Jordina Torrents-Barrena

TL;DR: A novel multi-task stacked generative adversarial framework is proposed to jointly learn synthetic fetal US generation, multi-class segmentation of the placenta, its inner acoustic shadows and peripheral vasculature, andplacenta shadowing removal and could be implemented in a TTTS fetal surgery planning software.

TL;DR: An auto-encoder based Generative Adversarial Network is adopted for synthetic fetal MRI generation that features a balanced power of the discriminator against the generator during training, provides an approximate convergence measure, and enables fast and robust training to generate high-quality fetal MRI in axial, sagittal and coronal planes.

TL;DR: The proposed TTTS planning software integrates all aforementioned algorithms to explore the intrauterine environment by simulating the fetoscope camera, determine the correct entry point, train doctors' movements ahead of surgery, and consequently, improve the success rate and reduce the operation time.

TL;DR: A fully-automated framework to achieve an accurate segmentation of the placenta and its peripheral vasculature in 3D US for twin-to-twin transfusion syndrome is proposed for the first time.

TL;DR: A semiautomatic algorithm to detect the placenta, both umbilical cord insertions and the placental vasculature from Doppler ultrasound and provides a near real-time user experience and requires short training without compromising the segmentation accuracy.