Table 4 Results from the post-curriculum survey consisting of 22 questions
Post-curriculum survey (n = 12 participants) | ||
|---|---|---|
What is your function at work? | Radiology resident | 11 (91.7%) |
Radiologist | 1 (8.3%) | |
What year of residency or experience are you at? | 3 | 5 (41.7%) |
4 | 3 (25%) | |
5 | 3 (25%) | |
10 | 1 (8.3%) | |
How confident were you about your knowledge and understanding of AI-based approaches in radiology before the AI course? (Scales 1–10 with 1 = not confident at all and 10 = very confident) | 1 | 2 (16.7%) |
2 | 1 (8.3%) | |
3 | 4 (33.3%) | |
4 | 3 (25%) | |
5 | 1 (8.3%) | |
6 | 1 (8.3%) | |
7 | 0 (0%) | |
8 | 0 (0%) | |
9 | 0 (0%) | |
10 | 0 (0%) | |
The course information increased my knowledge and skills about AI | Strongly disagree | 0 (0%) |
Disagree | 1 (8.3%) | |
Indifferent | 1 (8.3%) | |
Agree | 9 (75%) | |
Strongly agree | 1 (8.3%) | |
The course gave me more confidence on how to evaluate new AI projects | Strongly disagree | 0 (0%) |
Disagree | 0 (0%) | |
Indifferent | 7 (58.3%) | |
Agree | 4 (33.3%) | |
Strongly agree | 1 (8.3%) | |
After the course, I understand more about the shortcomings and strengths of AI | Strongly disagree | 0 (0%) |
Disagree | 0 (0%) | |
Indifferent | 2 (16.7%) | |
Agree | 9 (75%) | |
Strongly agree | 1 (8.3%) | |
The course addresses topics that are applicable to my work as a radiologist | Strongly disagree | 0 (0%) |
Disagree | 1 (8.3%) | |
Indifferent | 2 (16.7%) | |
Agree | 7 (58.3%) | |
Strongly agree | 2 (16.7%) | |
Which part of the course was most useful to you (multiple options are possible)? | Fundamentals of AI | 8 (66.7%) |
Hands-on laboratory sessions | 2 (16.7%) | |
Group discussions about AI | 8 (66.7%) | |
Presentation from a representative of a commercially available vendor | 4 (33.3%) | |
Which topics did you find most interesting and/or useful? | Which topics did you find most interesting and/or useful? I thought the talks and discussions about how these things come to life were very insightful. Also, our speaker was great! The fundamentals of course were important to get a grasp of the general concepts. The hands-on laboratory sessions were not something I could now implement in daily research practice, but I suppose that was not necessarily the goal. The commercial vendor was a great addition because it shows real clinical application of AI and its potential, as well as its downsides/pitfalls. It did not feel like a sales pitch The fundamentals are not the most interesting but definitely extremely important. Self-practicing with a model is very insightful and would be great to do that more independent and compare outcomes of individuals afterwards. Practical application by vendor was great, since it was not about the product but how work will change through new techniques Fundamentals of AI | |
The combination of theory and hands-on was useful | Strongly disagree | 2 (16.7%) |
Disagree | 1 (8.3%) | |
Indifferent | 4 (33.3%) | |
Agree | 3 (25%) | |
Strongly agree | 2 (16.7%) | |
The balance between theory, clinical application, and hands-on was well balanced | Strongly disagree | 2 (16.7%) |
Disagree | 1 (8.3%) | |
Indifferent | 2 (16.7%) | |
Agree | 7 (58.3%) | |
Strongly agree | 0 (0%) | |
What did you think about the group size (12 people)? | Good | 10 (83.3%) |
Larger groups are more effective | 1 (8.3%) | |
12–15 people would be ideal | 1 (8.3%) | |
What did you think about the length of the course (3 days)? | Good | 7 (58.3%) |
Bit long | 3 (25%) | |
2 days would be more efficient and is possible | 2 (16.7%) | |
The course was helpful in my progress towards my degree | Strongly disagree | 0 (0%) |
Disagree | 1 (8.3%) | |
Indifferent | 6 (50%) | |
Agree | 4 (33.3%) | |
Strongly agree | 1 (8.3%) | |
The course is likely to influence my radiology practice in the future | Strongly disagree | 0 (0%) |
Disagree | 1 (8.3%) | |
Indifferent | 5 (41.7%) | |
Agree | 5 (41.7%) | |
Strongly agree | 1 (8.3%) | |
After completing this course, how confident are you about your knowledge and understanding of AI-based approaches in radiology? (Scales 1–10 with 1 = not confident at all and 10 = very confident) | 1 | 0 (0%) |
2 | 0 (0%) | |
3 | 0 (0%) | |
4 | 0 (0%) | |
5 | 1 (8.3%) | |
6 | 6 (50%) | |
7 | 3 (25%) | |
8 | 2 (16.7%) | |
9 | 0 (0%) | |
10 | 0 (0%) | |
I would highly recommend this course to my colleagues and/or future radiologists | Strongly disagree | 0 (0%) |
Disagree | 0 (0%) | |
Indifferent | 3 (25%) | |
Agree | 7 (58.3%) | |
Strongly agree | 2 (16.7%) | |
The course should be included in the radiology residency program | Strongly disagree | 0 (0%) |
Disagree | 1 (8.3%) | |
Indifferent | 2 (16.7%) | |
Agree | 6 (50%) | |
Strongly agree | 3 (25%) | |
All radiologists should follow this course to understand more about AI | Strongly disagree | 0 (0%) |
Disagree | 2 (16.7%) | |
Indifferent | 4 (33.3%) | |
Agree | 3 (25%) | |
Strongly agree | 3 (25%) | |
What would you recommend on how to improve AI knowledge in the radiology residency program (multiple options are possible)? | Lectures about AI from guest speakers | 8 (66.7%) |
Journal clubs with the radiology department | 1 (8.3%) | |
Following online courses offered by radiology associations | 4 (33.3%) | |
Interdisciplinary conferences about implementing AI in clinical practice | 5 (41.7%) | |
Demos or simulations by AI companies | 7 (58.3%) | |
AI course with a small group | 9 (75%) | |
AI discussions with residents and/or radiologists | 5 (41.7%) | |
Discussions about ethics, financial, and insurance aspects of AI | 4 (33.3%) | |
Regional organized education | 1 (8.3%) | |
Are there any topics of discussions you missed during the course and you would like to discuss? | Views from our hospital on the matter or the input of radiologists that work with AI: how did they select a program, what is their cost-effectiveness strategy, etc | |
Do you have tips or ideas on how to improve this AI course? | Bit more hands-on. Maybe some easy examples with a little coding, do not know if that is feasible Communication beforehand should be much better Research part was not very useful to me During the UMCG resident day (for all residents, not just radiology residents) earlier this year, I attended a workshop offered by DASH in the UMCG during which we got to try out Google Teachable Machine (https://teachablemachine.withgoogle.com). With several small image datasets, we got to dabble with AI at a very basic level. Maybe this could also be implemented during the hands-on laboratory sessions of this course, as it is a very simple yet understandable way of showing how it works. Maybe afterwards, the Orange hands-on part is a bit easier to understand The hands-on laboratory sessions could be set up with more group participation; now it is as a loose part | |