-
Mashup Score: 5InVision Precision LVEF AI echo algorithm is now FDA cleared - 1 day(s) ago
Only AI-LVEF algorithm shown in blinded RCT to save time and improve precision
Categories: General Medicine News, Cardiologists1Tweet
-
Mashup Score: 2Clinical Validation of Medical AI - 5 month(s) ago
The story of software in healthcare is not straightforward or linear. While software is clearly eating the world, in healthcare there can be indigestion with poorly validated algorithms that actually do not improve care or efficiency. One such example, is the
Categories: General Medicine News, Cardiologists1Tweet-
https://t.co/wgtOsTOAE4
-
-
Mashup Score: 0How to choose research projects - 6 month(s) ago
Or AI projects in general
Categories: General Medicine News, Cardiologists1Tweet-
https://t.co/Z4Z531tFn1
-
-
Mashup Score: 0Cardiologists have an AUC of 0.81 with each other - 8 month(s) ago
Even worse R2 and other fun facts about clinical variability
Categories: Cardiologists1, Latest HeadlinesTweet-
Clinicians have so much intra observer variation that it masks the difference between methods. A preference for eyeball is recognizing there are limitations with any quantitative approach, but also incentivized be ease and lack of QC. https://t.co/GVhN9kX1Ii https://t.co/Xo28eBV3MR
-
-
Mashup Score: 4
This post is by Kai Christensen, a talented AI researcher, filmmaker, and artist on the team. Check him out at mkaic.blog or follow him on Twitter. I am an AI researcher, and I work at the CVAIR (CardioVascular AI Research) lab at Cedars-Sinai. I just finished a research project, and have
Categories: Cardiologists1, Latest HeadlinesTweet-
Guest post by @kayzayai with an approachable description of #Echofirst EchoCLIP and its implementation. https://t.co/GP3FwhgeWI https://t.co/eSAQNEB8OT
-
-
Mashup Score: 1A Rose by Any Other Name - 8 month(s) ago
The surprising nuances of diagnoses and what it means for training AI models
Categories: Cardiologists1, Latest HeadlinesTweet-
Potato, PoTATo. You might call it AF, I might call it AHRE, but why that's important for training AI algorithms. https://t.co/Sqq1aHitwz https://t.co/F9aMX2U27t
-
-
Mashup Score: 0Cardiologists have an AUC of 0.81 with each other - 9 month(s) ago
Even worse R2 and other fun facts about clinical variability
Categories: Cardiologists1, Latest HeadlinesTweet-
Please think of this next time you seen a sales pitch for an AI model with an AUC of 0.97 True in all forms of medical imaging. https://t.co/GVhN9kXzxQ #WeNeedClinicalTrials https://t.co/kADOX9NeIH
-
-
Mashup Score: 0How to choose research projects - 9 month(s) ago
Or AI projects in general
Categories: Cardiologists1, Latest HeadlinesTweet-
https://t.co/Z4Z531tFn1
-
-
Mashup Score: 0Non-Causal Inference - 9 month(s) ago
How colliders and confounders can introduce bias to AI models
Categories: Cardiologists1, Latest HeadlinesTweet-
This was written a few weeks ago, but seems timely given recent technologies for AI screening. I think it's an area we need clinical trials to validate efficacy. https://t.co/HtZeYTpOG3 Excellent coverage by @mariojoze at https://t.co/0if5Rd7C2l https://t.co/n1Sv7CIV4S
-
-
Mashup Score: 0Why there are so few AI clinical trials - 9 month(s) ago
Incentives, incentives, incentives
Categories: Cardiologists1, Latest HeadlinesTweet-
https://t.co/pknplT57s0
-
@InVision_AI More information here: https://t.co/QRnuCjeB3C