Purpose: Thyroid nodules are common, and ultrasound-based risk stratification using ACR's TIRADS classification is a key step in predicting nodule pathology. Determining thyroid nodule contours is necessary for the calculation of TIRADS scores and can also be used in the development of machine learning nodule diagnosis systems. This paper presents the development, validation, and multi-institutional independent testing of a machine learning system for the automatic segmentation of thyroid nodules on ultrasound.

Methods: The datasets, containing a total of 1595 thyroid ultrasound images from 520 patients with thyroid nodules, were retrospectively collected under IRB approval from University of Chicago Medicine (UCM) and Weill Cornell Medical Center (WCMC). Nodules were manually contoured by a team of UCM and WCMC physicians for ground truth. An AttU-Net, a U-Net architecture with additional attention weighting functions, was trained for the segmentations. The algorithm was validated through fivefold cross-validation by nodule and was tested on two independent test sets: one from UCM and one from WCMC. Dice similarity coefficient (DSC) and percent Hausdorff distance (%HD), Hausdorff distance reported as a percent of the nodule's effective diameter, served as the performance metrics.

Results: On multi-institutional independent testing, the AttU-Net yielded average DSCs (std. deviation) of 0.915 (0.04) and 0.922 (0.03) and %HDs (std. deviation) of 12.9% (4.6) and 13.4% (6.3) on the UCM and WCMC test sets, respectively. Similarity testing showed the algorithm's performance on the two institutional test sets was equivalent up to margins of DSC 0.013 and %HD 1.73%.

Conclusions: This work presents a robust automatic thyroid nodule segmentation algorithm that could be implemented for risk stratification systems. Future work is merited to incorporate this segmentation method within an automatic thyroid classification system.

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11548-024-03294-wDOI Listing

Publication Analysis

Top Keywords

thyroid nodules
16
ucm wcmc
12
test sets
12
thyroid
8
segmentation thyroid
8
risk stratification
8
thyroid nodule
8
machine learning
8
multi-institutional independent
8
independent testing
8

Similar Publications

Introduction: Until now, the thyroid cancer case number has increased, and it is not entirely possible to attribute this continuous growth to more meticulous thyroid nodule selection and more accurate diagnostic techniques. While there is currently no conclusive evidence linking dietary factors to thyroid cancer, certain dietary patterns seem to have an impact on the development of the disease. There are interesting connections among diet, environment, metabolism and thyroid carcinogenesis; a deeper comprehension of the underlying mechanisms should help the identification of modifiable risk factors for thyroid cancer.

View Article and Find Full Text PDF

Thyroid surgery under nerve auto-fluorescence & artificial intelligence tissue identification software guidance.

Langenbecks Arch Surg

January 2025

Department of General Surgery, Sanatorio Otamendi & Miroli (Otamendi & Miroli Hospital), University of Buenos Aires, Buenos Aires, Argentina.

Thyroid cancer is a common malignancy that requires comprehensive clinical evaluation prior to adequate surgical management. Over the last three decades thyroid surgery has tripled and is considered one of the most commonly performed procedures in general surgery. These procedures are associated with potential postoperative complications with significant deterioration in the patient's quality of life.

View Article and Find Full Text PDF

Thyroid surgery in pediatric age: a ten-year experience at a single center and literature review.

J Pediatr Endocrinol Metab

January 2025

Department of Otolaryngology, Pendik Training and Research Hospital, Marmara University, Istanbul, Türkiye.

Objectives: Surgery interventions for thyroid disorders are rare in pediatric population. This study aims to present our institution's 10-year experience regarding the surgical treatment and outcomes of thyroid pathologies in children and review the literature.

Methods: All pediatric patients who underwent thyroid surgery at our institution from April 2013 to October 2023 were retrospectively reviewed.

View Article and Find Full Text PDF

Objective Ultrasound is the predominant modality in medical practice for evaluating thyroid nodules. Currently, diagnosis is typically based on textural information. This study aims to develop an automated texture classification approach to aid physicians in interpreting ultrasound images of thyroid nodules.

View Article and Find Full Text PDF

Purpose: Thyroid nodules are common, and ultrasound-based risk stratification using ACR's TIRADS classification is a key step in predicting nodule pathology. Determining thyroid nodule contours is necessary for the calculation of TIRADS scores and can also be used in the development of machine learning nodule diagnosis systems. This paper presents the development, validation, and multi-institutional independent testing of a machine learning system for the automatic segmentation of thyroid nodules on ultrasound.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!