Deep-learning based segmentation of ultrasound adipose image for liposuction

BackgroundTo develop an automatic and reliable ultrasonic visual system for robot- or computer-assisted liposuction, we examined the use of deep learning for the segmentation of adipose ultrasound images in clinical and educational settings. MethodsTo segment adipose layers, it is proposed to use an Attention Skip-Convolutions ResU-Net (Attention SCResU-Net) consisting of SC residual blocks, attention gates and U-Net architecture. Transfer learning is utilised to compensate for the deficiency of clinical data. The Bama pig and clinical human adipose ultrasound image datasets are utilized, respectively. ResultsThe final model obtains a Dice of 99.06 & PLUSMN; 0.95% and an ASD of 0.19 & PLUSMN; 0.18 mm on clinical datasets, outperforming other methods. By fine-tuning the eight deepest layers, accurate and stable segmentation results are obtained. ConclusionsThe new deep-learning method achieves the accurate and automatic segmentation of adipose ultrasound images in real-time, thereby enhancing the safety of liposuction and enabling novice surgeons to better control the cannula.