Tumor Segmentation
Tumor segmentation, the automated identification and delineation of tumors in medical images, aims to improve diagnostic accuracy and treatment planning. Current research emphasizes robust segmentation across diverse imaging modalities (MRI, CT, PET) and scanners, often employing deep learning architectures like U-Net, Swin-UNet, and transformers, and addressing challenges such as missing modalities and domain shifts through techniques like knowledge distillation, multi-task learning, and data augmentation. These advancements hold significant promise for improving cancer diagnosis, treatment personalization, and ultimately, patient outcomes.
Papers
Towards SAMBA: Segment Anything Model for Brain Tumor Segmentation in Sub-Sharan African Populations
Mohannad Barakat, Noha Magdy, Jjuuko George William, Ethel Phiri, Raymond Confidence, Dong Zhang, Udunna C Anazodo
Bridging the Gap: Generalising State-of-the-Art U-Net Models to Sub-Saharan African Populations
Alyssa R. Amod, Alexandra Smith, Pearly Joubert, Confidence Raymond, Dong Zhang, Udunna C. Anazodo, Dodzi Motchon, Tinashe E. M. Mutsvangwa, Sébastien Quetin
Pre- to Post-Contrast Breast MRI Synthesis for Enhanced Tumour Segmentation
Richard Osuala, Smriti Joshi, Apostolia Tsirikoglou, Lidia Garrucho, Walter H. L. Pinaya, Oliver Diaz, Karim Lekadir
End-to-end autoencoding architecture for the simultaneous generation of medical images and corresponding segmentation masks
Aghiles Kebaili, Jérôme Lapuyade-Lahorgue, Pierre Vera, Su Ruan
Generating 3D Brain Tumor Regions in MRI using Vector-Quantization Generative Adversarial Networks
Meng Zhou, Matthias W Wagner, Uri Tabori, Cynthia Hawkins, Birgit B Ertl-Wagner, Farzad Khalvati
Iterative Semi-Supervised Learning for Abdominal Organs and Tumor Segmentation
Jiaxin Zhuang, Luyang Luo, Zhixuan Chen, Linshan Wu