AI and Cancer Research ── Brain ── 2026-11-09

Journals covered and how papers are chosen: see the index

Brain tumours: treatment, trials and AI

• Explainable DSC Perfusion Radiomics for Post-Treatment Glioblastoma Assessment Using a Soft-Voting Machine-Learning Ensemble
  Cancers · 2026-11-06 · Evidence: Retrospective observational (diagnostic model development on a public dataset with independent test-set validation)
  Abstract

  Background/Objectives: After chemoradiation for glioblastoma, residual or recurrent tumor (tumor-positive disease, TP) and treatment-related change (TRC), which includes pseudoprogression and radiation necrosis, look alike on conventional MRI. We tested whether radiomic features from dynamic susceptibility contrast (DSC) perfusion MRI alone can separate them.

  Methods: We analyzed 224 post-treatment MRI examinations from 168 patients in the public UCSD-PTGBM collection. A machine-learning ensemble was trained on 1223 radiomic features from the leakage-corrected relative cerebral blood volume (rCBV) map, with training (109 patients, 143 examinations) and test (59 patients, 81 examinations) data separated at the patient level and the model fixed before testing. The analysis was repeated on the cerebral blood flow (CBF) map from the same scans, and DSC was compared with arterial spin labeling (ASL) in patients who had both.

  Results: On the independent test set (patient-level analysis), the model distinguished TP from TRC with an area under the receiver operating characteristic curve (AUC) of 0.79 (95% CI 0.67–0.89). It detected 31 of 49 patients with tumor (sensitivity 63%) and correctly identified all 10 patients with TRC (specificity 100%); at a predefined higher-sensitivity cut-off, sensitivity was 69% and specificity 80%. With the CBF map, the AUC did not differ significantly (0.78; p = 0.89), but there was a different balance of sensitivity and specificity. In an exploratory subset of 36 patients, DSC and ASL did not differ significantly (AUC 0.82 vs. 0.71; p = 0.52).

  Conclusions: DSC perfusion radiomics alone identified tumor with high specificity but missed about one-third of tumors: a positive result may support a diagnosis of tumor, whereas a negative result cannot exclude it. External, multicenter validation is needed before clinical use.

  Journal IF-equivalent: 6.1 (OpenAlex 2-year mean citedness, value as of 2026-11-07, retrieved 2026-11-08; not the official Clarivate IF)
  Reference: Christodoulou RC, Christofi G, Vamvouras G, Esmeraldo MA, Papageorgiou PS, Vassiliou E, et al. Explainable DSC Perfusion Radiomics for Post-Treatment Glioblastoma Assessment Using a Soft-Voting Machine-Learning Ensemble. Cancers. 2026 Nov 6;18(21):3567. doi:10.3390/cancers18213567.
  Checked: Abstract only
• Brain Tumor Classification Accuracy Enhancement via DenseNet201 Compared to Four Alternative Models
  ELCVIA Electronic Letters on Computer Vision and Image Analysis · 2026-11-05 · Evidence: Retrospective model development and comparison using an MRI image dataset (no clinical validation)
  Abstract

  Early brain tumor staging is critical to the success of currently available treatment planning and modalities. Magnetic Resonance Imaging (MRI) is a necessary method for treatment planning and early diagnosis. Most current classification methods do not achieve a high level of classification accuracy, which may result in some tumor cases being misclassified or classified as suspicious. In our proposed work, we leverage the DenseNet201 deep learning architecture as the foundation for classifying brain tumors. A dataset in this methodology consisted of MRI scans, which included four unique types of tumors: meningioma, pituitary tumor, no tumor, and glioma. To improve classification precision, improve generalization, and reduce overfitting, we employed three strategies (a1, a2, and a3) during the classification process. When our model was evaluated on unseen test data, it achieved an impressive 99.77\% accuracy. The proposed technique effectively classifies malignant brain tumors, despite the similarities within classes and the varying tumor morphologies. There are four state-of-the-art techniques of deep learning that we compared our results with: ResNetRS50, ResNet50, VGG19, and EfficientNetV2, which achieved accuracies of 99.77\%, 98.71\%, 96.88\%, and 96.58\%, respectively. In the field of brain tumor classification, our approach has outperformed others in terms of performance, highlighting its potential for clinical applications.

  Journal IF-equivalent: 1.2 (OpenAlex 2-year mean citedness, value as of 2026-11-07, retrieved 2026-11-08; not the official Clarivate IF)
  Reference: Thamer Z, Sagheer N, Ata O, Abdulmunem A, Thamer H. Brain Tumor Classification Accuracy Enhancement via DenseNet201 Compared to Four Alternative Models. ELCVIA. 2026 Nov 5;25(3):65-80. doi:10.5565/rev/elcvia.2305.
  Checked: Abstract only

← Back to this issue