Reducing Overtreatment of Indeterminate Thyroid Nodules Using a Multimodal Deep Learning Model
This work addresses overtreatment in patients with indeterminate thyroid nodules by potentially reducing unnecessary benign resections, though it is incremental as it builds on existing molecular testing with added imaging.
The study tackled the problem of high false positives in molecular testing for indeterminate thyroid nodules by developing a multimodal deep learning model combining ultrasound images and molecular profiles, which matched sensitivity (0.946) and improved positive predictive value from 0.448 to 0.477, reducing false positives.
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.