多模态超声联合深度学习对乳腺癌新辅助化疗疗效及侵袭性评价的研究进展

近年来，医学影像诊断的迅速发展，越来越多的超声成像技术被应用于乳腺癌的诊疗中，新辅助化疗是乳腺癌患者的标准治疗方案，能有效降低肿瘤临床分期，提高患者的保乳率，从而改善患者的预后。目前，主要的超声诊断方式包括常规超声、超声造影、超声弹性成像、三维超声等，联合神经网络深度学习技术，可更高效、精确地实现对乳腺癌患者新辅助化疗后疗效及乳腺癌侵袭性的评价。本文就乳腺癌NAC疗效的评价方法及侵袭性的评价指标、多模态超声联合深度学习对乳腺癌新辅助化疗疗效评价、多模态超声联合深度学习等方面的研究进展进行综述。 

Research progress of multimodal ultrasound combined with deep learning in the evaluation of the efficacy of neoadjuvant chemotherapy and invasiveness for breast cancer

In recent years, with the speedy evolvement of medical imaging diagnosis, a growing number of ultrasound imaging technologies have been used in the diagnosis and treatment of breast cancer. Neoadjuvant chemotherapy is the standard treatment plan for breast cancer patients, which can effectively reduce the clinical stage of tumors and improve the breastconserving rate of patients, thereby improving the prognosis of patients. At present, the main ultrasound diagnostic methods include conventional ultrasound, contrast-enhanced ultrasound, ultrasound elastography, three-dimensional ultrasound, et cetera, which combines with neural network deep learning technology. It can be more efficiently and accurately evaluate the curative effect of neoadjuvant chemotherapy and the evaluation of breast cancer invasiveness. This article reviews the research progress of multi-modal ultrasound combined with deep learning in the evaluation of the efficacy of neoadjuvant chemotherapy and invasiveness for breast cancer.