InvYNet: an inverted Y shape network for prostate cancer segmentation using prostate zones information

Motivation: Accurate segmentation of prostate cancer is vital for effective treatment, yet challenging due Goal(s): We aim to enhance prostate cancer segmentation by leveraging anatomical zones of the prostate. Approach: Our novel InvYNet model features two branches: an auxiliary branch for segmenting anatomical zones and a main branch that utilizes this information, enhanced by a Dual Attention Gate (DAG). Results: Extensive experiments on public datasets show that InvYNet outperforms existing models, achieving a 5% improvement over the second-place method on the PROSTATEx dataset. Impact: InvYNet is the first model to integrate anatomical zones for prostate cancer segmentation, setting a new standard in this field.