Gunaratne Rajitha, Monteath Isaac, Goncalves Joshua, Sheh Raymond, Ironside Charles N, Kapfer Michael, Chipper Richard, Robertson Brett, Khan Riaz, Fick Daniel
Curtin University, Kent Street, Bentley 6102, Australia.
Australian Institute of Robotic Orthopaedics, 2 Centro Avenue, Subiaco 6008, Australia.
Biomed Opt Express. 2019 Jul 11;10(8):3889-3898. doi: 10.1364/BOE.10.003889. eCollection 2019 Aug 1.
To assess if incorporation of DRS sensing into real-time robotic surgery systems has merit. DRS as a technology is relatively simple, cost-effective and provides a non-contact approach to tissue differentiation. Supervised machine learning analysis of diffuse reflectance spectra was performed to classify human joint tissue that was collected from surgical procedures. We have used supervised machine learning in the classification of a DRS human joint tissue data set and achieved classification accuracy in excess of 99%. Sensitivity for the various classes were; cartilage 99.7%, subchondral 99.2%, meniscus 100% and cancellous 100%. Full wavelength range is required for maximum classification accuracy. The wavelength resolution must be larger than 8nm. A SNR better than 10:1 was required to achieve a classification accuracy greater than 50%. The 800-900nm wavelength range gave the greatest accuracy amongst those investigated DRS is a viable method for differentiating human joint tissue and has the potential to be incorporated into robotic orthopaedic surgery.
评估将漫反射光谱（DRS）传感技术整合到实时机器人手术系统中是否具有价值。DRS技术相对简单、成本效益高，并且提供了一种非接触式的组织区分方法。我们对从手术过程中收集的人体关节组织的漫反射光谱进行了监督式机器学习分析，以进行分类。我们在DRS人体关节组织数据集的分类中使用了监督式机器学习，并实现了超过99%的分类准确率。各类别的灵敏度分别为：软骨99.7%、软骨下骨99.2%、半月板100%和松质骨100%。为了获得最大分类准确率，需要全波长范围。波长分辨率必须大于8nm。要实现大于50%的分类准确率，信噪比需优于10:1。在研究的波长范围中，800 - 900nm波长范围的准确率最高。DRS是一种区分人体关节组织的可行方法，并且有潜力整合到机器人骨科手术中。
