Technologies like artificial intelligence and robotics are leaving marks in the healthcare sector. Regardless if we are on the supplying or receiving end of healthcare, these developments will impact us and our health. Here are some ways that AI and robotics are being used in healthcare and what we can expect in the near future.
Back-end Process Optimization
Hospitals are making use of AI for back-end processes like any other business. Hong Kong Health Authorities utilizes this technology to manage staff schedule, while Corti uses AI to dispatch emergency services.
Pharmaceutical Research
The explosion of data means more raw materials to run analysis and draw meaningful insights for all research fields, including medicine. AI algorithms improves research technology and allows pharmaceutical firms to make better sense of unstructured data available. Recognizing this, Teva Pharmaceuticals partnered with IBM Watson in 2016 to harness intelligence for drug repurposing purposes.
AtomNet developed deep convolutional neural network to study molecules and interaction of molecules. Tapping on this technology, Atomwise is able to pursue discovery with higher precision and accuracy.
Medical Diagnosis
Artificial intelligence has far-reaching applications for making diagnosis. Even though looking up Google for symptoms yield cancerous results in the present day and age, practitioners can still be optimistic about this. In fact, AI was able to identify 17 diseases from a breath test with over 80% accuracy.
Radiology algorithms also helped IBM Watson to achieve higher accuracy when providing lung cancer diagnosis. This is a very specific medical condition and there is still a long way before we can train machines with the vast array of possible health diagnosis. However, University of Rochester managed to work around this limitation to study CT exams for suspected condition and hand it off to radiologists who will review the results with priority. Instead of leaving the judgement call to AI, AI was used to pre-process the scan results so radiologists can prioritize and detect time-sensitive conditions first.
The Danger
Artificial intelligence has far-reaching applications for making diagnosis. Even though looking up Google for symptoms yield cancerous results in the present day and age, practitioners can still be optimistic about this. In fact, AI was able to identify 17 diseases from a breath test with over 80% accuracy.
Radiology algorithms also helped IBM Watson to achieve higher accuracy when providing lung cancer diagnosis. This is a very specific medical condition and there is still a long way before we can train machines with the vast array of possible health diagnosis. However, University of Rochester managed to work around this limitation to study CT exams for suspected condition and hand it off to radiologists who will review the results with priority. Instead of leaving the judgement call to AI, AI was used to pre-process the scan results so radiologists can prioritize and detect time-sensitive conditions first.
Patient-facing Improvements
AI can have direct interaction with patients. Chatbots is a widely accepted application for healthcare providers to address patient’s query instantaneously. Chatbots are able to resolve frequently asked questions and assist in administrative processes like appointment scheduling. These enables the medical team to dedicate resources to more pressing matters. AI is also able to generate personalized healthcare advice and interventions for healthcare providers. The advice can be disseminated to patients via different channels at appropriate intervals, thereby improving patient’s health.
Specialized businesses like DoctorAnywhere also tapped on technology advancements to provide an integrated healthcare solution. Through the app, common clinical visits can be conducted at the comfort of one’s home- users can receive clinic/ video consultations with prescriptions dispensed using speedy medication delivery services.
A recent breakthrough is the closed-loop insulin pump system. The system samples glucose levels and automatically adjust the insulin delivery rate for the diabetic patient accordingly. This technology is being reverse engineered to create a low-cost alternative to benefit more diabetic patients.
Healthcare providers are increasingly integrated with smart home systems to support the ageing population living in their own homes. Assisted living tools are also developed for patients with cognitive impairments such as dementia. Other technologies such as intelligent walkers and wheelchairs are also developed in anticipation for the ageing population.
3D printing technologies have brought transformation to the healthcare industry in many ways including creating personalized prosthetics, and 3D-printed skin to help burn patients achieve better aesthetics recovery. Printing of pills are also being tested and showing promising results for the pharmaceutical industry. One pill can be taken with different release time for each drug so you don’t have to set alarms for different medication at different hours.
Social Implementations
Mobile phone data are used to model spread of epidemics. In 2016, Malaysia predicted hotspots for the dengue outbreak with an accuracy of 86%. These are done using information such as the geography, whether, symptoms of dengue cases and documentation of preventive actions taken. AI is used to model spread of infectious and non-communicable diseases by recommending the right allocation of resources and preventive measures.
The Ideal State of Technology in Healthcare
Seeing the present and more forward measures applied to healthcare and its related fields, here are a few more things the industry is eagerly anticipating:
Wearable Technology
Track user’s health by collecting and storing health data. These data can analyze to provide healthcare providers with a ground to make faster and more accurate predictions and diagnosis for patients. These will come in handy when “spotting connections in implications that are not normally obvious” – Dr Byron Scott, Deputy Health Officer, IBM Watson Health. These data will be useful in early-detection of diseases, monitor progress, and outpatient observations.
3D Printing
More development can be looked forward to as scientists work towards being able to print pills which are self-regulated. Stem cells are also harvested to produce more complex organs beyond the skin to benefit the patients who are waiting for suitable transplants.
Drone Technology
In January 2019, drones are being used to deliver coffee in Australia. Fast forward a few months, May 2019, a drone delivered a kidney to a transplant patient. Technology is evolving quickly – it won’t be long until drones take care of organ and blood transportation. Healthcare professionals are also looking forward to implementing similar technology to provide timely medical supplies to the right place, no matter how remote the location might be. Think: your wearable technology detects a heart attack and the pill is delivered to you via a drone automatically.
Related Resources
- Vogel, L., C.M.A.J. (2017). Plan needed to capitalize on robots, AI in health care: CMAJ CMAJ. Canadian Medical Association.Journal, 189(8), E329-E330.
- https://hbr.org/2019/06/what-ai-app-stores-will-mean-for-radiology
- https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002689
- https://www.tevapharm.com/featuredstories/?itemid=%7B7E74A827-65B8-4A6E-B087-5A2300DF52ED%7D
- https://www.rdmag.com/article/2016/10/ibm-teva-use-ai-drug-repurposing-program
- https://www.atomwise.com/our-technology/
- https://techcrunch.com/2018/04/05/bioprinted-organs-skin-and-drugs-how-3d-printing-is-revolutionizing-healthcare-as-we-know-it/
- https://www.engadget.com/2019/05/01/drone-delivers-organ-for-first-time/
