Tracking Metastatic Brain Tumors in Longitudinal Scans via Joint Image Registration and Labeling

Area of Science:

• Medical imaging analysis
• Neurosurgery
• Computational oncology

Background:

• Accurate tumor volume measurement is crucial for stereotactic radiosurgery of metastatic brain tumors.
• Manual tumor delineation is time-consuming, complex, and susceptible to human error.
• Longitudinal tracking of tumor changes is essential for treatment assessment.

Purpose of the Study:

• To develop an automated method for detecting changes in metastatic brain tumor lesions over time.
• To assist clinicians in accurately determining tumor volumes from longitudinal scans.
• To improve the efficiency and reliability of tumor monitoring in neuro-oncology.

Main Methods:

• Jointly registering current and previous patient scans.
• Estimating intensity correspondence changes indicative of tumor growth or regression.
• Combining image registration, correspondence changes, and prior segmentations to estimate current tumor metastases.

Main Results:

• The automated method successfully detected changes in metastatic brain tumor lesions.
• Tumor tracking demonstrated promising results across 28 registrations using real patient data.
• The approach aids in estimating tumor volume changes over time.

Conclusions:

• The developed automated method shows potential for improving the assessment of metastatic brain tumors.
• This technique can enhance clinical decision-making in stereotactic radiosurgery.
• Automated longitudinal tumor analysis offers a more efficient and accurate alternative to manual methods.