AI and Cancer Research ── Brain ── 2026-10-25

Journals covered and how papers are chosen: see the index

Brain tumours: treatment, trials and AI

• Exploring supervised and unsupervised methods for brain tumor segmentation in MRI images
  Romanian Reports in Physics · 2026-10-22 · Evidence: Methodological comparison (image analysis)
  Abstract

  Accurate brain tumor segmentation from MRI images is crucial for diagnosis, treatment planning, and prognosis. In this study, we compare two fundamentally different approaches to tumor segmentation: a supervised deep learning model (2D U-Net) and an unsupervised clustering algorithm (K-means). We evaluate multiple variants of the 2D U-Net trained on different (Magnetic Resonance Imaging) MRI sequences (FLAIR and T1-Gd), as well as a fine-tuned model adapted across modalities, and compare them with K-means clustering applied separately on both modalities.

  Journal IF-equivalent: 2.4 (OpenAlex 2-year mean citedness, value as of 2026-10-23, retrieved 2026-10-24; not the official Clarivate IF)
  Reference: 1. Horia Hulubei National Institute for R&D in Physics and Nuclear Engineering, RO-077125, Măgurele-Bucharest, Romania 2. University of Bucharest, Faculty of Physics, RO-077125 Măgurele-Bucharest, Romania, BADEA RA, ORIAN AȘ, 1. Horia Hulubei National Institute for R&D in Physics and Nuclear Engineering, RO-077125, Măgurele-Bucharest, Romania 2. University of Bucharest, Faculty of Physics, RO-077125 Măgurele-Bucharest, Romania, VASILIU AA, University of Bucharest, Faculty of Physics, RO-077125 Măgurele-Bucharest, Romania, et al. Exploring supervised and unsupervised methods for brain tumor segmentation in MRI images. Rom Rep Phys. 2026 Dec 15;78(4):606-6. doi:10.59277/romrepphys.2026.78.606.
  Checked: Abstract only

← Back to this issue