Who wrote this? "ChatGPT may have inherent biases due to its training data, which could lead to discriminatory treatment recommendations. If healthcare professionals fail to recognize and mitigate these biases, it could result in unequal or harmful care for patients." #RadAIChat

A 59-year-old woman with right lower quadrant pain via AuntMinnie.com auntminnie.com/index.aspx?sec…

Now we’re on our way to #SIIM23 Then starting radiology residency at UF Department of Radiology in two weeks! Let’s go!!!

the entire @EmoryRadiology squad @ #SIIM23 Judy Gichoya HITI Lab Patricia Balthazar, MD, MPH, CIIP John T Moon Hanssen Li, MD Nabile Safdar Marta Heilbrun MD MS Brandon Price

"Your Chest-ray indicates you may have diabetes" nature.com/articles/s4146… The latest unexpected outgrowth of deep learning #AI Nature Communications Ayis Pyrros, MD and colleagues

Here is the video summary of our T2D paper. youtu.be/OBKFKCSJEtU

The first publication of my PhD and my first 10-rated review! Thrilled to share our NeurIPS Conference'23 Oral finding large differences in frequency tuning between humans and neural networks, which strongly explain shape bias and robustness! arxiv.org/pdf/2309.13190… 🧵(1/4)

Compute is all you need. For a given amount of compute, ViT and ConvNets perform the same. Quote from this DeepMind article: "Although the success of ViTs in computer vision is extremely impressive, in our view there is no strong evidence to suggest that pre-trained ViTs

Curious about understanding uncertainty and trusting AI predictions? Don't miss the upcoming webinar with Brad Erickson, Dr. Richard Bruce, Shahriar Faghani MD and me discussing the topic. Link to the webinar: ecs.page.link/ouEMw #AI #UncertaintyQunatification #deeplearning

I will be speaking at RSNA 2023! • Harnessing Deep Learning to Identify Type 2 Diabetes from Frontal Chest Radiographs: A Prospective Study with External Validation •Presentation ID: S2-SSCA01-6 •Date & Time: November 26, 2023, 10:30 AM - 11:30 AM •Location: Room N228

My fav night time host and show #JohnOliver Last Week Tonight touched on organ donations and some of the issues surrounding them. Perfect timing as we just published on disparities that exist in organ transplant and transplant interventions! Janice Newsome MD FSIR Judy Gichoya #transplant

Radiologists are increasingly overworked, spending significant time dictating reports, often transferring text literally. Despite solutions like DICOM SR, the IT investment remains substantial. Can LLMs help? Excited to launch: 🔗 github.com/apyrros/auto_r… #radiology

"The #LLM Will See You Now": Performance of #ChatGPT on the Brazilian Radiology and Diagnostic Imaging and Mammography Board Examinations doi.org/10.1148/ryai.2… @haritrivedimd Judy Gichoya @emoryradiology #GPT4 #ML #MachineLearning

Im excited for my first #ISMRM2024 oral presentation "Neural Shape Models Meaningfully Localize Features Relevant to Osteoarthritis Disease" We're also sharing our medRxiv paper and announcing our ShapeMed-Knee Dataset that we hope others will leverage to advance the field 🧵