A study shows large language models can predict immunotherapy responses in liver cancer as accurately as experienced doctors.
Key Details
- 1Research led by Prof. Hai Li at Hefei Institutes systematically assessed LLMs' ability to predict liver cancer treatment response.
- 2LLMs tested: GPT-4, GPT-4o, Google Gemini, DeepSeek, and a hybrid Gemini-GPT model.
- 3Dataset included clinical and imaging data from 186 unresectable HCC patients.
- 4Gemini-GPT matched senior (15+ years) doctor accuracy and surpassed less-experienced clinicians in both speed and accuracy.
- 5Hybrid and logical strategies improved LLM performance and consistency, especially in identifying likely responders to therapy.
- 6Published in the Journal of Medical Systems on May 15, 2025.
Why It Matters
This work demonstrates the emerging capability of LLMs to support and enhance personalized cancer treatment decisions, showing they can complement or even match the expertise of experienced radiologists and oncologists in outcome prediction.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.