machine-learning-algorithms-for-early-detection-bone-metastases-an

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine learning (ML) algorithms permit the integration of different features into a model to perform classification or regression tasks with an accuracy ...

German Cancer Research Center

Inaccurate conversion of CT data to water-equivalent path length (WEPL) is one of the most important uncertainty sources in ion treatment planning. Dual energy CT (DECT) imaging might help to reduce CT number ambiguities with the additional information. In our study we scanned a series of materials (tissue substitutes, aluminum, PMMA, and other polymers) in the dual source scanner (Siemens Somatom Definition Flash). Based on the 80kVp/140SnkVp dual energy images, the electron densities ϱe and effective atomic numbers Zeff were calculated. We introduced a new lookup table that translates the ϱe to the WEPL. The WEPL residuals from the calibration were significantly reduced for the investigated tissue surrogates compared to the empirical Hounsfield-look-up table (single energy CT imaging) from (-1.0±1.8)% to (0.1±0.7)% and for non-tissue equivalent PMMA from -7.8% to -1.0%. To assess the benefit of the new DECT calibration, we conducted a treatment planning study for three different idealized cases based on tissue surrogates and PMMA. The DECT calibration yielded a significantly higher target coverage in tissue surrogates and phantom material (i.e. PMMA cylinder, mean target coverage improved from 62% to 98%). To verify the DECT calibration for real tissue, ion ranges through a frozen pig head were measured and compared to predictions calculated by the standard single energy CT calibration and the novel DECT calibration. By using this method, an improvement of ion range estimation from -2.1% water-equivalent thickness deviation (single energy CT) to 0.3% (DECT) was achieved. If one excludes raypaths located on the edge of the sample accompanied with high uncertainties, no significant difference could be observed. 

Ion range estimation by using dual energy computed tomography.

To compare the relative biological effectiveness (RBE)-weighted tolerance doses for temporal lobe reactions after carbon ion radiation therapy using 2 different versions of the local effect model (LEM I vs LEM IV) for the same patient collective under identical conditions.

Temporal lobe reactions after carbon ion radiation therapy: comparison of relative biological effectiveness-weighted tolerance doses predicted by local effect models I and IV.

To derive the dose-response curve for temporal lobe reactions (TLRs) after proton therapy and to compare the resulting relative biological effectiveness (RBE)-weighted tolerance doses based on an RBE of 1.1 with published values for carbon ions, which were calculated by the two versions of the local effect model (LEM I or IV).

Dose-response curves for MRI-detected radiation-induced temporal lobe reactions in patients after proton and carbon ion therapy: Does the same RBE-weighted dose lead to the same biological effect?

Glioblastoma (GBM) is a typically lethal type of brain tumor with a median survival of 15 months postdiagnosis. This negative prognosis prompted the exploration of alternative treatment options. In particular, the reliance of GBM on angiogenesis triggered the development of anti-VEGF (vascular endothelial growth factor) blocking antibodies such as bevacizumab. Although its application in human GBM only increased progression-free periods but did not improve overall survival, physicians and researchers still utilize this treatment option due to the lack of adequate alternatives. In an attempt to improve the efficacy of anti-VEGF treatment, we explored the role of the  gene in malignant glioma. We found that the encoded extracellular matrix protein epidermal growth factor-like protein 7 (EGFL7) was secreted by glioma blood vessels but not glioma cells themselves, while no major role could be assigned to the parasitic miRNAs miR-126/126*. EGFL7 expression promoted glioma growth in experimental glioma models  and stimulated tumor vascularization. Mechanistically, this was mediated by an upregulation of integrin αβ on the cellular surface of endothelial cells, which enhanced fibronectin-induced angiogenic sprouting. Glioma blood vessels that formed  were more mature as determined by pericyte and smooth muscle cell coverage. Furthermore, these vessels were less leaky as measured by magnetic resonance imaging of extravasating contrast agent. EGFL7-inhibition using a specific blocking antibody reduced the vascularization of experimental gliomas and increased the life span of treated animals, in particular in combination with anti-VEGF and the chemotherapeutic agent temozolomide. Data allow for the conclusion that this combinatorial regimen may serve as a novel treatment option for GBM.

EGFL7 enhances surface expression of integrin αβ to promote angiogenesis in malignant brain tumors.

To compare the relative biological effectiveness (RBE)-weighted dose distributions in the target volume of chordoma and chondrosarcoma patients when using two different versions of the local effect model (LEM I vs. IV) under identical conditions.

RBE-weighted doses in target volumes of chordoma and chondrosarcoma patients treated with carbon ion radiotherapy: Comparison of local effect models I and IV.

Clarissa Gillmann

Department of Medical Physics in Radiation Oncology

German Cancer Research Center

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Clarissa Gillmann

.css-o250i3{font-size:18px;font-family:Open Sans,sans-serif;line-height:21.6px;}Department of Medical Physics in Radiation Oncology

German Cancer Research Center

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Department of Medical Physics in Radiation Oncology