Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
For many diseases, where a particular organ is affected, chemical by-products can be found in the patient's exhaled breath. Breath analysis is often done using gas chromatography and mass spectrometry, but interpretation of results is difficult and time-consuming. We performed characterization of patients' exhaled breath samples by an electronic nose technique based on an array of nanomechanical membrane sensors. Each membrane is coated with a different thin polymer layer. By pumping the exhaled breath into a measurement chamber, volatile organic compounds present in patients' breath diffuse into the polymer layers and deform the membranes by changes in surface stress. The bending of the membranes is measured piezoresistively and the signals are converted into voltages. The sensor deflection pattern allows one to characterize the condition of the patient. In a clinical pilot study, we investigated breath samples from head and neck cancer patients and healthy control persons. Evaluation using principal component analysis (PCA) allowed a clear distinction between the two groups. As head and neck cancer can be completely removed by surgery, the breath of cured patients was investigated after surgery again and the results were similar to those of the healthy control group, indicating that surgery was successful.
Kumar Varoon Agrawal, Kuang-Jung Hsu, Marina Micari, Xuekui Duan, Shuqing Song, Luis Francisco Villalobos Vazquez de la Parra, Shiqi Huang, Shaoxian Li, Heng-Yu Chi, Liping Zhong
Niels Quack, Dorian Giraud Herle
François Gallaire, Pier Giuseppe Ledda, Giuseppe Antonio Zampogna, Kevin Wittkowski