Radiomics-Based Thyroid Nodule Classification with Label-Conditional Conformal Prediction

Key points

• Thyroid nodule malignancy assessment relies on ultrasound imaging, yet current risk stratification systems suffer from substantial interobserver variability, contributing to a high rate of unnecessary biopsies.1
• To complement point predictions with uncertainty estimates, Cross-Conformal Prediction with label-conditional guarantees is applied, flagging ambiguous cases for clinical review rather than forcing a binary decision.1
• Experiments are conducted on TN5000, a publicly available ultrasound dataset of thyroid nodules.1
• At a significance level $ε=0.09$, selected to ensure empirical coverage guarantees on both classes, 63% of nodules on the test set receive a certain (i.e., single-label) prediction, with conditional precision of $0.94$ for certain malignant and conditional Negative Predictive Value (NPV) of $0.73$ for certain benign cases, while the remaining 37% are referred to the clinician.1

Sources (1)

• [1]Radiomics-Based Thyroid Nodule Classification with Label-Conditional Conformal Prediction
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 11:01 AM

  We propose a radiomics-based framework combining quantitative feature extraction via PyRadiomics with a Generalized Additive Model (GAM) classifier, selected among six candidates for its favorable trade-off between discriminative performance and interpretability.

  Thyroid nodule malignancy assessment relies on ultrasound imaging, yet current risk stratification systems suffer from substantial interobserver variability, contributing to a high rate of unnecessary biopsies.

Extractive summary: sentences quoted from the sources.