Moving away from generalized medicine to personalization
The article is based on a paper about the role of A.I. in Precision Medicine that was published in Expert Review of Precision Medicine and Drug Development.
Classical medical practice puts large groups of people in their focus and tries to develop clinical solutions, drugs or treatment based on the needs of the statistical average person. Disruptive technologies change that perspective completely. The basis of that transformation is data. Physicians are able to collect a vast amount of medical information about the individual through cheap genome sequencing, big data analytics, health sensors, wearables or artificial intelligence. Based on that specific knowledge, medical professionals can move away from generalistic solutions towards personalization and precision.
As disruptive technologies appear on the stage of healthcare, it becomes possible to get down even more deeply to the roots of diseases and treatments. The “one-size-fits-all” strategy will definitely start to crumble. It is the logical result of hundreds of years of medical research and accumulated knowledge. Currently, we know that everyone has a different genetic code, may react differently to pharmaceutics or may have a completely opposite reaction to treatment as assumed.
So why should we treat everyone with the same drugs or with the same method? And one of the most efficient means for precision medicine is artificial intelligence.
The place of artificial intelligence in precision medicine
As the National Institutes of Health (NIH) put it, precision medicine is “an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person.” To be able to ponder all those individual variations, medical professionals have to gather incredible amounts of information, and the ability to analyze, store, normalize or trace that data.
Big data analytics is one area where A.I., especially ANI comes into the picture. Within a couple of years, it will most probably analyze big medical data sets, draw conclusions, find new correlations based on existing precedences and support the doctor’s job e.g. in decision-making. Several companies recognized already the immense potential in A.I. for mining medical records (Google Deepmind and IBM Watson), identifying therapies (Zephyr Health), supporting radiology (Enlitic, Arterys, 3Scan) or genomics (Deep Genomics). My personal favorite is Atomwise, which uses supercomputers that root out therapies from a database of molecular structures. In 2015, Atomwise launched a virtual search for safe, existing medicines that could be redesigned to treat the Ebola virus. They found two drugs predicted by the company’s A.I. technology which may significantly reduce Ebola infectivity. This analysis, which typically would have taken months or years, was completed in less than one day.
Medical limitations and ethical issues around A.I.
To avoid over-hyping technology, the medical limitations of present-day A.I. have to be acknowledged. In the case of image recognition and using machine learning and deep learning algorithms for the purposes of radiology, there is the risk of feeding the computer not only with thousands of images but underlying bias. For example, the images tend to originate from one part of the U.S or the framework for conceptualizing the algorithm itself incorporates the subjective assumptions of the working team. Moreover, the forecasting and predictive abilities of smart algorithms are anchored in precedences – however, they might be useless in novel cases of drug side effects or treatment resistance.
Yet, medical as well as technological limitations of A.I. as well as ANI will still be easier to overcome than ethical and legal issues. Who is to blame if a smart algorithm makes a mistake and does not spot a cancerous nodule on a lung X-ray? To whom to turn to when A.I. comes up with a false prediction? Who will build in safety features? What will be the rules and regulations to decide on safety?
Although these burning questions cannot be answered in their entirety today, we have to do some preparations to be able to keep the human touch at the center of medicine and avert the possibility of A.I. becoming an existential threat to mankind feared by Elon Musk or Stephen Hawking.
What should stakeholders do to avoid the A.I. apocalypse?
- Set up ethical standards how to use A.I. on the micro and macro levels of the healthcare sector. We need specific guidelines starting from the smallest units (medical professionals) until the most complex ones (national-level healthcare systems). The principle of human comes first should stand at the core of these standards.
- A. I. should be implemented cautiously and gradually in order to give time and space for mapping the potential risks and downsides. Independent bioethical research groups, as well as medical watchdogs, should monitor the process closely.
- Medical professionals should familiarize with the basic concepts and working methods of A.I. in a medical setting to get over their potential fears and understand how the technology could help their work. There are concerns that A.I. will take over plenty of jobs in healthcare, yet, I believe the key is cooperation. Medical professionals should work together with technology if they want to achieve their full potential to heal patients.
- Patients should also explore A.I. in detail and how it might change their own everyday lives. It is important as in a couple of years, kids will probably play with A.I. friends such as the cute, dinosaur-shaped Cognitoys or learn from virtual reality teachers.
- Companies who develop A.I. solutions should communicate clearly and concisely towards the general public about the potential risks of utilizing A.I. in medicine. That’s also useful to avoid overhyping technology.
- Decision-makers at healthcare institutions & policy-makers should guide the process of implementing A.I. in healthcare along the principles and ethical standards they work out with other industry stakeholders. Moreover, they should push companies towards putting affordable A.I. solutions on the table and keeping the focus on the patient all the time.
I have no doubts about that A.I. will be the stethoscope of the 21st century and the backbone of precision medicine. It has the biggest potential to analyze vast amounts of data and offer insights to create personalized solutions and targeted treatments. Yet, we have to do everything in our power to ensure that A.I. remains safe, secure and efficient in fulfilling its mission as an aid in healing patients and helping the medical scene.