ML PhD at RWTH Aachen
Aachen, Germany
Joined June 2020
- Tweets125
- Following161
- Followers67
- Likes169
Tianyu Han retweeted
CLEAR: an auditable foundation model for radiology grounded in clinical concepts go.nature.com/3RPlTWI
Tianyu Han retweeted
📣Today, NIH joined federal partners in launching the Genesis Mission. The Bio Genesis Mission is NIH’s contribution to this effort.
This new initiative will help accelerate biomedical innovation and turn scientific discovery into meaningful patient impact.
Read the full story ➡️ bit.ly/4yGQUwP
ALT Three people in lab coats celebrate with a high-five. Text reads: "NIH Launches Biomedical Innovation Initiative. Introducing the Bio Genesis Mission, accelerating biomedical discovery with AI and strategic partnerships." NIH logos are visible at the bottom.
Excited to share our new @NatBME paper: "CLEAR — an auditable foundation model for radiology grounded in clinical concepts." CLEAR grounds its representations in clinical concepts, so you can actually audit why it decides what it decides. 🧵
🔗 [nature.com/articles/s41551-0…]
After 11 years in Germany, I'm thrilled to share that I've joined @PennMedicine as an Assistant Professor on the @PennAI2D team! Excited to meet new friends and collaborators.
I'm hiring postdocs in radiological image analysis & self-supervised learning—DM if interested! #hiring
If you're interested, check out my research profile here: scholar.google.com/citations….
Excited to connect and collaborate!
Just hit 1,000 citations on Google Scholar! Feeling incredibly grateful for the support, collaborations, and communities that made this milestone possible. #Research
scholar.google.com/citations…
Tianyu Han retweeted
I am very honored to receive this distinguished NIA award and give the annual lecture. It has been a long and enjoyable journey in aging research, with emphasis on AI in neuroimaging
🚀 New Research Alert! Our latest study, published online about a month ago in @CellRepMed , introduces DiffChest—a diffusion model for enhancing AI-driven radiology by tackling confounders. Here’s what makes it a game-changer. 🧵 Link doi.org/10.1016/j.xcrm.2024.…
🖼️ Visual Explanations: To boost clinician acceptance, DiffChest generates patient-specific visual explanations, aiding agreement in complex cases and supporting nuanced disease grading. (4/5)
🌍 Why it Matters: DiffChest's interpretability and accuracy are key to safer clinical AI. Published online last month, our findings aim to make AI diagnostics a trusted tool in healthcare. Read more here: doi.org/10.1016/j.xcrm.2024.… #MedAI #RadiologyAI (5/5)
Tianyu Han retweeted
How will large language models transform structured reporting in radiology and beyond? In our comprehensive review just published in @EurRadiology, we deeply dive into the past, present, and future of LLMs for radiology reporting. Read the article here: doi.org/10.1007/s00330-024-1…
Tianyu Han retweeted
Revealing a vulnerability in large language models (LLMs) used in medicine...
By altering just 1.1% of the model's weights, incorrect medical information can be introduced without affecting performance in other areas.
This finding highlights security concerns, emphasizing the need for strong safeguards and verification processes to ensure #LLMs remain trustworthy and reliable in healthcare settings.
nature.com/articles/s41746-0…
Tianyu Han retweeted
Author thread alert!
🚨 New Paper Alert! 🚨 We've discovered a major vulnerability in medical large language models (LLMs): they're highly susceptible to targeted misinformation attacks. This could have serious implications for healthcare AI! @DanielTruhn @jnkath
Full paper: doi.org/10.1038/s41746-024-0…
🚨 New Paper Alert! 🚨 We've discovered a major vulnerability in medical large language models (LLMs): they're highly susceptible to targeted misinformation attacks. This could have serious implications for healthcare AI! @DanielTruhn @jnkath
Full paper: doi.org/10.1038/s41746-024-0…
⚠️ This vulnerability poses a risk in scenarios where AI is used for critical healthcare decisions—doctors could unknowingly rely on false information, jeopardizing patient safety. We call for stronger verification mechanisms.
🛡️ Our findings underscore the need for robust safeguards, thorough audits, and stringent access control when using LLMs in medicine to ensure trustworthiness and reliability in healthcare AI.
📰 Read the full paper published in @npjDigitalMed : doi.org/10.1038/s41746-024-0…
Tianyu Han retweeted
Online now: Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining dlvr.it/TDd4MN