Rivera Zoe Calulo, González-Seguel Felipe, Horikawa-Strakovsky Arimitsu, Granger Catherine, Sarwal Aarti, Dhar Sanjay, Ntoumenopoulos George, Chen Jin, Bumgardner V K Cody, Parry Selina M, Mayer Kirby P, Wen Yuan
Department of Physiotherapy, School of Health Sciences, The University of Melbourne, Melbourne, Australia.
Center for Muscle Biology, University of Kentucky, Lexington, KY, USA.
Sci Rep. 2025 Apr 29;15(1):14936. doi: 10.1038/s41598-025-99522-7.
Muscle ultrasound has high utility in clinical practice and research; however, the main challenges are the training and time required for manual analysis to achieve objective quantification of muscle size and quality. We aimed to develop and validate a software tool powered by artificial intelligence (AI) by measuring its consistency and comparability of expert manual analysis quantifying lower limb muscle ultrasound images. Quadriceps complex (QC) and tibialis anterior (TA) muscle images of healthy, intensive care unit, and/or lung cancer participants were captured with portable devices. Manual analyses of muscle size and quality were performed by experienced physiotherapists taking approximately 24 h to analyze all 180 images, while automated analyses were performed using a custom-built deep-learning model (MyoVision-US), taking 247 s (saving time = 99.8%). Consistency between the manual and automated analyses was good to excellent for all QC (ICC = 0.85-0.99) and TA (ICC = 0.93-0.99) measurements, even for critically ill (ICC = 0.91-0.98) and lung cancer (ICC = 0.85-0.99) images. The comparability of MyoVision-US was moderate to strong for QC (adj. R = 0.56-0.94) and TA parameters (adj. R = 0.81-0.97). The application of AI automating lower limb muscle ultrasound analyses showed excellent consistency and strong comparability compared with human analysis across healthy, acute, and chronic population.
肌肉超声在临床实践和研究中具有很高的实用性；然而，主要挑战在于手动分析所需的培训和时间，以实现对肌肉大小和质量的客观量化。我们旨在通过测量其在量化下肢肌肉超声图像方面与专家手动分析的一致性和可比性，来开发和验证一种由人工智能（AI）驱动的软件工具。使用便携式设备采集健康、重症监护病房和/或肺癌参与者的股四头肌复合体（QC）和胫骨前肌（TA）肌肉图像。由经验丰富的物理治疗师对肌肉大小和质量进行手动分析，分析所有180张图像大约需要24小时，而使用定制的深度学习模型（MyoVision-US）进行自动分析则需要247秒（节省时间 = 99.8%）。对于所有QC（组内相关系数ICC = 0.85 - 0.99）和TA（ICC = 0.93 - 0.99）测量，手动和自动分析之间的一致性良好至优秀，即使对于危重症患者（ICC = 0.91 - 0.98）和肺癌患者（ICC = 0.85 - 0.99）的图像也是如此。MyoVision-US对于QC（调整后R = 0.56 - 0.94）和TA参数（调整后R = 0.81 - 0.97）的可比性为中等至强。与健康、急性和慢性人群的人工分析相比，人工智能自动化下肢肌肉超声分析的应用显示出极好的一致性和很强的可比性。
