial intelligence.
Presently 80% of the data generated through texts, Facebook posts, tweets, health record, clinical trial information, patient support program and others is unstructured and is predicted to reach 44,000,000,000,000 GB or 44 ZB (zettabytes) by 2020. Artificial intelligence will help to analyze this massive unstructured data.
Artificial Intelligence in Healthcare
Complimented by multiple technologies, Artificial Intelligence (AI) in healthcare is quickly rephrasing the modern conceptions of healthcare delivery by enabling machines to independently perform administrative and clinical healthcare functions such as predicting diseases, treating patients, and managing medical records.
In future, smart phones will not only be tracking your heart rate, counting on your steps, and keeping a log of your food but also AI-enabled devices will constantly be monitoring a number of health indicators – assisting in prevention of diseases such as diabetes and hypertension.
The AI storm has already hit the healthcare industry with the expansion of data management and communication applications or digital automation of medical records. In near-future, AI would frequently analyze tests, x-rays, CT scans, and other diagnosis as well as mundane tasks more quickly and more accurately.
Cardiology, Radiology, and Ophthalmology
Cardiology and radiology are the two disciplines where the amount of data to analyze could be excessive and arduous. With operationalization of AI in healthcare, cardiologists and radiologists may be relieved to most complicated cases only.
So with the development of AI in healthcare, not only large diagnostic data can be stored on computer systems for various kinds of diseases but also machines would analyze this information automatically to perform diagnosis.
AI will be able to detect suspected objects, normally human eyes won’t see and it will also enable the machine to learn, machine learning (ML). So instead of diagnosis coming from doctors in hundreds of cases, machine would diagnose in thousands of cases accurately and with a learning curve.
Tough radiological images (MT, CT. X-ray etc.) offer non-invasive study of human body however many diagnosis processes still need sample of physical tissues which are obtained through biopsies. Experts predict that AI will enable the next generation radiology tools to eliminate the need of physical tissue samples in some cases.
In ophthalmology, AI can detect eye diseases such as glaucoma in early stages which if not detected timely, can lead to lost vision. Retina images can be taken remotely and can be analyzed through AI techniques to identify the disease in early stages.
In coming times, machines would be doing all the diagnosis and the clinicians will be treating the patients on the basis of machine’s diagnosis.
Brain-Computer Interface (BCI)
The world is used to communicate through computers but creating direct interface between technology and human brain without any computer accessories is indeed a new concept.
Coupled with artificial intelligence, brain-computer interface (BCI) is a virtual communication between human mind and machine – enabling machine to receive signals from brain and transform brain intentions into control signals.
BCI technology thus allows people with neurological disorders to communicate with the external environments and other people through machine without the need of muscular or peripheral activity.
Remote Patient Monitoring (RPM)/telehealth and Digital Nurses/Virtual Assistants
Artificial intelligence is poised to impact lives through remote patients monitoring (RPM) or homecare telehealth and digital nurses or virtual assistants.
RPM technology is a kind of ambulatory healthcare service that allows a patient to use his smartphone medical device for performing routine test and send the data to healthcare professional in real-time. So, Patients in remote areas have a virtual access by telehealth system to doctors for treatment instead of visiting him physically. Critical cases however will certainly be referred to hospitals.
As AI grows, digital nurses or virtual assistants will be taking care of patients suffering from chronic diseases through smart platforms like Uber-type services. The startup Sensely’s digital nurse Molly helps to monitor patient’s condition and follow up with treatment between doctor visits, using machine learning (ML) program.
UK-based Babylon offers digital consultation to patients based on their personal medical history through mobile app. The patients could have virtual consultation with physicians and the app uses speech recognition to compare with against a database of illnesses. The service also provides drug prescriptions, referrals to health professionals, and book exams in nearby hospitals.
Precision Medicine (PM)
As disruptive technologies continue to surface in healthcare, the possibility of getting down deeply into the roots of disease and treatment has increased. The strategy of one-drug-for-all is bound to fail.
Everyone has a different genetic code so may suit one but at the same time may react differently or even may have a completely opposite reaction to treatment. So, why should be everyone treated with the same drug or with the same method?
Precision Medicine (PM) is the solution!
Precision medicine is the targeted therapy of every individual patient based on his or her genetics, environment, and lifestyle. PM can help to determine the best approach to prevent or treat a disease. And one of the most efficient means for precision medicine is Artificial Intelligence.
As genetics and genomics look for mutations and links to disease from DNA information, AI would help body scans to spot cancer and vascular diseases early and will predict health issues through genetics.
Drugs Discovery
Pharmaceutical development through clinical trials takes over a decade and costs billions of dollars. Artificial intelligence could foster and make cheaper this process. An AI-powered program was used to scan existing medicines that could be redesigned to fight the disease.
The program found two medications that may reduce Ebola infectivity in just one day while such kinds of analysis took months and years before. Thus AI can improve drugs discovery and save thousands of lives.
The Role of Human Physicians cannot be Ruled Out
In future, diagnosis and monitoring will become largely machine-driven and drugs discovery will also be aided, however creative scientific process is unlikely to be replaced completely.
The growth of AI will no way eliminate the role human physician which will shift the focus on providing expert advice and explanation. Artificial intelligence will provide the personalized information of the patient that physician will analyze. However, the physician will be needed improved data science skills in future to analyze vast data quantities.
Artificial intelligence should be used to assist human intelligence and not to replace it. If it is used as an early warning and recommendation, it can improve many lives. Otherwise, if health management is completely relied upon it, there are dangers.
Other important factors involved in AI are the technological limitations and scale of fallout. If a doctor makes a mistake, it will impact only a few people but if a smart algorithm diagnosis makes a mistake such as it does not spot a cancerous nodule or a lung x-ray, it can impact to thousands of them without any ethical and legal responsibility.
Experts are optimistic that AI can drive major improvements in predicting diseases, treating patients, and managing outcomes but warns that the technology is still in its early stages and there will be some growing pains along the way.
Sources:
OmniComHealthGroup: The Power of Artificial Intelligence in Healthcare
Longevity; AI Revolution: Artificial Intelligence Impacting Healthcare
The Medical Futurist: There is No Precision Medicine Without Artificial Intelligence
Novatio Solutions: 10 Common Applications of Artificial Intelligence in Healthcare
Health IT Analytics: Top 12 Ways Artificial Intelligence Will Impact Healthcare