close x
download

Brochure

Enter Your Info Below And We Will Send
You The Brochure To Your Inbox!

aibotics go-ditial brochure (en)

Thank you!
Your submission has been received!

Oops! Something went wrong while submitting the form

brochure
Blog

How AI Can Assist Physician in Healthcare

Dr Jit-Seng Tan
20.2.2020

Dr Jit-Seng Tan (Vice-President of APARA – Asia Pacific Assisted Robotics Association)

Artificial Intelligence, or AI as it is commonly known, has been a buzz word in the technology world for decades. There have been AI winters, AI summers and now it is more of an AI autumn where we are starting to understand more of what we can and cannot do based on how our inherent bias from our limited viewpoint and the historical data which constitutes one of the bigger source of bias when training the AI applications. Medicine is understandingly both an Art and an evolving Science. We have much data now, from the scans and imaging reports showing anatomical, and sometimes physiological changes, to epidemiological data helping the planning of impending medical needs like a flu epidemic or dengue clusters.

The AI technologies that are being used in healthcare can be patient facing and non-patient facing. And why AI? Is it more cost efficient, safer and faster than our current workflows?

The use of AI will likely be organized to assist with non-patient facing processes. They can be used in hospital processes, in medical records, in data analytics, in research etc. With a larger amount of information and complexity of modern funding and compensation mechanisms, AI can be useful and be trained to plough through the trillion bytes of data generated and make sense of them in order to further our understand on the on goings.

AI systems which are applied in patient facing solutions, like symptom checkers and simple advisory chatbots, can be very useful, especially in regions where accessibility to modern healthcare is difficult. Some of my fellow colleagues are doing very good work in locations which are remote and these applications provide timely advice and management plans to patients in need. AI can potentially lighten the workload of the overburdened healthcare system by collecting pertinent data of the patient, first, and presenting the data with recommendations to the clinician before the patient is seen. This can potentially provide another safety net for both the patients and the physicians as well.

Quality data collection is the key in getting AI to work and as such, the clinician imputing data might have to be skillfully trained in future. For example, writing alcohol abuse versus alcohol dependency may mean the same to the clinician but different to the engineers setting the AI system up to do further interpretation and research.

Data collected from more objective sources like images and numerical figures can be better utilized now to train the AI system, but for all the subjective inputs. Unfortunately, it is not a one size fits all concept. Perhaps we can start such training in medical schools in the future, if AI was to be deployed.

There are many cultural factors as well as environmental, genetics, social, and even family considerations when us physician do the consultation. I would argue it will not be possible for AI to replace clinician, but AI shall be assisting physicians in some way or another in this digitalized era of the 4th industrial revolution.


AIBotics 2020 will be held in August 2020 with a 2 Day Conference and Exhibition, plus an additional day of workshops and tutorials. The theme for 2020 is “Augmenting the Human Potential”. This understanding and implementation of AI in Robotics across the wide range of applications aim to bring together the most progressive end-users, first-class speakers, and innovative solution providers.

...
more posts