Automatic bone marrow segmentation for PETCT imaging in multiple myeloma

Introduction Multiple myeloma (MM) is a malignant hematologic disorder characterized by bone marrow infiltration with neoplastic plasma cells. Approximately 10% of all hematologic cancers are related to MM. Whole-body 18F-FDG PETCT is an extremely useful imaging tool for the assessment of patients with MM. The novel approach developed in this research automatically segments bone marrow regions of interest on both the PET and CT datasets. Purpose To automate bone marrow segmentation in PET. Materials and methods Firstly, affine linear transforms are applied to the PET dataset and it is aligned to the CT images. Next, a binary mask is created based on a pixel threshold value of cortical bone. A series of image processing steps are performed to remove noise and fill gaps that correspond to bone marrow locations. This process results in a binary mask relating to bone marrow only which can then be applied to the registered PET dataset. Conclusion The proposed method offers a fully automated and completely objective approach for segmentation of anatomical regions relating to bone marrow. With further development, this method could be used to evaluate clinical images in order to develop a database of PETCT images against which quantitative statistical comparisons between patients with normal bone marrow metabolism and those with myeloma can be made, establishing a baseline against which future scans may be referenced. In cases where the suspicion of myeloma exists, the tools could be used to support the diagnosis of the disease, and may be useful in staging of the disease in cases positive for myeloma.