Deep learning (DL) has been applied to glioblastoma (GBM) magnetic resonance imaging (MRI) assessment for tumor segmentation and inference of molecular, diagnostic, and prognostic information. We comprehensively overviewed the currently available DL applications, critically examining the limitations that hinder their broader adoption in clinical practice and molecular research. Technical limitations to the routine application of DL include the qualitative heterogeneity of MRI, related to different machinery and protocols, and the absence of informative sequences, possibly compensated by artificial image synthesis. Moreover, taking advantage from the available benchmarks of MRI, algorithms should be trained on large amounts of data. Additionally, the segmentation of postoperative imaging should be further addressed to limit the inaccuracies previously observed for this task. Indeed, molecular information has been promisingly integrated in the most recent DL tools, providing useful prognostic and therapeutic information. Finally, ethical concerns should be carefully addressed and standardized to allow for data protection. DL has provided reliable results for GBM assessment concerning MRI analysis and segmentation, but the routine clinical application is still limited. The current limitations could be prospectively addressed, giving particular attention to data collection, introducing new technical advancements, and carefully regulating ethical issues.

Background: Fluorescence-guided resection (FGR) of glioblastomas has been previously explored with the use of 5-amivelulinic acid (5-ALA) and sodium fluoresceine (SF), allowing us to maximize the extent of resection (EoR). In this study, we highlight the most relevant concerns regarding this technique and present the methods and results from the experience of our center. Methods: A case series of 100 patients operated on in AOU Città della Salute e della Scienza in Turin with a histological diagnosis of glioblastoma (grade IV, according to WHO 2021) was retrospectively analyzed. Both 5-ALA and SF were administered and intraoperatively assessed with an optical microscope. Results: 5-ALA is the only approved drug for FGR in glioblastoma, reporting an increased EoR. Nevertheless, SF can be positively used in addition to 5-ALA to reduce the risk of false positives without increasing the rate of adverse effects. In our experience, SF was used to guide the initial phase of resection while 5-ALA was used to visualize tumor spots within the surgical cavity. In 96% of cases, gross total resection was achieved, with supra-maximal resection in 11% of cases. Conclusions: Combined FGR using 5-ALA and SF seems to be a promising method of increasing the extent of resection and to improving the prognosis in glioblastoma patients.