Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Reducing Overtreatment of Indeterminate Thyroid Nodules Using a Multimodal Deep Learning Model

Shreeram Athreya, Andrew Melehy, Sujit Silas Armstrong Suthahar, Vedrana Ivezić, Ashwath Radhachandran, Vivek Sant, Chace Moleta, Henry Zheng, Maitraya Patel, Rinat Masamed, Corey W. Arnold, William Speier

TL;DR

This study tackles overtreatment of indeterminate thyroid nodules by integrating ultrasound imaging with molecular testing through an attention-based multiple instance learning framework. The multimodal AMIL model fuses US-derived features with MT outcomes to classify nodules as benign or malignant, aiming to preserve MT’s high sensitivity while reducing false positives. Across 333 cases, the ensemble approach achieved the highest AUROC (0.831) and improved PPV (0.477) compared to MT alone (0.448), suggesting fewer unnecessary surgeries. The method provides interpretable attention maps and presents a clinically meaningful step toward more precise, less invasive management of indeterminate nodules, pending prospective validation.

Abstract

Objective: Molecular testing (MT) classifies cytologically indeterminate thyroid nodules as benign or malignant with high sensitivity but low positive predictive value (PPV), only using molecular profiles, ignoring ultrasound (US) imaging and biopsy. We address this limitation by applying attention multiple instance learning (AMIL) to US images. Methods: We retrospectively reviewed 333 patients with indeterminate thyroid nodules at UCLA medical center (259 benign, 74 malignant). A multi-modal deep learning AMIL model was developed, combining US images and MT to classify the nodules as benign or malignant and enhance the malignancy risk stratification of MT. Results: The final AMIL model matched MT sensitivity (0.946) while significantly improving PPV (0.477 vs 0.448 for MT alone), indicating fewer false positives while maintaining high sensitivity. Conclusion: Our approach reduces false positives compared to MT while maintaining the same ability to identify positive cases, potentially reducing unnecessary benign thyroid resections in patients with indeterminate nodules.

Reducing Overtreatment of Indeterminate Thyroid Nodules Using a Multimodal Deep Learning Model

TL;DR

This study tackles overtreatment of indeterminate thyroid nodules by integrating ultrasound imaging with molecular testing through an attention-based multiple instance learning framework. The multimodal AMIL model fuses US-derived features with MT outcomes to classify nodules as benign or malignant, aiming to preserve MT’s high sensitivity while reducing false positives. Across 333 cases, the ensemble approach achieved the highest AUROC (0.831) and improved PPV (0.477) compared to MT alone (0.448), suggesting fewer unnecessary surgeries. The method provides interpretable attention maps and presents a clinically meaningful step toward more precise, less invasive management of indeterminate nodules, pending prospective validation.

Abstract

Objective: Molecular testing (MT) classifies cytologically indeterminate thyroid nodules as benign or malignant with high sensitivity but low positive predictive value (PPV), only using molecular profiles, ignoring ultrasound (US) imaging and biopsy. We address this limitation by applying attention multiple instance learning (AMIL) to US images. Methods: We retrospectively reviewed 333 patients with indeterminate thyroid nodules at UCLA medical center (259 benign, 74 malignant). A multi-modal deep learning AMIL model was developed, combining US images and MT to classify the nodules as benign or malignant and enhance the malignancy risk stratification of MT. Results: The final AMIL model matched MT sensitivity (0.946) while significantly improving PPV (0.477 vs 0.448 for MT alone), indicating fewer false positives while maintaining high sensitivity. Conclusion: Our approach reduces false positives compared to MT while maintaining the same ability to identify positive cases, potentially reducing unnecessary benign thyroid resections in patients with indeterminate nodules.
Paper Structure (12 sections, 2 equations, 3 figures, 2 tables)

This paper contains 12 sections, 2 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Dataset
  4. Attention Multiple Instance Learning
  5. Our Framework
  6. Experiments
  7. Results
  8. Conclusion
  9. Acknowledgements and Funding
  10. Author Contributions
  11. Conflict of Interest
  12. Data Availability

Figures (3)

  • Figure 1: Overview of the thyroid nodule diagnostic pipeline. Our method works as an additional step after molecular testing to combine molecular results with imaging to reduce unnecessary diagnostic surgeries.
  • Figure 2: An end-to-end overview of the multi-modal AMIL framework.
  • Figure 3: Results of the AMIL framework. (a) ROC curves comparing different model configurations, (b) and (c) Patch 256 attention maps of representative US scans. In (b), there is a nodule in the right lobe of the thyroid with the attention map including the majority of the nodule. The deep border has a notable rim of hyperechogenicity that is partially contained within the attention map. In (c), there is a large thyroid nodule in the center of the frame, containing a central solid component with hypoechoic peripheral areas. Notably, the attention map is located directly over an area containing isoechoic and hypoechoic nodule components at the periphery as well as the hyperechoic interface between the nodule and the deep surrounding tissue. BE: Bethesda scores, WF: Whole frame images.