Jagannadh Veerendra Kalyan, Gopakumar G, Subrahmanyam Gorthi R K Sai, Gorthi Sai Siva
Optics and Microfluidics Instrumentation Lab, Department of Instrumentation and Applied Physics, Indian Institute of Science, Malleshwaram, Bangalore, 560012, India.
Department of Earth and Space Sciences, Indian Institute of Space Science and Technology, Thiruvananthapuram, Kerala, 695547, India.
Med Biol Eng Comput. 2017 May;55(5):711-718. doi: 10.1007/s11517-016-1549-y. Epub 2016 Jul 22.
Each year, about 7-8 million deaths occur due to cancer around the world. More than half of the cancer-related deaths occur in the less-developed parts of the world. Cancer mortality rate can be reduced with early detection and subsequent treatment of the disease. In this paper, we introduce a microfluidic microscopy-based cost-effective and label-free approach for identification of cancerous cells. We outline a diagnostic framework for the same and detail an instrumentation layout. We have employed classical computer vision techniques such as 2D principal component analysis-based cell type representation followed by support vector machine-based classification. Analogous to criminal face recognition systems implemented with help of surveillance cameras, a signature-based approach for cancerous cell identification using microfluidic microscopy surveillance is demonstrated. Such a platform would facilitate affordable mass screening camps in the developing countries and therefore help decrease cancer mortality rate.
每年，全球约有700万至800万人死于癌症。超过一半的癌症相关死亡发生在世界较不发达地区。通过早期发现和后续治疗，癌症死亡率可以降低。在本文中，我们介绍了一种基于微流控显微镜的经济高效且无需标记的癌细胞识别方法。我们概述了针对该方法的诊断框架，并详细说明了仪器布局。我们采用了经典的计算机视觉技术，如基于二维主成分分析的细胞类型表示，随后基于支持向量机进行分类。类似于借助监控摄像头实现的刑事人脸识别系统，我们展示了一种基于微流控显微镜监测的基于特征的癌细胞识别方法。这样一个平台将有助于在发展中国家开展负担得起的大规模筛查活动，从而有助于降低癌症死亡率。
