Comparison of an AI-assisted system (S-detect) and sonographers of different experience levels in diagnosing thyroid nodules: a retrospective study
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Artificial intelligence (AI)-assisted ultrasonography, S-Detect, shows superior diagnostic performance compared to junior sonographers for thyroid nodules. Its efficacy is comparable to experienced sonographers, improving thyroid cancer diagnosis.
Area Of Science
- Oncology
- Medical Imaging
- Artificial Intelligence
Background
- Thyroid cancer (TC) is the most common neck malignancy with early metastasis potential.
- Conventional ultrasound diagnosis relies on subjective interpretation, necessitating objective tools for improved efficiency.
- Artificial intelligence (AI) offers potential for enhancing diagnostic accuracy in medical imaging.
Purpose Of The Study
- To compare the diagnostic efficacy of AI-assisted ultrasonography (S-Detect) against sonographers with varying experience levels.
- To evaluate the performance of S-Detect in differentiating benign from malignant thyroid nodules.
- To assess the impact of AI on diagnostic accuracy and efficiency in thyroid nodule evaluation.
Main Methods
- A retrospective study analyzed 315 thyroid nodules from 237 patients with biopsy/surgical confirmation.
- Sonographers were categorized into junior and advanced experience groups.
- Diagnostic performance metrics including sensitivity, specificity, accuracy, kappa, Youden's Index, and AUC were compared between S-Detect and sonographers.
Main Results
- S-Detect significantly outperformed junior sonographers in sensitivity, specificity, accuracy, kappa, and AUC (P<0.05).
- In comparison to advanced sonographers, S-Detect demonstrated comparable sensitivity for malignant nodules but lower specificity for benign nodules.
- While advanced sonographers showed higher accuracy and consistency, overall performance between S-Detect and advanced sonographers was not significantly different (P > 0.05).
Conclusions
- AI-assisted ultrasonography (S-Detect) surpasses junior sonographers in thyroid nodule diagnosis.
- S-Detect exhibits comparable overall diagnostic performance to experienced sonographers.
- AI tools like S-Detect show promise in improving the objective assessment of thyroid nodules.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.