PyPeCT2S: Pythonic paediatric computed tomography to strength with automatic landmarking for the automation of bone...

Area of Science:

• Biomedical Engineering
• Orthopedics
• Medical Imaging

Background:

• Quantitative computed tomography (QCT) and finite element analysis (FEA) are established methods for predicting bone strength.
• Current QCT-FEA pipelines are time-consuming and require specialized training, limiting their clinical application, especially for pediatric populations.

Purpose of the Study:

• To develop and validate an automated computed tomography to strength (CT2S) pipeline for pediatric femur analysis.
• To create a user-friendly, repeatable, and extensible platform (PyPeCT2S) that reduces the need for manual input and specialized training.

Main Methods:

• Development of an automated landmarking technique for pediatric CT scans.
• Creation of FEA models using the PyPeCT2S pipeline on CT scans from 69 children.
• Comparison of FEA results from automatic versus manual landmarking under four-point bending conditions.

Main Results:

• The automated PyPeCT2S pipeline demonstrated comparable FEA critical moment results to manual methods, with values ranging from 0.19-167.94 Nm.
• Automatic landmarking showed minimal differences in FEA outcomes compared to manual landmarking but was substantially faster.
• The pipeline achieved a mean time reduction of 49-61% (up to 70%), decreasing processing time from 35 to 13 minutes.

Conclusions:

• The pythonic approach significantly speeds up FE-based bone strength prediction in pediatric femurs.
• The automated pipeline simplifies operation, enhances efficiency, and reduces the potential for human error.
• PyPeCT2S is a promising tool for clinical settings, enabling broader use in analyzing pediatric bone health.