# A study on the diagnostic value of artificial intelligence combined with a contrast-enhanced ultrasound scoring system in partially cystic thyroid carcinoma

## TL;DR

This study shows that combining AI with a contrast-enhanced ultrasound scoring system improves thyroid nodule diagnosis accuracy across different experience levels.

## Contribution

A new CEUS scoring system combined with AI improves diagnostic accuracy for partially cystic thyroid nodules.

## Key findings

- A CEUS scoring system with 8 features was developed to differentiate benign and malignant thyroid nodules.
- AI combined with CEUS achieved high diagnostic accuracy (AUC=0.985 for senior and 0.967 for junior physicians).
- AI assistance significantly improved junior physicians' diagnostic performance for PCTNs.

## Abstract

The aim of this study was to investigate the diagnostic value of the contrast-enhanced ultrasound (CEUS) scoring system, artificial intelligence (AI) and the American College of Radiology Thyroid Imaging and Reporting Data System when used by sonographers of different seniority levels individually and in combination for the diagnosis of partial cystic thyroid nodules (PCTNs).

A retrospective analysis of conventional ultrasound and CEUS images of enrolled patients was performed, and a CEUS scoring system was established. The sensitivity, specificity, and area under the curve (AUC) of CEUS and AI individually and in combination for diagnosis were compared among sonographers with different seniority levels.

A total of 166 nodules (83 benign and 83 malignant) from 152 patients with PCTNs were analyzed in this study. Nine CEUS features of PCTNs were observed and summarized; eight of these features differed between the two groups (all p < 0.05) and were included in the CEUS scoring system. CEUS and AI used by junior and senior physicians effectively diagnosed PCTNs. AI improved the diagnostic efficacy of junior physicians. AI assistance combined with CEUS had the best diagnostic efficacy, with an AUC=0.985 for senior physicians and an AUC=0.967 for junior physicians, with no significant difference (P>0.05).

The CEUS scoring system established in this study has high diagnostic value for PCTNs. The use of CEUS and AI can improve the diagnostic accuracy of sonographers and improve the prognosis of PCTN patients.