Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning ...convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.
Purpose
Stereotactic radiosurgery (SRS) has become a standard of care for patients' with brain metastases (BMs). However, the manual multiple BMs delineation can be time‐consuming and could create an ...efficiency bottleneck in SRS workflow. There is a clinical need for automatic delineation and quantitative evaluation tools. In this study, building on our previous developed deep learning‐based segmentation algorithms, we developed a web‐based automated BMs segmentation and labeling platform to assist the SRS clinical workflow.
Method
This platform was developed based on the Django framework, including a web client and a back‐end server. The web client enables interactions as database access, data import, and image viewing. The server performs the segmentation and labeling tasks including: skull stripping; deep learning‐based BMs segmentation; and affine registration‐based BMs labeling. Additionally, the client can display BMs contours with corresponding atlas labels, and allows further postprocessing tasks including: (a) adjusting window levels; (b) displaying/hiding specific contours; (c) removing false‐positive contours; (d) exporting contours as DICOM RTStruct files; etc.
Results
We evaluated this platform on 10 clinical cases with BMs number varied from 12‐81 per case. The overall operation took about 4–5 min per patient. The segmentation accuracy was evaluated between the manual contour and automatic segmentation with several metrics. The averaged center of mass shift was 1.55 ± 0.36 mm, the Hausdorff distance was 2.98 ± 0.63 mm, the mean of surface‐to‐surface distance (SSD) was 1.06 ± 0.31 mm, and the standard deviation of SSD was 0.80 ± 0.16 mm. In addition, the initial averaged false‐positive over union (FPoU) and false‐negative rate (FNR) were 0.43 ± 0.19 and 0.15 ± 0.10 respectively. After case‐specific postprocessing, the averaged FPoU and FNR were 0.19 ± 0.10 and 0.15 ± 0.10 respectively.
Conclusion
The evaluated web‐based BMs segmentation and labeling platform can substantially improve the clinical efficiency compared to manual contouring. This platform can be a useful tool for assisting SRS treatment planning and treatment follow‐up.
Highlights • Hypoxia is associated with radiation resistance, particularly for SBRT. • Non-invasive OE-MRI is a potential biomarker to predict tumor response. • Breathing oxygen enhanced radiation ...response, with a wide range of responses. • OE-MRI demonstrates a correlation between T1 -weighted contrast and radiation response. • OE-MRI is highly translational.
Radiation dose is central to much of radiobiological research. Precision and accuracy of dose measurements and reporting of the measurement details should be sufficient to allow the work to be ...interpreted and repeated and to allow valid comparisons to be made, both in the same laboratory and by other laboratories. Despite this, a careful reading of published manuscripts suggests that measurement and reporting of radiation dosimetry and setup for radiobiology research is frequently inadequate, thus undermining the reliability and reproducibility of the findings. To address these problems and propose a course of action, the National Cancer Institute (NCI), the National Institute of Allergy and Infectious Diseases (NIAID), and the National Institute of Standards and Technology (NIST) brought together representatives of the radiobiology and radiation physics communities in a workshop in September, 2011. The workshop participants arrived at a number of specific recommendations as enumerated in this paper and they expressed the desirability of creating dosimetry standard operating procedures (SOPs) for cell culture and for small and large animal experiments. It was also felt that these SOPs would be most useful if they are made widely available through mechanism(s) such as the web, where they can provide guidance to both radiobiologists and radiation physicists, be cited in publications, and be updated as the field and needs evolve. Other broad areas covered were the need for continuing education through tutorials at national conferences, and for journals to establish standards for reporting dosimetry. This workshop did not address issues of dosimetry for studies involving radiation focused at the sub-cellular level, internally-administered radionuclides, biodosimetry based on biological markers of radiation exposure, or dose reconstruction for epidemiological studies.
The goal of this study was to explore conceptual benefits of characterizing delineated target volumes based on surface area and to utilize the concept for assessing risk of therapeutic toxicity in ...radiosurgery.
Four computer-generated targets, a sphere, a cylinder, an ellipsoid and a box, were designed for two distinct scenarios. In the first scenario, all targets had identical volumes, and in the second one, all targets had identical surface areas. High quality stereotactic radiosurgery plans with at least 95% target coverage and selectivity were created for each target in both scenarios. Normal brain volumes V12Gy, V14Gy and V16Gy corresponding to received dose of 12 Gy, 14 Gy and 16 Gy, respectively, were computed and analyzed. Additionally, V12Gy and V14Gy volumes and values for seven prospective toxicity variables were recorded for 100 meningioma patients after Gamma Knife radiosurgery. Multivariable stepwise linear regression and best subset linear regression analyses were performed in two statistical software packages, SAS/STAT and R, respectively.
In a phantom study, for the constant volume targets, the volumes of 12 Gy, 14 Gy and 16 Gy isodose clouds were the lowest for the spherical target as an expected corollary of the isoperimetric inequality. For the constant surface area targets, a conventional wisdom is confirmed, as the target volume increases the corresponding volumes V12Gy, V14Gy and V16Gy also increase. In the 100-meningioma patient cohort, the best univariate model featured tumor surface area as the most significantly associated variable with both V12Gy and V14Gy volumes, corresponding to the adjusted R2 values of 0.82 and 0.77, respectively. Two statistical methods converged to matching multivariable models.
In a univariate model, target surface area is a better predictor of spilled dose to normal tissue than target largest dimension or target volume itself. In complex multivariate models, target surface area is an independent variable for modeling radiosurgical normal tissue toxicity risk.
Multi-modality image-guided radiotherapy is the standard of care in contemporary cancer management; however, it is not common in preclinical settings due to both hardware and software limitations. ...Soft tissue lesions, such as orthotopic prostate tumors, are difficult to identify using cone beam computed tomography (CBCT) imaging alone. In this study, we characterized a research magnetic resonance (MR) scanner for preclinical studies and created a protocol for combined MR-CBCT image-guided small animal radiotherapy. Two in-house dual-modality, MR and CBCT compatible, phantoms were designed and manufactured using 3D printing technology. The phantoms were used for quality assurance tests and to facilitate end-to-end testing for combined preclinical MR and CBCT based treatment planning. MR and CBCT images of the phantoms were acquired utilizing a Varian 4.7 T scanner and XRad-225Cx irradiator, respectively. The geometry distortion was assessed by comparing MR images to phantom blueprints and CBCT. The corrected MR scans were co-registered with CBCT and subsequently used for treatment planning. The fidelity of 3D printed phantoms compared to the blueprint design yielded favorable agreement as verified with the CBCT measurements. The geometric distortion, which varied between -5% and 11% throughout the scanning volume, was substantially reduced to within 0.4% after correction. The distortion free MR images were co-registered with the corresponding CBCT images and imported into a commercial treatment planning software SmART Plan. The planning target volume (PTV) was on average 19% smaller when contoured on the corrected MR-CBCT images relative to raw images without distortion correction. An MR-CBCT based preclinical workflow was successfully designed and implemented for small animal radiotherapy. Combined MR-CBCT image-guided radiotherapy for preclinical research potentially delivers enhanced relevance to human radiotherapy for various disease sites. This novel protocol is wide-ranging and not limited to the orthotopic prostate tumor study presented in the study.
Ionizing radiation is genotoxic to cells. Healthy tissue toxicity in patients and radiation resistance in tumors present common clinical challenges in delivering effective radiation therapies. ...Radiation response is a complex, polygenic trait with unknown genetic determinants. The Drosophila Genetic Reference Panel (DGRP) provides a model to investigate the genetics of natural variation for sensitivity to radiation.
Radiation response was quantified in 154 inbred DGRP lines, among which 92 radiosensitive lines and 62 radioresistant lines were classified as controls and cases, respectively. A case-control genome-wide association screen for radioresistance was performed. There are 32 single nucleotide polymorphisms (SNPs) associated with radio resistance at a nominal p<10(-5); all had modest effect sizes and were common variants with the minor allele frequency >5%. All the genes implicated by those SNP hits were novel, many without a known role in radiation resistance and some with unknown function. Variants in known DNA damage and repair genes associated with radiation response were below the significance threshold of p<10(-5) and were not present among the significant hits. No SNP met the genome-wide significance threshold (p = 1.49 × 10(-7)), indicating a necessity for a larger sample size.
Several genes not previously associated with variation in radiation resistance were identified. These genes, especially the ones with human homologs, form the basis for exploring new pathways involved in radiation resistance in novel functional studies. An improved DGRP model with a sample size of at least 265 lines and ideally up to 793 lines is recommended for future studies of complex traits.
Detection and segmentation of brain metastases (BMs) play a pivotal role in diagnosis, treatment planning, and follow-up evaluations for effective BM management. Given the rising prevalence of BM ...cases and its predominantly multiple onsets, automated segmentation is becoming necessary in stereotactic radiosurgery. It not only alleviates the clinician's manual workload and improves clinical workflow efficiency but also ensures treatment safety, ultimately improving patient care. Recent strides in machine learning, particularly in deep learning (DL), have revolutionized medical image segmentation, achieving state-of-the-art results. This review aims to analyze auto-segmentation strategies, characterize the utilized data, and assess the performance of cutting-edge BM segmentation methodologies. Additionally, we delve into the challenges confronting BM segmentation and share insights gleaned from our algorithmic and clinical implementation experiences.
There is clinical interest in the modulation of regulatory T cells for cancer therapy. The safety of these therapies in combination with conventional anti-cancer therapies, including radiation ...therapy, can be studied in animal models. The effects of partial depletion of regulatory T (Treg) cells with an anti-CD25 antibody in conjunction with ionizing radiation on inflammation and tissue injury were analyzed in C57BL/6 mice. An anti-CD25 antibody (PC61) was administered 3 days prior to 13 Gy lower-half hemi-body irradiation (HBI). The blood, spleen, mesenteric lymph nodes (mLNs) and inguinal lymph nodes (iLNs) were harvested at various times thereafter. Alterations in the proportion of leukocyte subsets including CD4(+) T cells, CD8(+) T cells, Treg cells, B cells, NK cells, NK1.1(+) T cells, macrophages and granulocytes were analyzed by FACS. The lungs, liver, pancreas, stomach, jejunum, duodenum, ileum, colon and kidney were harvested and studied by H&E staining. Expression of inflammatory mediators in plasma and tissue were investigated by ELISA. HBI significantly decreased the leukocyte pool though the various leukocyte subsets had different sensitivities to HBI. The administration of PC61 significantly decreased the proportion of Treg cells in spleen, iLN, mLN and blood (reduction of approximately 60%). Irradiation significantly increased the proportion of Treg cells in the spleen, iLN and mLN. HBI induced a systemic inflammatory reaction as demonstrated by increased plasma levels of IL-6, KC/CXCL1 and circulating granulocytes in the blood. Neutrophils also infiltrated the small bowel. The same general patterns were observed whether or not Treg cells were partially depleted with PC61 prior to HBI. These data demonstrate that partial depletion of Treg cells in these mice does not influence HBI-induced inflammatory response and tissue injury, and that combining anti-CD25 therapy with radiation may be safe and well tolerated in a clinical setting.
PDF
