Want to keep learning?

This content is taken from the Taipei Medical University's online course, AI and Big Data in Global Health Improvement. Join the course to learn more.

Skip to 0 minutes and 14 seconds So the idea that I had for legends is that it should be similar to the Large Hadron Collider.

Skip to 0 minutes and 24 seconds So the Large Hadron Collider is a project where a lot of particles physicists came together and decided and, instead of each one of them doing a a small bit of research on themselves, together they would build the machine that would generate the evidence. And I hope in legend we can do the same where we can all get together and start working on this evidence generating machine. And so that evidence generating machine, we basically hold a lot of research questions that we want answers. So I was talking about these depression research questions. But we’re gonna add hybrid, we’re adding hypertension as a next area of questions that will answer. We’re working on improving new methods.

Skip to 1 minute and 5 seconds We’re working on expanding the number of databases that are involved there. And I hope that we can include the Taiwanese database at one point as well. And that would generates a large evidence base. Of course from that evidence base which I hope, we can make this thing publicly available, there are some snafus there. We don’t want people to do data dredging but anyway, well, assuming that it will be publicly available, of course, you can write lots of papers on it. We can build a web app like I just gave you the URL for it. But I’m also hoping that we get something like The Weather Channel. So The Weather Channel is this, a channel in the U.S. It’s a company.

Skip to 1 minute and 46 seconds It doesn’t actually predict the weather or collect any data. The National Weather Service does that. It collects all the data. It runs all the fancy model and then it puts all that information in a evidence base. And then the Weather Channel, all the Weather Channel does, is it picks up that database and then presents it in a way that is usable to the end-user. And so we hope actually the people will start doing that because we know that it doesn’t, Dr. Westbrook has already showed just having the information somewhere is not enough. We need to make sure that somehow it reaches the doctor and the patients. So closing remark. Current published evidence from observational research is unreliable.

Skip to 2 minutes and 27 seconds Because of study bias publication bias and P-hacking and we can address this by, well, using more advanced method by measuring bias and calibrating effect estimates and by answering many questions at one, and publishing that as a unit. And legend will apply all of these principles in practices. And with that, at the end, thank you very much.

A Framework for Evidence Base Network

In this video, Dr. Martijn Schuemie proposes the future scope of using a conceptual framework similar to the particle physicists project, Large Hadron Collider. In this framework, we will have an evidence-generating machine that can answer many questions by connecting with many databases.

He concludes the speech in this video. In order to avoid study bias, publication bias, and P-hacking, he mentions using more advanced methods by measuring bias and calibrating effect estimates.

Reflect upon what you have viewed on Dr. Martijn Schuemie’s speech. Does anything impress you? You are free to leave your comment to share with other learners.

Share this video:

This video is from the free online course:

AI and Big Data in Global Health Improvement

Taipei Medical University

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join: