Estimation of the tissue composition of the tumour mass in neuroblastoma using segmented CT images

Abstract

Neuroblastoma is the most common extra-cranial, solid, malignant tumour in children. Advances in radiology have made possible the detection and staging of the disease. Nevertheless, there is no method available at present that can go beyond detection and qualitative analysis, towards quantitative assessment of the tissue composition of the primary tumour mass in neuroblastoma. Such quantitative analysis could provide important information and serve as a decision-support tool to the radiologist and the oncologist, result in better treatment and follow-up and even lead to the avoidance of delayed surgery. The problem investigated was the improvement of the analysis of the primary tumour mass, in patients with neuroblastoma, using X-ray computed tomography (CT) images. A methodology was proposed for the estimation of the tissue content of the mass: it comprised a Gaussian mixture model for estimation, from segmented CT images, of the tissue composition of the primary tumour. To demonstrate the potential of the method, the results are presented of its application to ten CT examinations of four patients. The method provides quantitative information, and it was observed that the tumour in one of the patients reduced from 523 cm3 to 81 cm3 in volume, with an increase in calcification from about 20% to about 88% of the tumour volume, in response to chemotherapy over a period of five months. Results indicate that the proposed technique may be of considerable value in assessing the response to therapy of patients with neuroblastoma.

A curvature-based approach for left ventricular shape analysis from cardiac magnetic resonance imaging

“…In recent years, computerized image segmentation has played an increasingly important role in the field of medical imaging, and has been applied in various applications [2,4,11]. Some popular methods used for medical image segmentation include thresholding [30], edge-based methods [26], deformable models [24] and level-sets methods [6].…”

Landmarking and segmentation of computed tomographic images of pediatric patients with neuroblastoma

“…Once segmentation of the tumor mass is achieved, a Gaussian mixture model [33] could be applied to extract quantitative information corresponding to the tumor's tissue composition. Typically, in response to therapy, the tumor mass shrinks in size and the tissue composition changes toward increased proportions of necrotic and calcified tissues [33]. Quantifying such information could aid the radiologist in assessing a patient's response to therapy, and may help in guiding subsequent therapy.…”

“…As such, radiological evaluation of the tumor could provide important indications of the response of the disease to therapy [30]. In this context, computeraided analysis in the form of tumor segmentation and analysis [31][32][33] can be beneficial to radiologists, providing for quantitative and reproducible evaluation of the tumor; it could also assist in objective assessment of the response to therapy and in the planning of delayed surgery.…”

Segmentation and analysis of neuroblastoma

“…Ayres et al [36] proposed a statistical parametric model (Gaussian mixture model) for the tissue composition of the tumor, and used an expectation-maximization algorithm to estimate the parameters of the model. This model quantitatively assesses, using segmented image-based examinations, the amount of viable tumor, necrosis, calcification, and other tissues in a given tumor mass.…”

“…Brodeur et al [8] recommend that the volume of the primary tumor and large metastases be established to the best approximation, and sequentially used to determine response to therapy. Volume change may indicate response to therapy, and can be used as an index of the extent of disease [35,36].…”