Pseudoprogression prediction in high grade primary CNS tumors by use of radiomics

Abstract

Our aim is to define the capabilities of radiomics and machine learning in predicting pseudoprogression development from pre-treatment MR images in a patient cohort diagnosed with high grade gliomas. In this retrospective analysis, we analysed 131 patients with high grade gliomas. Segmentation of the contrast enhancing parts of the tumor before administration of radio-chemotherapy was semi-automatically performed using the 3D Slicer open-source software platform (version 4.10) on T1 post contrast MR images. Imaging data was split into training data, test data and an independent validation sample at random. We extracted a total of 107 radiomic features by hand-delineated regions of interest (ROI). Feature selection and model construction were performed using Generalized Boosted Regression Models (GBM). 131 patients were included, of which 64 patients had a histopathologically proven progressive disease and 67 were diagnosed with mixed or pure pseudoprogression after initial treatment. Our Radiomics approach is able to predict the occurrence of pseudoprogression with an AUC, mean sensitivity, mean specificity and mean accuracy of 91.49% [86.27%, 95.89%], 79.92% [73.08%, 87.55%], 88.61% [85.19%, 94.44%] and 84.35% [80.19%, 90.57%] in the full development group, 78.51% [75.27%, 82.46%], 66.26% [57.95%, 73.02%], 78.31% [70.48%, 84.19%] and 72.40% [68.06%, 76.85%] in the testing group and finally 72.87% [70.18%, 76.28%], 71.75% [62.29%, 75.00%], 80.00% [69.23%, 84.62%] and 76.04% [69.90%, 80.00%] in the independent validation sample, respectively. Our results indicate that radiomics is a promising tool to predict pseudo-progression, thus potentially allowing to reduce the use of biopsies and invasive histopathology.

Exploring the Application of the Artificial-Intelligence-Integrated Platform 3D Slicer in Medical Imaging Education

“…Shi et al also utilized 3D Slicer for preoperative assessments [43,45,47,50]. Additionally, Ari et al [10] and Brown et al [11] leveraged 3D Slicer for extracting and analyzing omics features. Despite the apparent distance from frontline teaching, it is imperative to recognize that a crucial facet of higher education is its application-oriented and employment-focused approach.…”

“…Moreover, it supports extensions, with over a hundred open-source extensions available on the platform. These extensions range from radiomics analysis [10,11] to artificial intelligence (AI)-based automatic organ segmentation for medical image analysis [12,13] and from surgical navigation [14,15] to target delineation [16,17] and dose calculation for radiation therapy clinical tools [18,19]. Its extensive functionality surpasses that of professional workstations utilized in clinical environments [20][21][22][23][24].…”

Advancements in Image-Based Models for High-Grade Gliomas Might Be Accelerated

“…131 patients with HGG. [42] Use of the tissue permeability and microcirculation parameters Ktrans, Kep, IAUC to differentiate PT from TM.…”

“…In a cohort of patients diagnosed with HGG, radiomics and machine learning methodologies were reported as able to help predict the development of pseudoprogression from pre-treatment MR images, thus potentially allowing to reduce the use of biopsy and invasive histopathology [42] (Table 1).…”

Use Test of Automated Machine Learning in Cancer Diagnostics

“…Quantitative factors extracted from conventional MRI and CT images (so-called radiomic features) can be used in conjunction with machine learning algorithms to answer important diagnostic questions in the diagnosis of brain tumors in a fully automated, objective, highly precise, and, above all, non-invasive way. For example, Ari et al showed that radiomic-based machine learning can be used to non-invasively predict pseudoprogression in high-grade gliomas [2]. Krähling et al developed an MRI-based radiomics model to predict mitotic cycles in intracranial meningiomas before surgery [3].…”