Haishun Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Haishun Liu has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Cunlin Zhang
Capital Normal Univ. (China)
1 shared publication- Cunlin Zhang
Capital Normal University
1 shared publication - Fan Wang
Department of Laboratory Medicine, Beijing Tongren Hospital Capital Medical University, Beijing 100730, China
1 shared publication - Xiangyi Liu
Department of Clinical Laboratory, Capital Medical University, Beijing Tongren Hospital
1 shared publication - Xin Zhang
Department of Chemistry, Capital Normal University, Beijing, China
1 shared publication - Yiding Han
Department of Pathology, Beijing Tongren Hospital Capital Medical University, Beijing 100730, China
1 shared publication
Latest Publications
- 9 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Terahertz Science and Technology
- Vol. 8 (3) , 271-277
Dimensionality Reduction for Identification of Hepatic Tumor Samples Based on Terahertz Time-Domain Spectroscopy
Terahertz time-domain spectroscopy (THz-TDS) combining with chemometrics methods was proposed for the identification of hepatic tumors. Two linear compression methods, principle component analysis and locality preserving projections (LPPs), and a nonlinear method, Isomap, were used to reduce the dimensionality of the measured dataset. Comparing two-dimensional (2-D) data reduced by these three dimensionality reduction techniques, only 2-D Isomap plot could separate the distances between two classes for the THz time-domain data and LPP had capacity of distinguishing two types of samples building on frequency-domain data. The best classification accuracies from 2-D time-domain data were ${\text{99.81}}\,\pm \,{\text{0.30}}\% $ and ${\text{99.69}}\,\pm \,{\text{0.61}}\% $ given by Isomap probabilistic neural network (PNN) and Isomap support vector machine (SVM), respectively, while the best classification results of 2-D frequency-domain data were ${\text{100.00}}\,\pm \,{\text{0.00}}\% $, ${\text{99.75}}\,\pm \,{\text{0.32}}\% $ provided by LPP-PNN, LPP-SVM. The results showed that Isomap and LPP are appropriate techniques to reflect the nonlinear manifold of the THz data. The THz technology either in time-domain or frequency-domain coupled with Isomap-PNN or LPP-PNN could offer a potential procedure to identify hepatic tumors.- 9 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Terahertz Science and Technology
- Vol. 8 (3) , 271-277
Dimensionality Reduction for Identification of Hepatic Tumor Samples Based on Terahertz Time-Domain Spectroscopy
Terahertz time-domain spectroscopy (THz-TDS) combining with chemometrics methods was proposed for the identification of hepatic tumors. Two linear compression methods, principle component analysis and locality preserving projections (LPPs), and a nonlinear method, Isomap, were used to reduce the dimensionality of the measured dataset. Comparing two-dimensional (2-D) data reduced by these three dimensionality reduction techniques, only 2-D Isomap plot could separate the distances between two classes for the THz time-domain data and LPP had capacity of distinguishing two types of samples building on frequency-domain data. The best classification accuracies from 2-D time-domain data were ${\text{99.81}}\,\pm \,{\text{0.30}}\% $ and ${\text{99.69}}\,\pm \,{\text{0.61}}\% $ given by Isomap probabilistic neural network (PNN) and Isomap support vector machine (SVM), respectively, while the best classification results of 2-D frequency-domain data were ${\text{100.00}}\,\pm \,{\text{0.00}}\% $, ${\text{99.75}}\,\pm \,{\text{0.32}}\% $ provided by LPP-PNN, LPP-SVM. The results showed that Isomap and LPP are appropriate techniques to reflect the nonlinear manifold of the THz data. The THz technology either in time-domain or frequency-domain coupled with Isomap-PNN or LPP-PNN could offer a potential procedure to identify hepatic tumors.- 12 January 2018
- Published by SPIE-Intl Soc Optical Eng
- Vol. 10623, 106230L
Removing the echoes from terahertz pulse reflection system and sample
Due to the echoes both from terahertz (THz) pulse reflection system and sample, the THz primary pulse will be distorted. The system echoes include two types. One preceding the main peak probably is caused by ultrafast laser pulse and the other at the back of the primary pulse is caused by the Fabry-Perot (F-P) etalon effect of detector. We attempt to remove the corresponding echoes by using two kinds of deconvolution. A Si wafer of 400μm was selected as the tested sample. Firstly, the method of double Gaussian filter (DGF) decnvolution was used to remove the systematic echoes, and then another deconvolution technique was employed to eliminate the two obvious echoes of the sample. The ultimate results indicated: although the combination of two deconvolution techniques could not entirely remove the echoes of sample and system, the echoes were largely reduced.
- 12 January 2018
- Published by SPIE-Intl Soc Optical Eng
- Vol. 10623, 106230L
Removing the echoes from terahertz pulse reflection system and sample
Due to the echoes both from terahertz (THz) pulse reflection system and sample, the THz primary pulse will be distorted. The system echoes include two types. One preceding the main peak probably is caused by ultrafast laser pulse and the other at the back of the primary pulse is caused by the Fabry-Perot (F-P) etalon effect of detector. We attempt to remove the corresponding echoes by using two kinds of deconvolution. A Si wafer of 400μm was selected as the tested sample. Firstly, the method of double Gaussian filter (DGF) decnvolution was used to remove the systematic echoes, and then another deconvolution technique was employed to eliminate the two obvious echoes of the sample. The ultimate results indicated: although the combination of two deconvolution techniques could not entirely remove the echoes of sample and system, the echoes were largely reduced.