Investigation of Energy-efficient AI Model Architectures and Compression Techniques for "Green" Fetal Brain Segmentation
This work addresses the environmental impact of AI in medical imaging, specifically for fetal brain segmentation, but is incremental as it applies known efficiency strategies to this domain.
The study tackled the problem of high energy consumption in training deep neural networks for fetal brain segmentation by exploring energy-efficient model architectures and compression techniques, achieving satisfactory performance with low energy consumption.
Artificial intelligence have contributed to advancements across various industries. However, the rapid growth of artificial intelligence technologies also raises concerns about their environmental impact, due to associated carbon footprints to train computational models. Fetal brain segmentation in medical imaging is challenging due to the small size of the fetal brain and the limited image quality of fast 2D sequences. Deep neural networks are a promising method to overcome this challenge. In this context, the construction of larger models requires extensive data and computing power, leading to high energy consumption. Our study aims to explore model architectures and compression techniques that promote energy efficiency by optimizing the trade-off between accuracy and energy consumption through various strategies such as lightweight network design, architecture search, and optimized distributed training tools. We have identified several effective strategies including optimization of data loading, modern optimizers, distributed training strategy implementation, and reduced floating point operations precision usage with light model architectures while tuning parameters according to available computer resources. Our findings demonstrate that these methods lead to satisfactory model performance with low energy consumption during deep neural network training for medical image segmentation.