Artificial Intelligence in Thyroid Nodules and Cancer: Clinical Validation and Real-World Performance

Area of Science:

• Endocrinology
• Medical Imaging
• Artificial Intelligence

Background:

• Thyroid nodules are common, but most are benign, posing a challenge for accurate risk stratification.
• Distinguishing benign from malignant nodules requires careful evaluation to avoid overdiagnosis and overtreatment.
• Artificial intelligence (AI) is emerging as a powerful tool in medical decision support.

Purpose of the Study:

• To review current evidence on AI applications in thyroid nodule and differentiated thyroid cancer management.
• To highlight AI's role in ultrasound-based diagnosis, cytology, histopathology, and surveillance.
• To discuss the potential and limitations of AI in clinical practice.

Main Methods:

• Synthesis of current evidence on AI applications in thyroid nodule evaluation.
• Emphasis on ultrasound-based computer-aided diagnosis and decision support systems.
• Review of AI integration across various diagnostic modalities including cytology and histopathology.

Main Results:

• AI systems demonstrate potential to enhance consistency and support less experienced clinicians.
• AI may reduce unnecessary fine-needle aspirations without compromising sensitivity.
• Commercial AI systems show promise in improving diagnostic accuracy and patient management.

Conclusions:

• AI shows significant potential as an adjunct to clinical judgment for thyroid nodule risk stratification and management.
• AI-enabled de-escalation strategies may support surveillance for low-risk nodules.
• Further prospective validation, transparent reporting, and regulatory oversight are crucial for widespread AI adoption.