Thoughts on Giving Oral Presentations

Content must be the core value of what you say, but the delivery and presentation often has the majority effect on people's perceptions. YOU are the presentation, not your slides -> Don't turn back to audience. Don't read slides. Try to get away from podium and walk around to talk with audience members. Avoid using oral presentations to transmit information / data / descriptions (e.g., lengthy description of methods or dense results tables, unless the novel method is itself the result). Use papers/writing for those details. Oral communication should be at the level of core ideas and storytelling. Just be able to answer the details if someone asks about them.

• How to avoid Death by PowerPoint - Provocative review on specific design techniques and guidelines for slide construction.

• Magic as Public Speaking - One of the best talks I've ever watched. Stage magician as an analogy on how to deliver a public presentation.

• Neil deGrasse Tyson's MasterClass - An incredibly well produced set of lessons on scientific communication and teaching to different audiences. This is valuable enough that I've paid for a lab subscription for people to watch.

• Designing Effective Presentations - Susan McConnell with extended discussion of designing slides for a scientific audience. Important foundational concepts, but just remember that your slides are NOT your presentation.

• Biomedical Informatics Training Resources - Resources collected by Stanford's Biomedical Informatics Training Program with tips on good presentations.

• Deliver Compelling Talks - Why, What, and How? - Jonathan's talk to motivate everyone to enhance their own presentation skills.

More general tips

• Arresting Opening: Immediately start with a compelling story or motivation. NOT with reading your name, affiliation, and title. Imagine watching a blockbuster movie. It doesn't start with a long scroll of the credits. It starts with an action scene to get you excited and engaged. Once you have them hooked, you can then breathe for agenda, structure, etc.

• If leave a last slide on screen, try to have a brief graphical abstract of your subject. Will remind people of what they wanted to ask about.

• Rehearse with video recordings so you can check timing (leave ~1/3 for Q&A) and practice eliminating visual and auditory clutter (distracting body language, utterances and non useful words).

• It's worth writing out a script in the notes section of your slides at least a single draft pass. It will let you spot weird transition points, so you're not stuck in a live setting trying to decide what to say. For the first and last 30 seconds, worth deliberately scripting and polishing to hook listeners and leave with something memorable or inspiring.

• Asked a question you don't know the answer to? It's okay to say "I don't know," but even better to redirect towards something you do know or something you can hypothesize. Try something like, That's a great question and suggestion. We haven't gotten to that phase of the work yet, but here's how I imagine tackling it... I'm not sure, that's somewhat of an open research question, but it does relate to XXX which makes me think of YYY... (politician's technique: 30 seconds later, people won't even remember what the question was. They'll just think that you said a bunch of stuff that sounded smart).

• Never ask: "Did I answer your question?" Only two possible results. One is you did answer the question, so it's just wasting air time. Second is you still didn't answer their question, so you're trapped in an even more awkward spot. As the speaker, you should dominate the space and just break eye contact with the person and move on to the next.

Preparing Poster Presentations

Scientific Poster Design Issues
I don't necessarily like the "solution" offered in the above video, but it does a pretty good job of pointing out the problems in typical poster design.
Less is more. A poster isn't meant to stand alone (don't expect anyone to read it). It's meant to be your "banner ad" to prompt people walking by to ask you about it, and occasionally a visual prop you can point to while talking to them.

Example Poster That I won a couple student research presentation awards with. Included in the Notes of the Poster is the "script" I used when presenting within 2-3 minutes. Note in particular how sparse the text is and that the emphasis is just on a couple figures I point to briefly in my presentation.

Conferences to Consider Submitting a Paper or Abstract to and their usual dates

Medical / Clinical Informatics

• AMIA Annual Symposium (Papers due March, conference November)
• AMIA Informatics Summits (Papers due August, conference March)
• AMIA Clinical Informatics (Papers due November, conference May)
• Pacific Symposium Biocomputing (Papers due August, conference January)
• Stanford Big Data in Precision Health (Internal Abstracts, conference May)

AI / Machine Learning

• NeurIPS Healthcare Workshop? (Extended) Abstracts due in September
• ICML (Papers due January, conference in July)
• Machine Learning for Healthcare (Papers due April, conference August)

General Medical

• ACP National Meeting (Abstracts due November, conference April) Scientific Session Submissions due May, two years in advance!?
• SGIM National Meeting (Abstracts due January, conference May)

Others

• Epic XGM or UGM???
• ACM, IEEE, ICHI, SMDM ???