Method for quantifying volumetric lesion change in interval liver CT examinations

Area of Science:

• Medical Imaging
• Radiology
• Computational Anatomy

Background:

• Accurate measurement of lesion interval change in liver computed tomography (CT) exams is crucial for treatment monitoring.
• Existing methods may be limited by registration accuracy and sensitivity to minor misalignments.

Purpose of the Study:

• To develop and validate a novel method for precise interval change measurement of homogeneous contrast lesions in serial liver CT scans.
• To assess the accuracy of the proposed method using simulated data and clinical validation with implanted microcoils.

Main Methods:

• A low-degree-of-freedom (DOF) warping technique (3 x 24 DOF) is employed to align serial liver CT examinations.
• Lesion volume change is calculated using the difference volume between registered scans, incorporating partial volume contributions.
• A probabilistic liver atlas aids in automatic liver masking and registration refinement.

Main Results:

• The proposed method accurately quantifies lesion volume changes, demonstrating insensitivity to slight misregistration.
• Accuracy was validated using a mathematically generated phantom under noisy conditions.
• CT liver registration accuracy was further confirmed using implanted chemotherapy microcoils.

Conclusions:

• The developed low-DOF warping method offers a robust and accurate approach for assessing interval changes in liver lesions from serial CT data.
• This technique enhances the reliability of lesion volume change quantification, supporting clinical decision-making in liver cancer management.