Automatic Parotid Gland Segmentation in MVCT Using Deep Convolutional Neural Networks

Radiation-induced xerostomia, as a major problem in radiation treatment of the head and neck cancer, is mainly due to the overdose irradiation injury to the parotid glands. Helical Tomotherapy-based megavoltage computed tomography (MVCT) imaging during the Tomotherapy treatment can be applied to monitor the successive variations in the parotid glands. While manual segmentation is time consuming, laborious, and subjective, automatic segmentation is quite challenging due to the complicated anatomical environment of head and neck as well as noises in MVCT images. In this article, we propose a localization-refinement scheme to segment the parotid gland in MVCT. After data pre-processing we use mask region convolutional neural network (Mask R-CNN) in the localization stage after data pre-processing, and design a modified U-Net in the following fine segmentation stage. To the best of our knowledge, this study is a pioneering work of deep learning on MVCT segmentation. Comprehensive experiments based on different data distribution of head and neck MVCTs and different segmentation models have demonstrated the superiority of our approach in terms of accuracy, effectiveness, flexibility, and practicability. Our method can be adopted as a powerful tool for radiation-induced injury studies, where accurate organ segmentation is crucial.

Comparison of Absolute Dose Achievable Between Helical Tomotherapy and RapidArc in Total Dura Mater Irradiation for Child Cancer

Background and Purpose: In this study, the absolute dose achievable between helical tomotherapy (HT) plans and RapidArc (RA) plans for total dura mater irradiation (TDMI) was compared. Materials and methods: A planning study was conducted on nine children's case datasets with dura mater metastasis of neuroblastoma. The target included the entire calvarium and skull base and formed a closed volume with a certain thickness around the brain. HT and RA plans with four coplanar full arcs (RA4) with half-field technique were generated for the comparison of absolute dose achievable. In total, 30.6 Gy was prescribed as D95% (ie, dose to 95% of PTV volume). Results: In the dosimetric comparison between the two modalities, HT provided more homogenous dose distribution than RA4 (mean HI5−95%: 1.046 vs 1.088, P < .001). The V107% and D2Gy of PTV in HT versus RA4 were 3.06% versus 30.47% and 32.59 Gy versus 33.45 Gy, respectively. HT reduced the Dmean and V5Gy of the brain, brainstem, and hippocampus by 25%–48% and 27%–56% compared with RA4, respectively. Conclusion: Both techniques could provide sufficient coverage for targets, but HT offered more homogenous dose to PTV and lower dose to the central region of the brain involving the brainstem and hippocampus. RA4 could be completed in a shorter time with lower MUs, but with relatively higher dose to the brain or hippocampus. In terms of dosimetry, HT may improve long-term cognitive decline in these young pediatric patients with TDMI.

Analysis of clinical patient-specific pre-treatment quality assurance with the new helical tomotherapy platform, following the AAPM TG-218 report

Purpose This study presents patient-specific quality assurance (QA) results from the first 395 clinical cases for the new helical TomoTherapy® platform (Radixact) coupled with dedicated Precision TPS. Methods The passing rate of the Gamma Index (GP%) of 395 helical QA of patient-specific tomotherapy, acquired with ArcCHECK, is presented, analysed and correlated to various parameters of the plan. Following TG-218 recommendations, the clinic specific action limit (ALcs) and tolerance limit (TLcs) were calculated for our clinic and monitored during the analysed period. Results The mean values ​​(± 1 standard deviation) of GP% (3%/2 mm) (both global and local normalization) are: 97.6% and 90.9%, respectively. The proposed ALcs and TLcs, after a period of two years’ process monitoring are 89.4% and 91.1% respectively. Conclusions The phantom measurements closely match the planned dose distributions, demonstrating that the calculation accuracy of the new Precision TPS and the delivery accuracy of the Radixact unit are adequate, with respect to international guidelines and reports. Furthermore, a first correlation with the planning parameters was made. Action and tolerance limits have been set for the new Radixact Linac.

Stereotactic Radiosurgery for Limited Brain Metastasis using Three Different Techniques; Helical Tomotherapy, Volumetric Modulated Arc Therapy, and Cone-Based Linac Radiosurgery.

Background: Specific radiation delivered to the tumor by stereotactic radiosurgery (SRS) has become widely used in the treatment of brain metastasis. This study aimed to compare radiation dosimetry and its parameters from SRS using three different modalities: Helical Tomotherapy (HT), Volumetric Modulated Arc Therapy (VMAT), and Cone-based Linac Radiosurgery (Cone-based). Methods: Patients who had experienced oligo-brain metastasis received SRS in our treatment center. Each contouring dataset was re-planned to create radiation dosimetry in all three treatment systems (HT, VMAT, and Cone-based). The parameters of conformity index (CI), homogeneity index (HI), CI50, and gradient index (CGI) were analyzed to compare the effects of the three techniques. Results: A total of 21 patients with 39 lesions were included in this study. For single lesion, Cone-based and VMAT revealed statistically identical CI, CI50, and CGI values, while exhibiting the poorest HI value. For multiple lesions, Cone-based provided the best CI50 and CGI values. VMAT displayed better CI50 and CGI values than HT. Moreover, VMAT exhibited the lowest BrainV5Gy value and displayed the shortest beam on time (BoT) calculation. Lastly, Decision Score analysis demonstrated better performance in VMAT when compared to the Cone-based approach. Conclusions: Our data indicated the dosimetric comparison between three radiation techniques for single and multiple lesions. The Cone-based technique revealed the poorest HI value. On the other hand, VMAT provided the best estimated BoT value. Moreover, we have reported on the feasibility of SRS using HT, which has been associated with well-tolerability and an acceptable level of radiation output.Trial registration: Thai Clinical Trials Registry (TCTR) ID No. TCTR20200803006Date of registration: 24 July 2020 (Retrospectively registered)URL: http://www.thaiclinicaltrials.org/show/TCTR20200803006

Assessing the dose rate delivery of helical TomoTherapy prostate and head & neck treatments

The dose rate distributions delivered to 55 prostate and head & neck (H&N) cancer patients treated with a helical TomoTherapy (HT) system were resolved and assessed with regard to pitch and field width defined during treatment planning. Statistical analysis of the studied cases showed that the median treatment delivery time was 4.4 min and 6.3 min for the prostate and H&N cases, respectively. Dose rate volume histogram data for the studied cases showed that the 25% and 12% of the volume of the planning target volumes of the prostate and H&N cases are irradiated with a dose rate of greater or equal to 1 Gy/min. Quartile dose rate (QDR) data confirmed that in HT, where the target is irradiated in slices, most of the dose is delivered to each voxel of the target when it travels within the beam. Analysis of the planning data from all cases showed that this lasts for 68 s (median value). QDRs results showed that using the 2.5 cm field width, 75% of the prescribed dose is delivered to target voxels with a median dose rate of at least 3.2 Gy/min and 4.5 Gy/min, for the prostate and H&N cases, respectively. Systematically higher dose rates were observed for the H&N cases due to the shallower depths of the lesions in this anatomical site. Delivered dose rates were also found to increase with field width and pitch setting, due to the higher output of the system which, in general, results in accordingly decreased total treatment time. The biological effect of the dose rate findings of this work needs to be further investigated using in-vitro studies and clinical treatment data.