Automatic Segmentation and Measurement on Knee Computerized Tomography Images for Patellar Dislocation Diagnosis

Area of Science:

• Medical Imaging
• Orthopedic Surgery
• Computer-Aided Diagnosis

Background:

• Diagnosing patellar dislocation using traditional manual geometric measurements on CT images is complex and prone to errors.
• Accurate assessment of patellar alignment is crucial for effective treatment and patient outcomes.

Purpose of the Study:

• To develop and validate a prototype Computer-Aided Diagnosis (CAD) system for automatic measurement and diagnosis of patellar dislocation.
• To automate the measurement of key geometric parameters, namely patellar tilt angle (PTA) and patellar lateral shift (PLS).

Main Methods:

• Segmentation of the patella and femur bone regions from knee CT images.
• Automatic calculation of patellar tilt angle (PTA) and patellar lateral shift (PLS) based on segmented images.
• Experimental validation of the proposed measurement quantities and algorithms.

Main Results:

• The developed CAD system successfully segmented patellar and femur regions.
• Automatic measurement of PTA and PLS was achieved with high accuracy.
• Experimental results demonstrated the validity of the proposed geometric quantities and the effectiveness of the algorithms.

Conclusions:

• The prototype CAD system offers an effective and automated solution for diagnosing patellar dislocation.
• Automated measurement of PTA and PLS can significantly improve diagnostic accuracy and efficiency in clinical practice.
• This technology has the potential to reduce errors associated with manual measurements in orthopedic imaging.