PET/MRI Radiomics in Patients With Brain Metastases

Abstract

Although a variety of imaging modalities are used or currently being investigated for patients with brain tumors including brain metastases, clinical image interpretation to date uses only a fraction of the underlying complex, high-dimensional digital information from routinely acquired imaging data. The growing availability of high-performance computing allows the extraction of quantitative imaging features from medical images that are usually beyond human perception. Using machine learning techniques and advanced statistical methods, subsets of such imaging features are used to generate mathematical models that represent characteristic signatures related to the underlying tumor biology and might be helpful for the assessment of prognosis or treatment response, or the identification of molecular markers. The identification of appropriate, characteristic image features as well as the generation of predictive or prognostic mathematical models is summarized under the term radiomics. This review summarizes the current status of radiomics in patients with brain metastases.

Standardization of imaging methods for machine learning in neuro-oncology

“…Whereas other techniques are generally based on predefined features, deep learning methods allow for automatic extraction of high-level features by using multiple layers of neural networks that imitate the functions of the human visual network. 3 , 48 While this technique has the advantage of not requiring image segmentation or predefined features, it also increases the difficulty in performing step-wise quality control due to the intrinsic nature of complex, high-integrated neural network layers. In addition, deep learning techniques often need significantly larger datasets compared to traditional machine learning techniques to reduce the chance of over-fitting due to a much greater number of features generated within the neural networks.…”

“…Furthermore, radiomics can sometimes be categorized into predefined feature-based techniques versus deep learning-based techniques that do not require predefinition or the intermediate feature extraction step. 3 , 4 Despite the promises of this developing field, most radiomic tools remain far from clinical implementation due to a number of challenges such as technical complexity, poor study design, overfitting of data, and lack of standards for validating results. 5 , 6 The purpose of this article is to review current efforts of standardization for neuroimaging, more specifically to discuss the advantages of standardized methodology and reporting within radiomic and neuro-oncology research.…”

Algorithm of traumatic brain injury management at Indonesia in the COVID 19 pandemic ERA. Retrospective cohort study

“…Next up is screening for COVID-19 patients who enter the emergency room. There are many guidelines that explain screening methods in neurosurgery patients in general, as well as neurotrauma in particular [ 6 , 9 ]. But here, we proposed our own scoring and screening algorithm that has been developed based on conditions in Indonesia ( Table 1 and Fig.…”

Radiomics in radiation oncology—basics, methods, and limitations

“…However, including information contained in the infiltration zone of the lesion by also considering signal abnormalities on T2-weighted or fluid-attenuated inversion recovery (FLAIR) MRI provides a more realistic representation of the whole tumor and allows the radiomics analysis to be performed on a larger segment, potentially encoding more information and resulting in a better diagnostic performance. Although the number of studies using amino acid PET images for radiomics analysis is still low, especially the combined analysis of amino acid PET and MRI radiomics encodes more diagnostic information than either modality alone [13,14] and might gain clinical relevance. In patients with brain tumors, image segmentation in clinical routine is usually performed manually on CT or MRI for the purpose of radiotherapy planning or volumetric assessment of therapy response.…”