Glottis segmentation is a crucial step to quantify endoscopic footage in laryngeal high-speed videoendoscopy. Recent advances in deep neural networks for glottis segmentation allow for a fully automatic workflow. However, exact knowledge of integral parts of these deep segmentation networks remains unknown, and understanding the inner workings is crucial for acceptance in clinical practice. Here, we show that a single latent channel as a bottleneck layer is sufficient for glottal area segmentation using systematic ablations. We further demonstrate that the latent space is an abstraction of the glottal area segmentation relying on three spatially defined pixel subtypes allowing for a transparent interpretation. We further provide evidence that the latent space is highly correlated with the glottal area waveform, can be encoded with four bits, and decoded using lean decoders while maintaining a high reconstruction accuracy. Our findings suggest that glottis segmentation is a task that can be highly optimized to gain very efficient and explainable deep neural networks, important for application in the clinic. In the future, we believe that online deep learning-assisted monitoring is a game-changer in laryngeal examinations.

No abstract

In the deep learning community, increasingly large models are being developed, leading to rapidly growing computational costs and energy costs. Recently, a new trend has been arising, advocating that researchers should also report the energy efficiency besides their model's performance in their papers. Previous research has shown that reduced precision can be helpful to improve energy efficiency. Based on this finding, we propose a simple practice to effectively improve the energy efficiency of training and inference, i.e., training the model with mixed precision and deploying it on Edge TPUs. We evaluated its effectiveness by comparing the speed-up of a state-of-the-art semantic segmentation architecture with respect to different typical usage scenarios, including using different devices, deep learning frameworks, model sizes, and batch sizes. Our results show that enabled mixed precision can gain up to a 1.9× speedup compared to the most common and default float32 data type on GPUs. Deploying the models on Edge TPU further boosted the inference by a factor of 6. Our approach allows researchers to accelerate their training and inference procedures without jeopardizing the model's accuracy, meanwhile reducing energy consumption and electricity cost easily without changing their model architecture or retraining. Furthermore, our approach is helpful in reducing the carbon footprint used to train and deploy the neural network and thus has a positive effect on environmental resources.INDEX TERMS Deep learning, green AI, energy efficiency, mixed precision training, quantization, edge TPU.