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.

The aim of this study was to investigate the potential of combined textural feature analysis of contrast-enhanced MRI (CE-MRI) and static O-(2-[18F]fluoroethyl)-L-tyrosine (FET) PET for the differentiation between local recurrent brain metastasis and radiation injury since CE-MRI often remains inconclusive. Fifty-two patients with new or progressive contrast-enhancing brain lesions on MRI after radiotherapy (predominantly stereotactic radiosurgery) of brain metastases were additionally investigated using FET PET. Based on histology (n = 19) or clinicoradiological follow-up (n = 33), local recurrent brain metastases were diagnosed in 21 patients (40%) and radiation injury in 31 patients (60%). Forty-two textural features were calculated on both unfiltered and filtered CE-MRI and summed FET PET images (20-40 min p.i.), using the software LIFEx. After feature selection, logistic regression models using a maximum of five features to avoid overfitting were calculated for each imaging modality separately and for the combined FET PET/MRI features. The resulting models were validated using cross-validation. Diagnostic accuracies were calculated for each imaging modality separately as well as for the combined model. For the differentiation between radiation injury and recurrence of brain metastasis, textural features extracted from CE-MRI had a diagnostic accuracy of 81% (sensitivity, 67%; specificity, 90%). FET PET textural features revealed a slightly higher diagnostic accuracy of 83% (sensitivity, 88%; specificity, 75%). However, the highest diagnostic accuracy was obtained when combining CE-MRI and FET PET features (accuracy, 89%; sensitivity, 85%; specificity, 96%). Our findings suggest that combined FET PET/CE-MRI radiomics using textural feature analysis offers a great potential to contribute significantly to the management of patients with brain metastases.

Over the past years, the quantity and complexity of imaging data available for the clinical management of patients with solid tumors has increased substantially. Without the support of methods from the field of artificial intelligence (AI) and machine learning, a complete evaluation of the available image information is hardly feasible in clinical routine. Especially in radiotherapy planning, manual detection and segmentation of lesions is laborious, time consuming, and shows significant variability among observers. Here, AI already offers techniques to support radiation oncologists, whereby ultimately, the productivity and the quality are increased, potentially leading to an improved patient outcome. Besides detection and segmentation of lesions, AI allows the extraction of a vast number of quantitative imaging features from structural or functional imaging data that are typically not accessible by means of human perception. These features can be used alone or in combination with other clinical parameters to generate mathematical models that allow, for example, prediction of the response to radiotherapy. Within the large field of AI, radiomics is the subdiscipline that deals with the extraction of quantitative image features as well as the generation of predictive or prognostic mathematical models. This review gives an overview of the basics, methods, and limitations of radiomics, with a focus on patients with brain tumors treated by radiation therapy.

Mutations in the isocitrate dehydrogenase (IDH mut) gene have gained paramount importance for the prognosis of glioma patients. To date, reliable techniques for a preoperative evaluation of IDH genotype remain scarce. Therefore, we investigated the potential of O-(2-[¹⁸F]fluoroethyl)-L-tyrosine (FET) PET radiomics using textural features combined with static and dynamic parameters of FET uptake for noninvasive prediction of IDH genotype. Prior to surgery, 84 patients with newly diagnosed and untreated gliomas underwent FET PET using a standard scanner (15 of 56 patients with IDH mut) or a dedicated high-resolution hybrid PET/MR scanner (11 of 28 patients with IDH mut). Static, dynamic and textural parameters of FET uptake in the tumor area were evaluated. Diagnostic accuracy of the parameters was evaluated using the neuropathological result as reference. Additionally, FET PET and textural parameters were combined to further increase the diagnostic accuracy. The resulting models were validated using cross-validation. Independent of scanner type, the combination of standard PET parameters with textural features increased significantly diagnostic accuracy. The highest diagnostic accuracy of 93% for prediction of IDH genotype was achieved with the hybrid PET/MR scanner. Our findings suggest that the combination of conventional FET PET parameters with textural features provides important diagnostic information for the non-invasive prediction of the IDH genotype.