Hepatocellular carcinoma is a serious threat to human health and life, so early diagnosis of hepatocellular carcinoma is particularly important. A new method based on FTIR spectroscopy and classification tree is proposed in this paper to develop a rapid and accurate diagnosis method for hepatocellular carcinoma. FTIR spectroscopy was firstly used to compare the spectra of hepatocellular carcinoma and normal tissues. The spectra of hepatocellular carcinoma and normal tissues have showed remarkable differences, which implied that the structure and compositions of hepatocellular carcinoma tissues have changed compared with those of normal tissues. 12 peak locations from both hepatocellular carcinoma tissues and normal tissues were analyzed and they had statistical differences by t test, or Wilcoxon rank test with a significance level of 0.05. Thus, peak locations were served as feature vectors for construction of diagnostic models based on classification tree. Diagnostic models based on classification tree were constructed and validated via a 10- fold cross validation method. The classification tree model based on Gdi split criterion achieved an accuracy of 99.24% for discrimination between hepatocellular carcinoma and normal tissue. The results demonstrated that FTIR spectroscopy combined with classification tree could be utilized for rapid and accurate diagnosis of hepatocellular carcinoma.
Terahertz time domain spectroscopy has been widely used in tumor detection, chemical analysis and nondestructive testing. However, the measurement errors of terahertz time domain spectrum frequently occur because of vibration of experiment instrument platform or temperature and humidity changes. Lifting wavelet transform based on different wavelet basis functions was applied to the denoising of terahertz time domain spectrum of PTFE. The denoising results were compared with denoising results of wavelet soft threshold method. The wavelet soft threshold method got a highest signal to noise ratio (SNR) of 58.75 dB and a least root mean square error (RMSE) of 3.56*10^ (-5), while lifting wavelet transform method achieved a highest SNR of 60.69 dB and a least RMSE of 2.85*10^ (-5). These results imply that lifting wavelet transform performs better in terahertz spectrum denoising than wavelet soft threshold.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.