Dr Vishal Beri, CEO, Hinduja Hospital, Khar elucidates on the importance of time when it comes to healthcare delivery an how AI can change the scenario when applied strategically
Artificial intelligence is a branch of computer science that probes into the possibility of simulating human intelligence into a machine. John McCarthy, an American computer scientist and cognitive scientist, considered to be the father of AI coined the term ‘artificial intelligence; in 1956 at the Dartmouth conference.
These days, artificial intelligence is almost everywhere. It is impacting our lives in various fields, such as development of smart cars, detecting and avoiding potential financial frauds, identifying fake news, customer service etc. When applied strategically, AI has the potential to profoundly change the healthcare landscape altogether. The best thing about AI is that it can be a part and parcel of a clinician’s daily routine in multiple ways, patients’ safety and privacy and can eventually help to control overall costs.
Time is of crucial essence in healthcare delivery as any interventionist move is time sensitive. And since automation can take care of the processing part in regard to patient data and allied stats, it would, as well, go a long way in offering clinicians the extra-time advantage which is a rarity in the healthcare delivery space today, at least in this part of the world.
There is no denying that AI is surely a better and valuable appendage to a clinician’s constellation and goad ‘patient-facing activities’ in the longer run. Broad areas in healthcare have been earmarked as viable areas when it comes to attracting AI investment. Digitisation, engagement and diagnostics are three areas where AI is envisioned to be used in a phased manner. Digital tools to speed up operational process and improving on the patients- healthcare providers dynamics to harnessing algorithms aimed at providing diagnoses or health advice to patients, AI can be customised to alter operational processes eventually making healthcare delivery less expensive than what they are today.
AI is set to usher in radical changes in diagnostics, therapeutics and practices relating to maintaining patients’ safety and privacy- two critical pillars at the forefront of healthcare delivery. No one disagrees that it will be the primary driving force behind novel therapeutics, especially when it comes to difficult-to-treat diseases and healthcare administration vis-à-vis those diseases. AI today translates patient-driven biology/ data and catalyses predictive hypotheses. And by extension, accentuates patients’ centricity, more than anything else, in the entire spectrum of things. It is expected to play a seminally critical role in automated operations, precision surgery and preventive intervention which in turn rests on predictive diagnostics.
With its phenomenal number-crunching ability, AI can evaluate potential drug candidates exponentially faster and better than humans. A built-in artificial neural network, which mimics the human brain in recognising patterns and adapting to change, is attuned to process information faster and neater than our brains could ever handle.
Since a machine is fed with a lot of data and the problem, at hand, is formulated correctly, artificial intelligence has a chance to capture patterns which humans may not be able to capture. Pattern recognition is central to how AI is being harnessed to make sense of big data. And once the machine (read computers) encounters a large pool of data, the machine learning algorithms can identify patterns and in turn, use those patterns to make predictions or classify new data much faster than any human can possibly imagine. AI is equipped to sift systematically through a large amount of data and is, simultaneously, capable of answering questions to offer insights into how to do things the way they should be done.
This ability of AI has tremendous potential in certain medical sub-specialities such as radiology, especially CT and MRI, screening programmes for ophthalmology, diabetes etc. Initially, Watson Analytics was developed as a computing system for pure question and answer (QA). Over time, the system evolved dramatically with the advantage of cloud technology, improved machine learning and hardware capabilities; implementation of Watson Analytics in healthcare has significantly revolutionised the sector by assisting both patients and healthcare professionals. Improved organisational performance, effective diabetes management, advanced oncology care and ameliorated drug discovery are prominent trends of Watson that are transforming the healthcare sector.
Scientists are now capable of creating 3D models of the disease protein discovered in the drug candidates. At a later stage, the 3D model is used to perform molecular dynamics functions and ‘docking’ studies to further help evaluate the drug candidates, Achieving a 3D model of the diseased protein happens after proteins are crystallised – where crystals of the disease proteins are grown cosmetically- and X-rays then, used to suss out its structure.
Experts opine that the impact of various investments will likely be realised first in the operational and administrative side of the healthcare system rather than the clinical side. A PwC study reveals that “The majority of AI’s economic impact will come from the consumption side, through higher- quality, more personalised and more data-driven products and services.”
The success of AI in healthcare largely depends on assumptions over ‘data fitness’ and creeping biases when choosing data, in the first place. Alternatively, if the algorithm and its users do not sufficiently factor in those biases, then the output will also remain biased and skewed. Despite its potential to unlock new insights and streamline the way providers and patients interact with healthcare data, AI may bring considerable threats of privacy problems, ethics concerns and medical errors.
AI will require collaborative efforts from various pillars like regulators, technology developers, consumers and businesses.
4. How Artificial Intelligence is Changing Drug Discovery. Accessed from Nature, a multidisciplinary scientific journal https://www.nature.com/articles/d41586-018-05267-x