New research demonstrates that high BMI negatively impacts both human and AI performance in chest low-dose CT interpretation, highlighting dataset diversity concerns.
Key Details
- 1Higher BMI leads to increased image noise and reduced CT scan quality, especially in low-dose protocols.
- 2Many AI models lack diverse training data reflecting a broad range of BMIs.
- 3Study compared top 1.5% highest and lowest BMI patients using chest LDCT scans from the Lifelines cohort.
- 4AI and human readers both showed impacted nodule detection sensitivity and false positive rates at extreme BMIs.
- 5Evaluation was performed by two expert chest radiologists.
Why It Matters
This study exposes a critical blind spot in radiology AI—model performance can deteriorate for patients at BMI extremes due to suboptimal imaging and lack of representativeness in training datasets. Ensuring AI equity and reliability in clinical practice requires addressing these biases.
Source
Health Imaging
Related News
•AuntMinnie
Toronto Study: LLMs Must Cite Sources for Radiology Decision Support
University of Toronto researchers found that large language models (LLMs) such as DeepSeek V3 and GPT-4o offer promising support for radiology decision-making in pancreatic cancer when their recommendations cite guideline sources.
•AuntMinnie
AI Model Using Mammograms Enhances Five-Year Breast Cancer Risk Assessment
A new image-only AI model more accurately predicts five-year breast cancer risk than breast density alone, according to multinational research presented at RSNA 2025.
•Radiology Business
AI Model Uses CT Scans to Reveal Biomarker for Chronic Stress
Researchers developed an AI model to measure chronic stress using adrenal gland volume on routine CT scans.