Название: Generative Machine Learning Models in Medical Image Computing
Автор: Le Zhang, Chen Chen, Zeju Li, Greg Slabaugh
Издательство: Springer
Год: 2025
Страниц: 373
Язык: английский
Формат: pdf
Размер: 29.7 MB

Generative Machine Learning Models in Medical Image Computing" provides a comprehensive exploration of generative modeling techniques tailored to the unique demands of medical imaging. This book presents an in-depth overview of cutting-edge generative models such as GANs, VAEs, and diffusion models, examining how they enable groundbreaking applications in medical image synthesis, reconstruction, and enhancement. Covering diverse imaging modalities like MRI, CT, and ultrasound, it illustrates how these models facilitate improvements in image quality, support data augmentation for scarce datasets, and create new avenues for predictive diagnostics.

Beyond technical details, the book addresses critical challenges in deploying generative models for healthcare, including ethical concerns, interpretability, and clinical validation. With a strong focus on real-world applications, it includes case studies and implementation guidelines, guiding readers in translating theory into practice. By addressing model robustness, reproducibility, and clinical utility, this book is an essential resource for researchers, clinicians, and data scientists seeking to leverage generative models to enhance biomedical imaging and deliver impactful healthcare solutions. Combining technical rigor with practical insights, it offers a roadmap for integrating advanced generative approaches in the field of medical image computing.

In the past decade, the advances in Deep Learning technologies have enabled their application to medical image segmentation, showing great potential. Nonetheless, the scarcity of available labelled data can result in a lack of model generalisability. This is especially true for supervised methods requiring annotated data. Data augmentation can be used to partially alleviate data scarcity when training Deep Learning models. In particular, the use of deep learning-based generative modelling, which allows for the sampling of synthetic data from the modelled datadistribution, has shown its potential for data augmentation in the past years. In this work, we address the topic of generative modelling to generate images and annotations, going over brainSPADE, a 2D and 3D generative model of healthy and pathological segmentations and corresponding multi-modal images for brain MRI, and how the synthetic data it produces can be applied to a range of segmentation tasks to mitigate the effects of data scarcity or domain shift.