Díaz-Fernández et al. studied how the incorporation of intraoperative electrocorticography (ECoG) can distinguish tumor tissue from peritumoral and healthy cortex during the resection of gliomas. In 29 surgeries, the researchers recorded ECoG activity directly from electrodes placed over tumor, peritumoral, and healthy regions. They analyzed frequency bands and then correlated these with histological measures of tumor infiltration and applied machine learning classification models. They found that the tumoral cortex showed decreased activity across most frequency bands but increased delta power, alongside flatter spectral slopes and stronger local connectivity. By contrast, the peritumoral cortex exhibited increased beta activity and steeper 20–40 Hz slopes, features consistent with hyperexcitability, a known characteristic of peritumoral neuronal cells. Biopsies confirmed that higher levels of tumor cell infiltration corresponded with reduced physiological brain rhythms. Machine learning algorithms could classify tumoral versus non-tumoral tissue with specificity and sensitivity approaching 70%. The authors conclude that intraoperative electrocorticography is capable of recognizing distinct signatures for tumor and peritumoral regions with some degree of accuracy, suggesting potential for intraoperative mapping beyond standard imaging as advancements improve the capability of this technology.