SummaryThe use of Cone-beam Computed Tomography (CBCT) in radiotherapy is increasing due to the widespread implementation of kilovoltage systems on the currently available linear accelerators. Cone beam CT acts as an effective Image-Guided Radiotherapy (IGRT) tool for the verification of patient position. It also opens up the possibility of real-time re-optimization of treatment plans for Adaptive Radiotherapy (ART). This paper reviews the most prominent applications of CBCT (linac-mounted) in radiation therapy, focusing on CBCT-based planning and dose calculation studies. This is followed by a concise review of the main issues associated with CBCT, such as imaging artifacts, dose and image quality. It explores how medical physicists and oncologists can best apply CBCT for therapeutic applications.

Cone beam computed tomography (CBCT) images obtained from linac-based kV imagers are typically used for image-guided radiotherapy, in particular to perform three-dimensional image matching. CBCT image sets can also be used for adaptive radiotherapy where the treatment plan is modified on the basis of periodic imaging throughout the treatment course. CBCT images provide both anatomical information and Hounsfield unit (HU) values, which are required for dose calculations. This study evaluates treatment plans based on CBCT datasets calibrated using the Catphan 504 phantom to investigate the feasibility of using CBCT for adaptive replanning. The CBCT images were acquired from a Varian On-Board Imager system. Conventional planning CT (PCT) images obtained from a Philips Brilliance Big Bore CT scanner were used as reference images. The HU-density calibration curves of CBCT were obtained using a Catphan 504 phantom and a CIRS density phantom and compared with the clinical PCT calibration curve (obtained using the CIRS density phantom). Treatment plans created using the different calibration curves were compared. Identical targets were delineated on CBCT and PCT images on four different-sized phantoms and planar dose maps were generated. The dose-volume histograms of PCT-and CBCT-based plans were compared and evaluated by gamma analysis. To extend the study to a typical clinical situation, two prostate cases were included. The dose distribution comparison between PCT-and CBCT-based plans for patients yielded similar results to those obtained using phantoms. The study also analyzed the effect of phantom dimensions on HU values and its impact on dose calculations. The isodose distributions computed based on PCT and CBCT using the Catphan calibration curve agree to within ± 1% compared to that based on CBCT using the density phantom calibration curve. However, for phantoms of larger diameter, there is a pronounced discrepancy in the 50% and 60% isodose lines, with the dose difference being about ± 3%. For phantoms whose thickness is less than the cone beam scan length (16 cm) and for phantoms whose diameter is less than that of the calibration phantom, the variation in HU values is high. The effect of a change in radial diameter has a larger impact on dose calculations. This study shows that the CIRS density phantom is not suitable for CBCT calibration and that individual calibration curves obtained using phantoms of appropriate dimensions should be used for planning individual treatment sites.

Cone-beam CT (CBCT) using kV imagers integrated with linear accelerators is now widely used in verifying patient position during radiation therapy. Current CBCT acquisition protocols have lowered tube current to keep the imaging dose to a minimum. This affects the usability of CBCT data sets in treatment planning by reducing the soft tissue contrast and accuracy of CT numbers (Hounsfield values). The purpose of this study is to investigate the effect of reconstruction filters on full-fan and half-fan acquisition modes of CBCT and assess the image quality parameters of contrast- to -noise ratio, spatial resolution, pixel stability and uniformity. The results of this study show the relation between the noise and resolution of a CBCT image by using different reconstruction filters and provide possible estimations of the impact of filters on image quality and subsequent optimization for image-guided radiotherapy purposes.

Brain tumor is a dreadful disease which occurs when abnormal cells form uncontrollably. The modality adopted to detect abnormalities is Magnetic Resonance Imaging (MRI). MRI brain images contain nonbrain tissues. One of the important preprocessing steps is the whole brain segmentation, the process of skull stripping which isolates brain tissue and non-brain tissue. Segmentation is tedious and consumes more time only well experienced radiologist or a clinical expert can perform it with best accuracy. In order to overcome these limitations, computer aided medical diagnosis is essential. In this work, an intelligent and a robust skull stripping algorithm using mathematical morphology suited for all types of MR sequences is proposed. The method was validated on the international database collected from whole brain Atlas. The performance was evaluated using the metrics Jaccard Similarity Coefficient (JSC), Dice Similarity Coefficient (DSC), False Positive Rate (FPR), False Negative Rate (FNR), sensitivity, specificity and accuracy. An average of 97.25% indicates better overlap between proposed skull stripping and manual stripping by radiologists as a gold standard. The simulation results proved high accuracy in comparison to the ground truth results which is evident from the similarity coefficient metrics.

No abstract

No abstract

In this paper, an optimized bilevel brain tumor diagnostic system for identifying the tumor type at the first level and grade of the identified tumor at the second level is proposed using genetic algorithm, decision tree, and fuzzy rule-based approach. The dataset is composed of axial MRI of brain tumor types and grades. From the images, various features such as first and second order statistical and textural features are extracted (26 features). In the first level, tumor type classification was done using decision tree constructed with all features. Further evolutionary computing using genetic algorithms (GA) was applied to select the optimal discriminating feature set (5 features) and classification using the decision tree constructed with the reduced feature set resulted in better performance. In the second level, grade classification, a fuzzy rule-based approach was used to resolve the uncertainty in discriminating the tumor grades II and III. Membership functions of all grades were defined for all features extracted from brain tumor grade images, to derive the fuzzy inference rules for grade discrimination. Similar to type classification with GA, better grade discrimination performance was exhibited with fuzzy inference rules derived using optimal feature set (13 features) using GA. Overall performance comparison of the proposed bilevel classifier with all features vs GA-based feature selection, shows that evolutionary computing combined with fuzzy rule-based approach is successful in reducing false positives, thereby enhancing classifier performance.

No abstract