Poster Session

Start Date

16-5-2017 1:45 PM

Document Type

Poster Abstract

Description

Despite numerous research and development efforts that provide important tools to fight cancer, this disease still poses great challenges to diagnosis and treatment, and it remains one of the leading causes of death worldwide. Early diagnosis is crucial to increase the survival rate and quality of life of cancer patients. Thus, developing non-invasive screening methods would represent a key step towards point-of-care large scale screening and prevention of asymptomatic tumors such as Epithelian Ovarian Cancer (EOC) and others. Our group has developed two experimental strategies to pursue early cancer diagnosis through Laser-Induced Breakdown Spectroscopy (LIBS), a versatile atomic spectroscopy technique whose main advantages are: little or no sample preparation required; real-time multi-elemental response; virtually no limitation about the kind of sample that can be analyzed. The first is a LIBS-based immunoassay (Tag-LIBS), where a cancer biomarker is tagged with a suitably functionalized inorganic microparticles, which are in turn quantitatively and sensitively detected by LIBS. The second is based on the direct analysis of biological fluids through the combined use of LIBS and machine learning algorithms. By combining femtosecond LIBS with unsupervised classification techniques, we have shown that it is possible to discriminate blood samples extracted from healthy and diseased mice with an accuracy that approaches 80%. We will present our most recent results obtained with both approaches, and in particular we will report about the effects of various substrates used for LIBS measurements on the classification accuracy of blood samples extracted from cancerous and healthy mice.

Keywords

cancer, diagnosis, non-invasive screening methods, atomic spectroscopy, machine learning

Creative Commons License

Creative Commons Attribution-Noncommercial-Share Alike 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.

 
May 16th, 1:45 PM

Cancer Diagnosis using LIBS and Machine Learning Tools: Progress and Challenges

Despite numerous research and development efforts that provide important tools to fight cancer, this disease still poses great challenges to diagnosis and treatment, and it remains one of the leading causes of death worldwide. Early diagnosis is crucial to increase the survival rate and quality of life of cancer patients. Thus, developing non-invasive screening methods would represent a key step towards point-of-care large scale screening and prevention of asymptomatic tumors such as Epithelian Ovarian Cancer (EOC) and others. Our group has developed two experimental strategies to pursue early cancer diagnosis through Laser-Induced Breakdown Spectroscopy (LIBS), a versatile atomic spectroscopy technique whose main advantages are: little or no sample preparation required; real-time multi-elemental response; virtually no limitation about the kind of sample that can be analyzed. The first is a LIBS-based immunoassay (Tag-LIBS), where a cancer biomarker is tagged with a suitably functionalized inorganic microparticles, which are in turn quantitatively and sensitively detected by LIBS. The second is based on the direct analysis of biological fluids through the combined use of LIBS and machine learning algorithms. By combining femtosecond LIBS with unsupervised classification techniques, we have shown that it is possible to discriminate blood samples extracted from healthy and diseased mice with an accuracy that approaches 80%. We will present our most recent results obtained with both approaches, and in particular we will report about the effects of various substrates used for LIBS measurements on the classification accuracy of blood samples extracted from cancerous and healthy mice.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.