M. Sheikh, Urmi Kirtonia, Nuzath Tabassum Arthi, Md Al-Imran
The increasing use of artificial intelligence-generated deepfakes creates major challenges in maintaining digital authenticity. Four AI-based models, consisting of three CNNs and one Vision Transformer, were evaluated using large face image datasets. Data preprocessing and augmentation techniques im...
Md Al-Imran, Md Sifatullah Sheikh, Urmi Kirtonia, Nuzath Tabassum Arthi et al.
The increasing realism of GAN-generated facial forgeries has intensified the need for reliable deepfake detection systems capable of operating under diverse generative conditions. Traditional detectors often struggle to simultaneously capture global semantic structures and fine-grained local artifac...
Sajid Mahmud, Ankon Saha, Nuzath Tabassum Arthi, Musharrat Khan et al.
Manual review of brain MRI alone struggles to deliver timely, reliable tumor diagnosis. We study transfer-learning baselines and a simple ensemble for four-class MRI classification (glioma, meningioma, pituitary, no tumor). Using the BRISC 2025 T1-weighted corpus (6,000 images across axial, coronal,...
Ankon Saha, Sajid Mahmud, Nuzath Tabassum Arthi, Musharrat Khan et al.
Manual review of brain MRI image alone struggles for timely and reliable tumor diagnosis. We tested a custom ensemble framework for four-class MRI image classification (glioma, meningioma, pituitary, no tumor). On the BRISC 2025 T1-weighted corpus, we do a uniform preprocessing: grayscale and denois...