Browsing by Subject "Gliomas"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Dosimetric Impact of Source Displacement in GammaTile Surgically Targeted Radiation Therapy for Gliomas
    (Springer Nature, 2023-05-02) Ng, Sook Kien; Yue, Yong; Shiue, Kevin; Shah, Mitesh V.; Le, Yi; Radiation Oncology, School of Medicine
    Background: This study aims to evaluate dosimetric changes that happened during the first month after GammaTile surgically targeted radiation therapy (STaRT) for gliomas due to Cesium-131 (Cs-131) seed displacement caused by cavity shrinkage in brain brachytherapy. Methodology: In this study, 10 glioma patients had 4-11 GammaTiles placed along the resection bed after maximal safe resection during craniotomy. Each GammaTile is composed of four Cs-131 seeds embedded in a biodegradable collagen sponge to minimize seed movement and maintain seed-to-cavity surface distance. The Cs-131 seed positions were identified using VariSeed on day one. On day 30, post-implant computed tomography (CT) images and dosimetry parameters were calculated. An iterative closest point (ICP) algorithm was used to compute rigid transformation between the day one and day 30 seed clouds. The seed displacement was calculated after registration. The volume receiving 100% of the prescription dose (V100), the dose received by 90% of the planning target volume (D90_PTV), the planning target volume receiving 100% of the prescription dose (V100_PTV), and the dose to organs at risk (OARs) were calculated for both CT images to determine the dosimetric changes from any seed displacement. Results: The mean seed displacement of 1.8 ± 1.0 mm for all patients was observed between day one and day 30. The maximum seed displacement for each patient ranged from 2.3 mm to 7.3 mm. The mean V100 difference between day one and day 30 was 2.5 cc (range = 0.5-6.5 cc). The mean D90_PTVs were 95.5% (range = 69.0%-131.0%) and 98.1% (range = 19.9%-149.0%) on day one and day 30, respectively. The mean V100_PTVs were 88.4% (range = 81.3%-99.1%) and 87.9% (range = 47.0%-99.7%) on day one and day 30, respectively. On day one, the brainstem dose was 63.5 Gy for one case and 28.1 Gy for another case; while on day 30, the brainstem dose was 55.8 Gy and 20.6 Gy for the same patients, contributing to 7.7 Gy (12.8%) and 7.5 Gy (12.5%) dose reductions to brainstem for these patients, respectively. Only two patients received a dose to the optic nerves (34.1 Gy and 5.2 Gy). There were small changes (1.8 Gy and 0.5 Gy, respectively) in the dose to optic nerves when comparing the dose calculated on day one and the dose calculated on day 30 CT images. The same two patients received 30.4 Gy and 6.8 Gy to the chiasm, respectively. Small changes in the dose to the chiasm (≤1.1 Gy) were noted between day one and day 30. Conclusions: A maximum seed displacement of up to 7.3 mm and a mean seed displacement of 1.8 mm caused by cavity shrinkage were observed during the first month after GammaTile STaRT for gliomas. There were noticeable changes in dosimetry parameters. Changes in the doses to OARs, particularly the brainstem, were large (up to 12.8% of the prescription dose). These changes in dosimetry should be considered when evaluating treatment outcomes and planning future GammaTile treatments.
  • Thumbnail Image
    Item
    Enhancing clinical decision-making: An externally validated machine learning model for predicting isocitrate dehydrogenase mutation in gliomas using radiomics from presurgical magnetic resonance imaging
    (Oxford University Press, 2024-10-03) Lost, Jan; Ashraf, Nader; Jekel, Leon; von Reppert, Marc; Tillmanns, Niklas; Willms, Klara; Merkaj, Sara; Cassinelli Petersen, Gabriel; Avesta, Arman; Ramakrishnan, Divya; Omuro, Antonio; Nabavizadeh, Ali; Bakas, Spyridon; Bousabarah, Khaled; Lin, MingDe; Aneja, Sanjay; Sabel, Michael; Aboian, Mariam; Pathology and Laboratory Medicine, School of Medicine
    Background: Glioma, the most prevalent primary brain tumor, poses challenges in prognosis, particularly in the high-grade subclass, despite advanced treatments. The recent shift in tumor classification underscores the crucial role of isocitrate dehydrogenase (IDH) mutation status in the clinical care of glioma patients. However, conventional methods for determining IDH status, including biopsy, have limitations. Exploring the use of machine learning (ML) on magnetic resonance imaging to predict IDH mutation status shows promise but encounters challenges in generalizability and translation into clinical practice because most studies either use single institution or homogeneous datasets for model training and validation. Our study aims to bridge this gap by using multi-institution data for model validation. Methods: This retrospective study utilizes data from large, annotated datasets for internal (377 cases from Yale New Haven Hospitals) and external validation (207 cases from facilities outside Yale New Haven Health). The 6-step research process includes image acquisition, semi-automated tumor segmentation, feature extraction, model building with feature selection, internal validation, and external validation. An extreme gradient boosting ML model predicted the IDH mutation status, confirmed by immunohistochemistry. Results: The ML model demonstrated high performance, with an Area under the Curve (AUC), Accuracy, Sensitivity, and Specificity in internal validation of 0.862, 0.865, 0.885, and 0.713, and external validation of 0.835, 0.851, 0.850, and 0.847. Conclusions: The ML model, built on a heterogeneous dataset, provided robust results in external validation for the prediction task, emphasizing its potential clinical utility. Future research should explore expanding its applicability and validation in diverse global healthcare settings.