Researchers have proposed a new AI model called MambaX-Net for segmenting prostate tumors in MRI scans. The model is designed to handle longitudinal data, where multiple time points and limited expert labels make it challenging to train accurate models. MambaX-Net uses a combination of cross-attention and shape extraction modules to capture temporal evolution and spatial dependencies. It also employs a semi-supervised self-training strategy that leverages pseudo-labels generated from a pre-trained model. The authors claim that MambaX-Net outperforms state-of-the-art models on a longitudinal dataset, even when trained on limited data. --- Why it matters: This matters to researchers in AI because it provides a new approach for segmenting prostate tumors in MRI scans, which is an important step in diagnosing and monitoring cancer progression. The model's ability to handle longitudinal data and limited expert labels makes it a valuable tool for real-world applications. Source: https://arxiv.org/abs/2510.17529