Had the incredible opportunity to present work w @cshahrad at #RSNA25!
We evaluated the real-world performance of an automated ASPECTS scoring tool (RapidAI, iSchemaView) on one year of stroke alert NCCTs at Cleveland Clinic and found that AI routinely overcalls ischemia, especially on LVO negative cases which, while outside the validated use case for many AI ASPECTS tools, is far more common than LVO positive cases.
New @AJR_Radiology Accepted Manuscript:
"Beyond the AJR: Robust Ability of Artificial Intelligence to Detect Race Underscores the Need for Inclusivity and Transparency"
By Drs Shah & @howardpchen@ClevelandClinic
https://t.co/bfoZXigMJh
Inclusivity during development, data transparency, and performance monitoring for bias are key to ensure fairness in the future of AI.
https://t.co/yghS3hlcjZ
Using a smartphone app, investigators recruited 419,297 participants to be monitored for irregular pulses. Patterns suggesting atrial fibrillation were detected in 2161 participants who then received ECG monitoring devices. Read the Apple Heart Study. https://t.co/QhXpHzqcVL
Marty Pomper's Pendergrass Lecture: only 1/3 of cancer risk is environmental, 2/3 bad luck. No matter how much kale you eat, #radiology and precision imaging adds value thru early detection. @PennRadiology
What a perfect conclusion of the #datasceince session in #OHBM2018. Hopefully in the coming years we will see a gradual transition to more data driven analyses and in suit, also greater #reproducibility ...