⚡ Quick Summary
This clinical trial explored the impact of the AI-based diagnostic system S-Detect on the ultrasound diagnosis of thyroid nodules among various ultrasound users. The findings revealed that while the mean diagnostic accuracy for S-Detect was 71.3%, the operator’s experience significantly influenced the diagnostic outcomes, with accuracy ranging from 40% to 100%.
🔍 Key Details
- 👥 Participants: 20 individuals, including medical students, novice physicians, and experienced physicians.
- 🧪 Cases: 5 patients with thyroid nodules (1 malignant, 4 benign).
- 📊 Diagnostic System: S-Detect for Thyroid.
- 📈 Accuracy Metrics: Mean diagnostic accuracy of 71.3%, biopsy recommendation accuracy of 69.8% before AI and 69.2% after AI.
🔑 Key Takeaways
- 🤖 AI Integration: S-Detect did not significantly improve diagnostic accuracy among novice and intermediate ultrasound operators.
- 📉 Operator Influence: The operator’s experience had a substantial impact on AI-generated diagnoses.
- 🔍 Diagnostic Range: Variability in diagnostic accuracy was observed, ranging from 40% to 100% across participants.
- 📋 Biopsy Recommendations: No significant change in biopsy recommendation accuracy was noted post-AI analysis.
- 🧑⚕️ Clinical Implications: The study highlights the need for further training and integration strategies for AI in clinical practice.
📚 Background
The integration of artificial intelligence (AI) in medical diagnostics has the potential to enhance accuracy and efficiency. However, understanding how AI tools like S-Detect for Thyroid interact with various levels of user expertise is crucial. This study aims to shed light on these dynamics, particularly in the context of thyroid nodule diagnosis, which is a common clinical challenge.
🗒️ Study
Conducted with a diverse group of 20 participants, this clinical trial involved performing ultrasound scans on five patients with thyroid nodules. Each participant utilized the same ultrasound systems equipped with S-Detect, allowing for a controlled comparison of diagnostic capabilities across different experience levels. The study aimed to evaluate both the AI’s performance and the influence of the operator’s experience on diagnostic outcomes.
📈 Results
The results indicated a mean diagnostic accuracy of 71.3% for S-Detect, with no significant differences between the groups (p = 0.31). The accuracy of biopsy recommendations was 69.8% before AI analysis and 69.2% after, showing no significant improvement (p = 0.75). These findings suggest that while AI can assist in diagnostics, it does not replace the need for skilled operators.
🌍 Impact and Implications
The implications of this study are significant for the future of AI in healthcare. It emphasizes the importance of operator training and the need for a collaborative approach between AI systems and healthcare professionals. As AI continues to evolve, understanding its limitations and the critical role of human expertise will be essential for improving diagnostic accuracy and patient outcomes.
🔮 Conclusion
This clinical trial highlights the complex interplay between AI technology and human operators in the field of ultrasound diagnostics. While S-Detect shows promise, the variability in diagnostic accuracy underscores the necessity for ongoing training and integration of AI tools in clinical practice. Future research should focus on enhancing user proficiency and exploring how AI can best support healthcare professionals in their diagnostic efforts.
💬 Your comments
What are your thoughts on the integration of AI in ultrasound diagnostics? Do you believe that operator experience will always play a crucial role? Let’s discuss! 💬 Share your insights in the comments below or connect with us on social media:
Human-AI collaboration for ultrasound diagnosis of thyroid nodules: a clinical trial.
Abstract
PURPOSE: This clinical trial examined how the articifial intelligence (AI)-based diagnostics system S-Detect for Thyroid influences the ultrasound diagnostic work-up of thyroid ultrasound (US) performed by different US users in clinical practice and how different US users influences the diagnostic accuracy of S-Detect.
METHODS: We conducted a clinical trial with 20 participants, including medical students, US novice physicians, and US experienced physicians. Five patients with thyroid nodules (one malignant and four benign) volunteered to undergo a thyroid US scan performed by all 20 participants using the same US systems with S-Detect installed. Participants performed a focused thyroid US on each patient case and made a nodule classification according to the European Thyroid Imaging Reporting And Data System (EU-TIRADS). They then performed a S-Detect analysis of the same nodule and were asked to re-evaluate their EU-TIRADS reporting. From the EU-TIRADS assessments by participants, we derived a biopsy recommendation outcome of whether fine needle aspiration biopsy (FNAB) was recommended.
RESULTS: The mean diagnostic accuracy for S-Detect was 71.3% (range 40-100%) among all participants, with no significant difference between the groups (p = 0.31). The accuracy of our biopsy recommendation outcome was 69.8% before and 69.2% after AI for all participants (p = 0.75).
CONCLUSION: In this trial, we did not find S-Detect to improve the thyroid diagnostic work-up in clinical practice among novice and intermediate ultrasound operators. However, the operator had a substantial impact on the AI-generated ultrasound diagnosis, with a variation in diagnostic accuracy from 40 to 100%, despite the same patients and ultrasound machines being used in the trial.
Author: [‘Edström AB’, ‘Makouei F’, ‘Wennervaldt K’, ‘Lomholt AF’, ‘Kaltoft M’, ‘Melchiors J’, ‘Hvilsom GB’, ‘Bech M’, ‘Tolsgaard M’, ‘Todsen T’]
Journal: Eur Arch Otorhinolaryngol
Citation: Edström AB, et al. Human-AI collaboration for ultrasound diagnosis of thyroid nodules: a clinical trial. Human-AI collaboration for ultrasound diagnosis of thyroid nodules: a clinical trial. 2025; (unknown volume):(unknown pages). doi: 10.1007/s00405-025-09236-9