Artificial Intelligence

The robot will see you now


This is truly the age of science and technology, which now touches almost every aspect of our life. Every week, new apps and devices pop up to make our routines and tasks easier. However, I have some reservations about artificial intelligence and the direction in which research on cloning is headed.

Programmes and features on our phones are increasingly being driven by artificial intelligence, and I worry about the data being collected day in day out. It seems by reviewing your phone usage, technology companies can trace you to any corner of the world, whether you are at a cabin in the Swiss Alps or inside a toilet in Mexico. Convenience offered by technology has come at the cost of privacy. Government agencies around the world use the excuse of surveillance to look into your private life in different ways.

Artificial intelligence is slowly making inroads in the field of medicine now. An article published in Medscape last month talked about Google’s patent application for a deep machine learning system, which uses longitudinal electronic health record (EHR) data to predict future health events. To study various parameters of its system, Google collaborated with the University of California, San Francisco, Stanford Medicine, and the University of Chicago Medicine. The company said its system was able to predict in-hospital mortality, prolonged stay, unplanned readmission, and final discharge diagnosis with an accuracy superior to that of existing predictive models, Medscape reported.

In its patent application, Google has said that its predictive system could help doctors prioritise patients and could point to more relevant information in a patient’s chart. These details could help healthcare providers identify areas of concern or intervene to reduce the likelihood of an adverse event, the application added.

One medical centre in the US used machine learning to predict which patients are likely to develop sepsis. Another facility saved $4 million by using AI-driven predictive analytics to prevent readmissions.

In a trial, Google’s EHR system predicted 86 per cent of the time whether patients would have a lengthy hospital stay, and it had accuracy rates of 77 per cent and 95 per cent while forecasting readmissions and mortality, respectively. These are astonishing figures, but we must remember that the system was tested in a developed country. The results may be quite different in Indian settings where hospitals and doctors attend to a large number of patients. Even without predictive technology, seasoned physicians have been correctly predicting patients’ chances of recovery and survival for years.

The journal BMJ addressed interesting facets of artificial intelligence in medicine. Dr Goldhahn from Switzerland believes that one day, doctors will become obsolete as machines become make more accurate diagnoses in certain specialties. In 2017, a robot cleared the medical exam in China, getting more than minimum score of 96 points. Such machines will have no conflict of interest, and AI could drastically reduce healthcare costs. Many people believe that machines won’t be able to provide the comfort and assurance a trusted physician can. However, no human can keep up with the evolving complexity of medicine and a physician could produce better results with the help of AI tools and systems.

The so-called ‘death clocks’ on the internet calculate your life expectancy by reviewing details such as your date of birth, height, weight and lifestyle choices (whether you smoke or drink). But these programmes currently don’t take exceptions into account: some people will live longer despite the odds. I think these programmes will evolve and produce more accurate results by studying all factors.

After a few decades, perhaps you will have the option of visiting a physician or a robot. Whom will you choose? I think it’s going to be a difficult choice.



READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.