Clinical Application of an Artificial Intelligence System for Diagnosing Thyroid Disease Based on a Computer Neural...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Accurate differentiation between benign and malignant thyroid nodules is crucial for effective patient management.
• Current diagnostic methods can have limitations, necessitating advanced tools for improved accuracy.

Purpose of the Study:

• To develop an artificial intelligence (AI) diagnosis system utilizing a stereoscopic neural learning network for predicting benign and malignant thyroid diseases.
• To create AI-assisted thyroid diagnosis software with high sensitivity and a low false-negative rate for clinical application.

Main Methods:

• A stereoscopic neural learning network was established using deep learning techniques.
• Ultrasonic diagnostic characteristics from 36 patients were used to build a database for the AI system.
• The AI system's data was iteratively adjusted based on conformity rates and misjudgment analysis to exceed 90% accuracy.

Main Results:

• The initial AI software achieved an 88% conformity rate in identifying benign and malignant thyroid conditions.
• After database utilization and correction, the AI software's conformity rate increased to 94%.

Conclusions:

• A stereoscopic neural learning network and AI diagnosis system were successfully established for thyroid disease prediction.
• The developed AI system demonstrates potential for clinical practice with high sensitivity and a low false-negative rate in differentiating thyroid pathologies.