Detecting Tumor Infiltration in Diffuse Gliomas with Deep Learning

Glioblastoma tumor recurrences often occur in brain tissue areas harboring infiltrating cells, resembling healthy imaging. Demarcating infiltrative regions for aggressive resections is critical improving prognostic outcomes but challenging neurosurgery. Herein, a multilayer sigmoid‐activated convolutional neural network (MLS‐CNN) developed rapidly distinguishing glioma infiltration histology. Unlike conventional multiclass classifiers, the MLS‐CNN employs sigmoidal activation to accommodate coexisting classes within patch images. 59 811 image patches (25 807 edge, 15 178 normal brain, 18 826 cellular tumor) from 73 samples are extracted train classifier. The model achieves an accuracy of 91.70% (sensitivity: 91.62%; specificity: 91.78%) and area under curve (AUC) 0.964 edges, outperforming current state‐of‐the‐art Vision Transformer (ViT) (accuracy: 89.45; AUC: 0.947). computationally efficient, generating predictions 11.5 s comparison 81.4 ViT. strongly correlate with In Situ Hybridization expression intensities, validating utility spatial genomics investigations gliomas. robust can therefore serve as automatic accurate classifier help pathologists identify better diagnosis patient oncology.