AI For Healthcare was a course created by Health Education England and the University of Manchester, to provide a general overview of AI and how it can and is being used in the health sector. Anyone could access this course for a limited amount of time, although it was designed for healthcare workers in mind.
As someone who’s really interested in AI and machine learning (and a big fan of the Topol Review), I took the plunge and had a go. The course was incredibly useful, providing a great introduction to AI. It showed working examples of how it could be utilised, and the pros and cons of implementing new technologies.
Discussion was actively encouraged, and I chatted with wide variety of people working within the healthcare sector. There was the occasional quiz, but mostly people benefited from the rich conversations taking place in the comments sections.
The course was split into five weeks:
- Week 1: Motivating AI in Healthcare
- Week 2: What is Artificial Intelligence
- Week 3: Data in Healthcare
- Week 4: Making it Work
- Week 5: Supporting and Skilling the Workforce
The first week was a brief introduction to the course, and looked at the opportunities and challenges of working within the health sector; using technologies to assist with healthcare in an increasingly demanding setting. It was also an opportunity to introduce ourselves within the discussion, and how we believe our roles could utilise AI in the future. I mentioned monitoring library usage (seeing what resources/topics are popular) and targeted promotion, making resources more accessible and findable for users, more relevant current awareness updates and taking the edge out of literature searching.
We focused on ethical and social aspects of AI and machine learning, generating interesting discussion around if we would be comfortable with being provided personal information and news regarding our health by AI, and whether AI should be used by healthcare professionals to inform decision making. There was also debate on whether AI could essentially ‘replace’ certain services, such as GPs. The general consensus was that as the technology is designed to support, rather than replace services, that it is not capable or desirable for technology to replace human roles.
Further down the line, we looked at cases of AI in action with regards to identifying cancer in breast images. This was particularly topical as it had been recently reported in the news.
There was also an introduction to ‘team science’ theory, creating interdisciplinary teams to work together on projects. Experts from all kinds of different fields and backgrounds will be required for the development of AI in healthcare. Having a diverse range of professionals with different backgrounds, expertise and insights would be highly beneficial, both to reduce bias in software and to create something which can be used by a wide variety of people. I was keen to point out that LKS workers have great skills around Knowledge Management, accessibility and user-centred design, and that having LKS staff embedded into multidisciplinary teams would be an excellent use of our expertise.
We also looked at the challenges of AI; its implementation, management, and the need to educate and train staff on how to use it effectively. I believe this in particular is a golden opportunity for LKS staff; to educate, train, and advocate for the user, enabling them access to quality technology and providing them a safe space to learn and develop new skills.
All in all, the course was an excellent introduction. Being able to network with healthcare professionals was also very useful, as I was able to gage their thoughts and feelings about AI. The course tutors and mentors were fantastic, contributing to discussion and encouraging people to think outside the box. It was heartening also to see the support and interest from others in the roles of LKS staff, and how AI can be a useful tool in our libraries.
Below is a list of some resources which were recommended by the course:
- Part 1: Artificial Intelligence Defined
- The Emma Watch: A Wearable Device to Ease PD Tremors
- Parkinson’s diagnosis set to be sped up by Tencent’s AI
- How Medopad and Tencent are Helping People with Parkinson’s Disease
- Artificial intelligence apps, Parkinson’s and me
- AI Trends glossary
- How AI Can Unlock Data in CT and MRI Scans
- Learn from ML experts at Google
- A.I. Bias In Healthcare
- How much data is generated each day?
- Mapping opportunities for earlier detection of Bipolar Disorder – Linking big data to improve patient outcomes (MOBILISE)
- Creating the right framework to realise the benefits for patients and the NHS where data underpins innovation
- 5 Applications of Facial Recognition Technology
- Deep Learning and Computer Vision
- HEE digital literacy
Weston Area Health Library