Learning to Measure and Diagnose: A Clinical-Dictionary Radiomic Bottleneck Model for Thyroid Nodule Classification

Guibin Zhang; Weier Luo; Lei Cao

cs.CVOct 21, 2026

Learning to Measure and Diagnose: A Clinical-Dictionary Radiomic Bottleneck Model for Thyroid Nodule Classification

Authors: Guibin Zhang, Weier Luo, Lei Cao

Organizations: School of Biomedical Engineering, Southern Medical University, Guangzhou, China

Abstract

Thyroid ultrasound supports the assessment of nodule malignancy, but classification predictions often lack evidence linked to familiar clinical descriptions. Radiomics provides explicitly defined measurements that can support such explanations through a clinical dictionary. However, using these measurements as interpretable evidence requires reliable estimation from ultrasound images and preservation of measurement fidelity when optimizing classification performance. We propose a clinical-dictionary radiomic bottleneck model for thyroid nodule classification. Within this model, Family-Aware Radiomic Learning (FARL) estimates measurements with balanced supervision across feature families. Fidelity-Guided Bottleneck Modeling (FGBM) constructs a fixed bottleneck, followed by joint optimization of measurement estimation and classification under fidelity guidance. Decision evidence decomposition then traces feature contributions through a linear classifier and organizes them by clinical dictionary categories. Experiments on ThyroidXL and TN3K show competitive classification performance, while case studies illustrate feature contributions for benign and malignant nodules. These findings indicate interpretable classification with inspectable quantitative evidence informed by clinical knowledge. Code is available online.

Figures & tables

Figure 1: The proposed model is trained in two stages: Family-Aware Radiomic Learning (FARL, Stage 1) and Fidelity-Guided Bottleneck Modeling (FGBM, Stage 2). FARL learns per-image radiomic features with equally weighted family losses using offline targets from image–mask pairs, as shown in (a). Predictions are averaged per case and passed through a fixed dictionary bottleneck constructed through the reliability- and redundancy-based selection shown in (b). FGBM jointly optimizes the encoder and linear classifier with a fidelity loss anchored to frozen FARL predictions. Classification uses only bottleneck features, and inference requires ultrasound images alone.

Radiomics references
Method	ThyroidXL		TN3K
Method	AUC	BACC	AUC	BACC
TU1.0 (Eur. J. Radiol., 2026)	$\mathbf{99.21}$ $\pm 0.03$	$91.70$ $\pm 0.18$	$63.33$ $\pm 1.67$	$54.04$ $\pm 0.52$
Oracle radiomics (Cancers, 2024)	$98.90$ $\pm 0.00$	$\mathbf{94.72}$ $\pm 0.00$	$68.67$ $\pm 0.00$	$63.07$ $\pm 0.00$
Seg-first radiomics (JRRAS, 2025)	$87.41$ $\pm 0.05$	$79.68$ $\pm 0.50$	$\mathbf{69.84}$ $\pm 0.12$	$\mathbf{65.32}$ $\pm 0.13$
Black-box references

Table 1: Comparison results (mean

\pm

SD, %). Bold indicates the highest mean per column within each block.

Configuration	ThyroidXL		TN3K
	AUC	$\bar{\rho}$	AUC	$\bar{\rho}$
Full model	$\mathbf{94.77}$ $\pm 0.14$	$\mathbf{91.34}$ $\pm 0.08$	$\mathbf{79.49}$ $\pm 1.62$	$84.19$ $\pm 0.55$
w/o family balancing	$94.05$ $\pm 0.33$	$91.00$ $\pm 0.18$	$78.82$ $\pm 0.55$	$84.66$ $\pm 0.19$
w/o joint optimization	$92.20$ $\pm 0.64$	$91.22$ $\pm 0.19$	$73.24$ $\pm 0.09$	$\mathbf{84.90}$ $\pm 0.18$
w/o fidelity loss	$93.90$ $\pm 0.85$	$90.42$ $\pm 0.50$	$79.26$ $\pm 1.75$	$83.06$ $\pm 0.65$

Table 2: Ablation results (mean

\pm

SD; AUC in %,

\bar{\rho}\times 100

). Bold indicates the highest mean per column.

Figure 2: Feature contributions for representative TN3K cases. (a, c) Ultrasound images of the benign and malignant nodules. (b, d) Corresponding contributions organized by clinical dictionary categories. Sector percentages indicate the relative magnitudes of category contributions $C_{k}$ . Radial bands show cumulative absolute feature contributions within each category, with the dominant feature labeled.

Thyroid Nodule Detection

See all 9 papers

Radiomics-Based Thyroid Nodule Classification with Label-Conditional Conformal Prediction
Oct 9, 2026Laura Testa, Pierpaolo Brutti, Umberto Ferraro PetrilloThyroid Nodule Detection Medical Image Classification
Department of Statistical Sciences, Sapienza University of Rome, Italy.
CMCNet: Aligning Ultrasound Image Embeddings with Textual TI-RADS Representations for Fine-Grained Thyroid Classification
Aug 14, 2026Bingxin Yu, Xueli Wang, Jerry Zhou +6Thyroid Nodule Detection Ultrasound Image Classification
China-Japan Union Hospital of Jilin University, Changchun, 130012, P.R. China. · College of Computer Science and Technology, Jilin University, Changchun, 130012, P.R. China. · Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun, 130012, P.R. China. +3
Representation-Level Adversarial Regularization for Clinically Aligned Multitask Thyroid Ultrasound Assessment
Mar 22, 2026Dina Salama, Mohamed Mahmoud, Nourhan Bayasi +2Multi-Task Learning Clinical Risk Prediction
University of British Columbia, Vancouver, Canada · University of Calgary, Calgary, Canada

Oct 9, 2026J/K move · Enter open · S save

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

Laura Testa, Pierpaolo Brutti, Umberto Ferraro Petrillo

Department of Statistical Sciences, Sapienza University of Rome, Italy.

Thyroid Nodule Detection Medical Image Classification Radiomics

Open paper AskPDF arXiv

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. 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. 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. Experiments are conducted on TN5000, a publicly available ultrasound dataset of thyroid nodules. After Spearman correlation filtering, 36 features are retained and the GAM achieves an AUC of

0.89

, a sensitivity of

0.83

, and a specificity of

0.82

on the validation set. 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.

Abstract

Thyroid Nodule Detection

Explore similar work

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

CMCNet: Aligning Ultrasound Image Embeddings with Textual TI-RADS Representations for Fine-Grained Thyroid Classification

Representation-Level Adversarial Regularization for Clinically Aligned Multitask Thyroid Ultrasound Assessment