At SIIM 2026, LLMs demonstrated high accuracy on radiology-based numerical tasks, particularly in extraction and judgment tests.
Key Details
- 1LLMs evaluated included Llama 3.1 8B, DeepSeek R1-distilled Llama 8B, OpenAI o1-mini, and OpenAI GPT 5-mini.
- 2Tasks tested involved extraction and judgment from DEXA, ultrasound, CT, and PET radiology reports.
- 3Most models, except Llama, achieved over 95% accuracy on extraction tasks; Llama ranged from 86% to 98.7%.
- 4GPT 5-mini achieved highest minimum accuracy (judgment tasks: 91.7%) among tested models.
- 5o1-mini and GPT 5-mini reached perfect accuracy in detecting osteoporosis and made no mathematical errors.
- 6Answer-only output formats reduced accuracy for Llama and DeepSeek, but not OpenAI models.
Why It Matters
Reliable numerical extraction and judgment by LLMs could streamline radiology workflows, increasing efficiency for repetitive data extraction. However, caution remains due to ongoing risks of non-mathematical and medical knowledge-based errors.
Source
AuntMinnie
Related News
•AuntMinnie
Radiologists Struggle to Spot AI-Generated Radiology Images
Radiologists correctly identify AI-generated images 75% of the time, with CT and MRI images being particularly challenging to spot.
•AuntMinnie
Radiology Leads FDA AI Device Approvals Over Three Decades
Radiology accounts for 76% of all FDA-cleared AI/ML-enabled medical devices as of the end of 2025.
•Radiology Business
Automation Bias: How AI Can Compromise Radiologist Accuracy
AI decision support can induce automation bias, leading radiologists to accept incorrect interpretations and potentially reduce their diagnostic accuracy.