Design and implementation of a low-cost gimbal-based angular ultrasound gantry for optimal tissue slice selection...

Area of Science:

• Medical Imaging
• Computational Pathology
• Artificial Intelligence in Medicine

Background:

• Ultrasound (US) is crucial for diagnosing tumors and soft tissue pathology.
• Manual slicing of excised tumors is time-consuming, resource-intensive, and prone to human error.
• Automating tissue slice selection can enhance diagnostic accuracy and efficiency.

Purpose of the Study:

• To develop a cost-effective ultrasound gantry system integrated with a deep learning algorithm.
• To automate the process of selecting optimal tissue slices for microscopic analysis.
• To improve the accuracy and efficiency of tumor diagnosis and treatment planning.

Main Methods:

• Development of a linear ultrasound gantry system for B-mode image acquisition.
• Enhancement to an angular ultrasound gantry system capable of multi-angle scanning for comprehensive geometric analysis.
• Integration of a deep learning algorithm to predict optimal tissue slices from acquired US images.

Main Results:

• The angular ultrasound gantry system demonstrated significant improvement over the linear design.
• The system achieved 98% accuracy in selecting the optimal tissue slice.
• Automation reduces time, resources, and human error in tissue preparation for analysis.

Conclusions:

• The developed angular ultrasound gantry system with deep learning effectively automates optimal tissue slice selection.
• This automated approach minimizes diagnostic uncertainty, aiding in accurate tumor grading and typing.
• The system has the potential to reduce treatment risks by improving diagnostic accuracy and planning.