With the advent of gene-editing, synthetic biology, and stem cells technology, there is no question that healthcare is a sector that is far from being antiquated. However, it is paramount that one does not merely dismiss its inefficiencies.
According to a review published in the Journal of Stroke and Vascular Neurology, there are three areas in which the room for improvement is still very much vacant in the current healthcare practices. First, the large volume of healthcare data that remained unstructured in every hospital meant more valuable time for patients being spent in waiting areas rather than the examining rooms. Second, in this highly dynamic industry, doctors find it tenuous to be up-to-date with the latest medical information from journals, textbooks, or technological breakthroughs due to its overwhelming amount. Third, the inevitability of diagnostic and therapeutic errors in every clinical procedure.
AI for a More Accurate Medical Sphere
How can we address these issues and make healthcare better not only for the patients but for doctors as well? The use of artificial intelligence (AI) might provide an answer.
Clinical data often came in the form of, but not limited to, information on “demographics, medical notes, electronic recordings from medical devices, physical examinations, and clinical laboratory, and images.” The use of AI allows the conversion of the vast range of data into the machine-understandable electronic medical record (EMR) which in turn could map a patient’s medical history and compare it with their current state of well-being. This prescribes for the integration of machine learning application within the existing medical information infrastructure, which is ever-increasing in types vis-à-vis variety, velocity, and volume of data over the last decade. Such technology could open the possibility of illnesses being detected and diagnosed before they develop in severity.
Misdiagnosis is considered to be one of the most dangerous instances that exist in the healthcare system. Similarly, the EMR formulated by AI would alleviate errors because the difference between doctors and AI in making a diagnosis is the efficiency in detecting patterns from the broad range of patients’ medical records. The latter has been proven to out-perform the former. AI can connect the large yet disparate supply of healthcare data and make a diagnosis accurate 87% of the time.
It is a commonplace for doctors to do computer searches to identify new research studies, then determine which ones are relevant to the area they specialized in. Nevertheless, with the assistance of AI, doctors can sieve through the high density of information to get new medical treatments that are suitable for their specific patients. The estimated amount of scientific paper related to healthcare is around 2.5 million publications online. AI with abilities in understanding various concepts through medical literature will be proactive in remembering which studies have the doctors read and gives suggestions on new ones based on case descriptions they are working on. This also poses unique opportunities for terminal illnesses to be treated faster in the event of a relevant medical breakthrough.
Types of concepts that AI is able to recognized through medical literatures. (Courtesy of Stroke and Vascular Neurology)
The Challenges and Future of AI in Medical Sphere
Granted, to reap the full potential of what AI brings to the healthcare industry, significant challenges need to be overcome. One of them is the substantial amount of time necessary to ‘train’ the AI in the first place. Current research still requires the input of the complex machine learning algorithms and baseline indicators such as age, gender, disease history, and patients’ medical outcomes that are often collected in clinical research. And, of course, the integration of AI is also faced with skepticism in the medical community. The fact that it is devoid of the “warmth, sympathy, and understanding may outweigh the surgeon’s knife or the chemist’s drug” associated with the human doctors have taken the Hippocratic Oath.
Yet, since the business-as-usual healthcare industry is swallowing a greater share of national wealth in most countries, greater eﬃciency is precisely what is needed, at least so far as governments and insurers are concerned. Otherwise, prosperous societies may fail to cope with the needs of aging and growing populations and poor communities have to allocate more and more funding to healthcare instead of other areas of development.
Editor: Janitra Haryanto
Read another article written by Fikry Ghibran
 Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y., Dong, Q., Shen, H. and Wang, Y. (2017). Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology, 2(4), pp.230-243.
 The AI will see you now. (2019). The Economist, (9135), p.77.
 The World's Number One Portal for Artificial Intelligence in Business. (2019). An Epidemic of Misdiagnosis: Using AI To Solve A Quiet Crisis In Healthcare. [online] Available at: https://aibusiness.com/ai-misdiagnosis-optum-ai-healthcare/ [Accessed 20 Apr. 2019].
 U.S. (2019). Artificial intelligence may help doctors keep up with new research. [online] Available at: https://www.reuters.com/article/us-health-ai-research-reviews/artificial-intelligence-may-help-doctors-keep-up-with-new-research-idUSKBN1952WR [Accessed 20 Apr. 2019].
 Topol, E. (2019). Deep medicine. New York: Basic Books.