The type of proposal may vary for each of the three conferences according to the submission guidelines. Participants are encouraged to submit full paper manuscript to Stage III.
Decisions will be sent to the primary author approximately 30 to 45 days after the deadline for each of the three phases of the AI Evaluation Showcase.
The SPC requests original work however preliminary results that have been expanded are welcome.
Yes. We would like to see the description of the evaluation of the system’s technical performance in Stage I, but the abstract should include an idea of what the rest of the evaluation will look like in the usability and health impact studies. A brief description of the anticipated Stages II and III can be in the Methods or Future Work sections). We will not hold you to do exactly what you anticipated you would do, but the Stage I submission should have a sense of what the research plan is for all three parts.
For this initiative we are interested in the study of tools that go the full length of development into usage. You are welcome to submit abstracts on studies that achieved results only in Stage I through the regular submission process.
The amount of data you have at each stage will influence the type of abstract you submit. The first Stage could be a podium abstract or manuscript, although a poster is acceptable as well. Stage III will most likely be a poster, which would include a recap of the first two stages.
No. There should be one PI or senior leader for all three studies, but the research teams and order of authors can be different for each of the three submissions.
We don’t recommend the resubmission of the same submission. New data on that original submission would make it a new submission.
We welcome submissions from any organization that is willing to evaluate its work. AI vendors are welcome to submit their abstracts to this peer-reviewed process.
Absolutely. Sharing what is not working is one of the most important things we can do to elevate the field.
Yes, we are looking at a broad definition here and we will consider any tools that mimic human intelligence, such as ML or DL.
You can decide what study results to submit for each meeting.
No. Give us what you’ve got! There isn’t necessarily any limit. You may not have a lot of data, in which case you would submit a poster. If you have bigger datasets you can submit a podium abstract or manuscript.
Submissions must be in line with the submission deadlines. We want to see submissions of things that are viable through Stage III.
Methodologic methods papers are reasonable to submit, but they should fit in with the stages of submission: 1. technical, 2. usability/workflow, 3. health impact.