Evaluation Exploration of Atlas-Based and Deep Learning-Based Automatic Contouring for Nasopharyngeal Carcinoma.

Purpose: The purpose of this study was to evaluate and explore the difference between an atlas-based and deep learning (DL)-based auto-segmentation scheme for organs at risk (OARs) of nasopharyngeal carcinoma cases to provide valuable help for clinical practice. Methods: 120 nasopharyngeal carcinoma cases were established in the MIM Maestro (atlas) database and trained by a DL-based model (AccuContour®), and another 20 nasopharyngeal carcinoma cases were randomly selected outside the atlas database. The experienced physicians contoured 14 OARs from 20 patients based on the published consensus guidelines, and these were defined as the reference volumes (Vref). Meanwhile, these OARs were auto-contoured using an atlas-based model, a pre-built DL-based model, and an on-site trained DL-based model. These volumes were named Vatlas, VDL-pre-built, and VDL-trained, respectively. The similarities between Vatlas, VDL-pre-built, VDL-trained, and Vref were assessed using the Dice similarity coefficient (DSC), Jaccard coefficient (JAC), maximum Hausdorff distance (HDmax), and deviation of centroid (DC) methods. A one-way ANOVA test was carried out to show the differences (between each two of them). Results: The results of the three methods were almost similar for the brainstem and eyes. For inner ears and temporomandibular joints, the results of the pre-built DL-based model are the worst, as well as the results of atlas-based auto-segmentation for the lens. For the segmentation of optic nerves, the trained DL-based model shows the best performance (p < 0.05). For the contouring of the oral cavity, the DSC value of VDL-pre-built is the smallest, and VDL-trained is the most significant (p < 0.05). For the parotid glands, the DSC of Vatlas is the minimum (about 0.80 or so), and VDL-pre-built and VDL-trained are slightly larger (about 0.82 or so). In addition to the oral cavity, parotid glands, and the brainstem, the maximum Hausdorff distances of the other organs are below 0.5 cm using the trained DL-based segmentation model. The trained DL-based segmentation method behaves well in the contouring of all the organs that the maximum average deviation of the centroid is no more than 0.3 cm. Conclusion: The trained DL-based segmentation performs significantly better than atlas-based segmentation for nasopharyngeal carcinoma, especially for the OARs with small volumes. Although some delineation results still need further modification, auto-segmentation methods improve the work efficiency and provide a level of help for clinical work.

Early Assessment of Treatment Responses During Radiation Therapy for Lung Cancer Using Quantitative Analysis of Daily Computed Tomography.

PURPOSE: To investigate early tumor and normal tissue responses during the course of radiation therapy (RT) for lung cancer using quantitative analysis of daily computed tomography (CT) scans. METHODS AND MATERIALS: Daily diagnostic-quality CT scans acquired using CT-on-rails during CT-guided RT for 20 lung cancer patients were quantitatively analyzed. On each daily CT set, the contours of the gross tumor volume (GTV) and lungs were generated and the radiation dose delivered was reconstructed. The changes in CT image intensity (Hounsfield unit [HU]) features in the GTV and the multiple normal lung tissue shells around the GTV were extracted from the daily CT scans. The associations between the changes in the mean HUs, GTV, accumulated dose during RT delivery, and patient survival rate were analyzed. RESULTS: During the RT course, radiation can induce substantial changes in the HU histogram features on the daily CT scans, with reductions in the GTV mean HUs (dH) observed in the range of 11 to 48 HU (median 30). The dH is statistically related to the accumulated GTV dose (R2 > 0.99) and correlates weakly with the change in GTV (R2 = 0.3481). Statistically significant increases in patient survival rates (P=.038) were observed for patients with a higher dH in the GTV. In the normal lung, the 4 regions proximal to the GTV showed statistically significant (P<.001) HU reductions from the first to last fraction. CONCLUSION: Quantitative analysis of the daily CT scans indicated that the mean HUs in lung tumor and surrounding normal tissue were reduced during RT delivery. This reduction was observed in the early phase of the treatment, is patient specific, and correlated with the delivered dose. A larger HU reduction in the GTV correlated significantly with greater patient survival. The changes in daily CT features, such as the mean HU, can be used for early assessment of the radiation response during RT delivery for lung cancer.

Radiation-induced CT number changes in GTV and parotid glands during the course of radiation therapy for nasopharyngeal cancer.

OBJECTIVE: To investigate the changes in CT number (CTN) in gross tumour volume (GTV) and organs at risk (OARs) during the course of radiation therapy (RT) for nasopharyngeal cancer (NPC). METHODS: Daily megavoltage CT (MVCT) data collected from 30 patients with NPC treated with a prescription dose of 70 Gy in 30-33 fractions using helical tomotherapy were retrospectively analyzed. The contours of GTV and OARs on daily MVCTs were obtained by populating the planning contours from planning CT to daily MVCTs with manual editing, if necessary. The changes of GTV and OAR volumes and the histograms of CTN in the GTV and OARs during the course of RT delivery were analyzed. RESULTS: Volumes of GTV and parotid glands were reduced during the course of radiation treatment, with an average shrinkage rate of 0.23% per day (range, 0.02-0.8%) and 1.2% per day (range, 0.2-2.3%), respectively. The mean CTN changes in GTV and ipsilateral and contralateral parotid glands were reduced by 52 ± 35 HU, 18 ± 20 HU and 17 ± 22 HU, respectively. For GTV, the CTN and GTV volume decreases were found to be correlated with each other (p < 0.0001). No noticeable CTN change was found in the spinal cord and non-specified tissue irradiated with low doses. CONCLUSION: The CTN changes in GTV and parotids are measurable during the delivery of fractionated radiotherapy for NPC, were associated with the doses received (the number of fractions delivered) and were patient specific. ADVANCES IN KNOWLEDGE: The CTN change during radiotherapy is dose dependent and is measurable for NPC.

Computed tomography number changes observed during computed tomography-guided radiation therapy for head and neck cancer.

PURPOSE: To investigate CT number (CTN) changes in gross tumor volume (GTV) and organ at risk (OAR) according to daily diagnostic-quality CT acquired during CT-guided intensity modulated radiation therapy for head and neck cancer (HNC) patients. METHODS AND MATERIALS: Computed tomography scans acquired using a CT-on-rails during daily CT-guided intensity modulated radiation therapy for 15 patients with stage II to IVa squamous cell carcinoma of the head and neck were analyzed. The GTV, parotid glands, spinal cord, and nonspecified tissue were generated on each selected daily CT. The changes in CTN distributions and the mean and mode values were collected. Pearson analysis was used to assess the correlation between the CTN change, organ volume reduction, and delivered radiation dose. RESULTS: Volume and CTN changes for GTV and parotid glands can be observed during radiation therapy delivery for HNC. The mean (±SD) CTNs in GTV and ipsi- and contralateral parotid glands were reduced by 6 ± 10, 8 ± 7, and 11 ± 10 Hounsfield units, respectively, for all patients studied. The mean CTN changes in both spinal cord and nonspecified tissue were almost invisible (<2 Hounsfield units). For 2 patients studied, the absolute mean CTN changes in GTV and parotid glands were strongly correlated with the dose delivered (P<.001 and P<.05, respectively). For the correlation between CTN reductions and delivered isodose bins for parotid glands, the Pearson coefficient varied from -0.98 (P<.001) in regions with low-dose bins to 0.96 (P<.001) in high-dose bins and were patient specific. CONCLUSIONS: The CTN can be reduced in tumor and parotid glands during the course of radiation therapy for HNC. There was a fair correlation between CTN reduction and radiation doses for a subset of patients, whereas the correlation between CTN reductions and volume reductions in GTV and parotid glands were weak. More studies are needed to understand the mechanism for the radiation-induced CTN changes.

Spectral decomposition of susceptibility artifacts for spectral-spatial radiofrequency pulse design.

Susceptibility induced signal loss is a limitation in gradient echo functional MRI. The through-plane artifact in axial slices is particularly problematic due to the inferior position of air cavities in the brain. Spectral-spatial radiofrequency pulses have recently been shown to reduce signal loss in a single excitation. The pulses were successfully demonstrated assuming a linear relationship between susceptibility gradient and frequency, however, the exact frequency and spatial distribution of the susceptibility gradient in the brain is unknown. We present a spiral spectroscopic imaging sequence with a time-shifted radiofrequency pulse that can spectrally decompose the through-plane susceptibility gradient for spectral-spatial radiofrequency pulse design. Maps of the through-plane susceptibility gradient as a function of frequency were generated for the human brain at 3T. We found that the linear relationship holds well for the whole brain with an optimal slope of -1.0 µT/m/Hz.

Accelerated multidimensional radiofrequency pulse design for parallel transmission using concurrent computation on multiple graphics processing units.

Multidimensional radiofrequency (RF) pulses are of current interest because of their promise for improving high-field imaging and for optimizing parallel transmission methods. One major drawback is that the computation time of numerically designed multidimensional RF pulses increases rapidly with their resolution and number of transmitters. This is critical because the construction of multidimensional RF pulses often needs to be in real time. The use of graphics processing units for computations is a recent approach for accelerating image reconstruction applications. We propose the use of graphics processing units for the design of multidimensional RF pulses including the utilization of parallel transmitters. Using a desktop computer with four NVIDIA Tesla C1060 computing processors, we found acceleration factors on the order of 20 for standard eight-transmitter two-dimensional spiral RF pulses with a 64 × 64 excitation resolution and a 10-µsec dwell time. We also show that even greater acceleration factors can be achieved for more complex RF pulses.

Simple analytical dual-band spectral-spatial RF pulses for B(1) + and susceptibility artifact reduction in gradient echo MRI.

Susceptibility artifacts and transmission radio frequency (RF) field (B(1) +) inhomogeneity are major limitations in high-field gradient echo MRI. Previously proposed numerical 2D spectral-spatial RF pulses have been shown to be promising for reducing the through-plane signal loss susceptibility artifact by incorporating a frequency-dependent through-plane phase correction. This method has recently been extended to 4D spectral-spatial RF pulse designs for reducing B(1) + inhomogeneity as well as the signal loss. In this manuscript, we present simple analytical pulse designs for constructing 2D and 4D spectral-spatial RF pulses as an alternative to the numerical approaches. The 2D pulse capable of exciting slices with reduced signal loss and is lipid suppressing. The 4D pulse simultaneously corrects signal loss as well as the B(1) + inhomogeneity from a body coil transmitter. The pulses are demonstrated with simulations and with gradient echo phantom and brain images at 3T using a standard RF body coil. The pulses were observed to work well for multiple slices and several volunteers.

Four-dimensional spectral-spatial RF pulses for simultaneous correction of B1+ inhomogeneity and susceptibility artifacts in T2*-weighted MRI.

Susceptibility artifacts and excitation radiofrequency field B(1)+ inhomogeneity are major limitations in high-field MRI. Parallel transmission methods are promising for reducing artifacts in high-field applications. In particular, three-dimensional RF pulses have been shown to be useful for reducing B(1)+ inhomogeneity using multiple transmitters due to their ability to spatially shape the slice profile. Recently, two-dimensional spectral-spatial pulses have been demonstrated to be effective for reducing the signal loss susceptibility artifact by incorporating a frequency-dependent through-plane phase correction. We present the use of four-dimensional spectral-spatial RF pulses for simultaneous B(1)+ and through-plane signal loss susceptibility artifact compensation. The method is demonstrated with simulations and in T(2)*-weighted human brain images at 3 T, using a four-channel parallel transmission system. Parallel transmission was used to reduce the in-plane excitation resolution to improve the slice-selection resolution between two different pulse designs. Both pulses were observed to improve B(1)+ homogeneity and reduce the signal loss artifact in multiple slice locations and several human volunteers.

Simultaneous z-shim method for reducing susceptibility artifacts with multiple transmitters.

The signal loss susceptibility artifact is a major limitation in gradient-echo MRI applications. Various methods, including z-shim techniques and multidimensional tailored radio frequency (RF) pulses, have been proposed to mitigate the through-plane signal loss artifact, which is dominant in axial slices above the sinus region. Unfortunately, z-shim techniques require multiple steps and multidimensional RF methods are complex, with long pulse lengths. Parallel transmission methods were recently shown to be promising for improving B1 inhomogeneity and reducing the specific absorption rate. In this work, a novel method using time-shifted slice-select RF pulses is presented for reducing the through-plane signal loss artifact in parallel transmission applications. A simultaneous z-shim is obtained by concurrently applying unique time-shifted pulses on each transmitter. The method is shown to reduce the signal loss susceptibility artifact in gradient-echo images using a four-channel parallel transmission system at 3T.