Rapid Identification and Quality Evaluation of Medicinal Centipedes in China Using Near-Infrared Spectroscopy Integrated with Support Vector Machine Algorithm

Abstract

To investigate the feasibility of rapid identification and quality evaluation of Chinese medicinal centipedes using NIR spectroscopy, the qualitative and quantitative analysis models were explored. A PCA-SVC model was optimized to differentiate five species of the genus Scolopendra . When the model was validated with the calibration and prediction sets, the prediction accuracy was 100% and 81.82%, respectively; it can meet the requirement for rapid and preliminary identification. Based on nitrogen content detected by the chemical method, and the dimensionality of spectral data reduced with PLS, the quantitative analysis models were successfully built by PLSR and SVR algorithms. After spectra were pretreated and parameters were optimized, the performance, rationality, and prediction ability of the models were validated and evaluated with RMSECV, RMSEP, RMSEE, R² , and RPD. Compared with the features and advantages of these two models, the PLS-SVR model had better performance and stronger prediction capacity, and it was finally regarded as the optimal quantitative analysis model to predict nitrogen content. The relative deviation between the predictive value and the reference was 2.69%, and the average recovery was 99.02%, which indicated it has potential for rapid prediction and evaluation of the quality of medicinal centipedes. This research suggested that NIR spectroscopy can be used as a rapid detection method to identify species and evaluate the quality of medicinal centipedes in China.

Full text

Research Article
Rapid Identification and Quality Evaluation of Medicinal
Centipedes in China Using Near-Infrared Spectroscopy
Integrated with Support Vector Machine Algorithm
Sihe Kang ,
1
,
2
Haiying Deng ,
3
Long Chen ,
1
,
4
Xiaoxuan Zeng,
1
Yimei Liu ,
1
and
Keli Chen
1
1
Key Laboratory of Ministry of Education on Traditional Chinese Medicine Resource and Compound Prescription,
Hubei University of Chinese Medicine, Wuhan 430065, China
2
NMPA Key Laboratory of Quality Control of Chinese Medicine, Hubei Institute for Drug Control, Wuhan 430075, China
3
Medical College of Wuhan University of Science and Technology,
Hubei Province Key Laboratory of Occupational Hazard Identification and Control, Wuhan 430065, China
4
Xiangyang Central Hospital, Xiangyang 441021, China
Correspondence should be addressed to Yimei Liu; liuyimei1971@126.com and Keli Chen; kelichen@126.com
Received 14 May 2019; Revised 23 July 2019; Accepted 2 August 2019; Published 22 September 2019
Academic Editor: Eugen Culea
Copyright ©2019 Sihe Kang et al. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
To investigate the feasibility of rapid identification and quality evaluation of Chinese medicinal centipedes using NIR spec-
troscopy, the qualitative and quantitative analysis models were explored. A PCA-SVC model was optimized to differentiate five
species of the genus Scolopendra. When the model was validated with the calibration and prediction sets, the prediction accuracy
was 100% and 81.82%, respectively; it can meet the requirement for rapid and preliminary identification. Based on nitrogen
content detected by the chemical method, and the dimensionality of spectral data reduced with PLS, the quantitative analysis
models were successfully built by PLSR and SVR algorithms. After spectra were pretreated and parameters were optimized, the
performance, rationality, and prediction ability of the models were validated and evaluated with RMSECV, RMSEP, RMSEE, R
2
,
and RPD. Compared with the features and advantages of these two models, the PLS-SVR model had better performance and
stronger prediction capacity, and it was finally regarded as the optimal quantitative analysis model to predict nitrogen content. e
relative deviation between the predictive value and the reference was 2.69%, and the average recovery was 99.02%, which indicated
it has potential for rapid prediction and evaluation of the quality of medicinal centipedes. is research suggested that NIR
spectroscopy can be used as a rapid detection method to identify species and evaluate the quality of medicinal centipedes in China.
1. Introduction
Animals of the genus Scolopendra are widely distributed in
the world, especially in tropical and subtropical areas [1]. In
China, there are 14 species which are mainly distributed in
the southern region [2]. Recently, the medicinal value of
centipedes had become a research hotspot; the venom was
reported to be used for relieving pain and anticoagulation
[3, 4]. As an important source of Chinese medicinal ma-
terials, five species of Scolopendra are commonly used in
China [2], which were reported to possess analgesic [3], anti-
inflammatory [5], and antitumor [6, 7] activities and to
improve blood rheology [8]. However, Scolopendra mutilans
is the only species recorded in Chinese Pharmacopoeia 2015
(ChP 2015), and the other four species are just used in local
regions; for instance, S. multidens is used in Guangxi and
S. mojiangica in Yunnan [2]. As the animals of Scolopendra
are poisonous, the venom and toxic ingredients of some
species will bring high risk to humans [9], and the activity
and quality of species are different. erefore, in order to
ensure safety and clinical efficacy, a simple, rapid, and
accurate method is needed to identify the species and
control the quality of medicinal centipedes.
Previously, medicinal centipedes were mostly identified
using morphological description, but some similar charac-
teristics were probably shown among closely related species.
Hindawi
Journal of Spectroscopy
Volume 2019, Article ID 9636823, 11 pages
https://doi.org/10.1155/2019/9636823

If samples were damaged or powdered, they were difficult to
be identified, and confusion and misuse would be un-
avoidable. Presently, molecular methods are gradually ap-
plied to identify Scolopendra species [10]. However, the
complexity of operation and high technical requirements
make it difficult to obtain rapid and accurate results, es-
pecially in mixed samples. Proteins or amino acids are
recognized as the main active ingredient of medicinal
centipedes [11–13], and their content is usually used for
quality evaluation. Because of the diversity of ingredients,
and the complexity of chemical determination methods, this
measurement is usually cumbersome [14].
Near-infrared (NIR) spectroscopy combined with che-
mometrics is a fast, nondestructive, and environmentally
friendly analysis technique that can realize multicomponent
analysis. Nowadays, it is widely used in agriculture and
medicine [15–17]. NIR spectroscopy mainly reflects the
absorption of overtone and combination peaks containing
hydrogen bonds of C-H, O-H, and N-H [16]. Lipids and
proteins are considered to be important medicinal com-
ponents, which are rich in centipedes and show charac-
teristic absorption in the near-infrared region. However, the
application of NIR spectroscopy on medicinal centipedes
has not yet been reported. In this study, the NIR spec-
troscopy analysis methods were investigated, a PCA-SVC
model was explored to identify the species of Scolopendra,
and in light of nitrogen content determined by the chemical
method, a quantitative model was established for quality
prediction using regression algorithms.
2. Materials and Methods
2.1. Instruments and Software. e nitrogen content of
samples was determined with the DK 20 Heating Digester
(VELP, Italy) and UDK 149 Automatic Distillation Unit
(VELP, Italy). Spectra were collected with an MPA FT-NIR
spectrometer (Bruker Optics Co., Ltd., Germany) and an-
alyzed using the OPUS 7.5 spectrum analysis software
(Bruker), MATLAB R2014a data analysis software (Math-
Works, Inc., USA), and Unscrambler 9.7 data analysis
software (CAMO Software AS, Norway).
2.2. Samples and Identification. A total of 64 samples from
28 batches have been collected from field surveys or market
commodity in China since 2015. All samples were identified
into five nominal species according to characteristics
recorded by Siriwut et al. [1], Kang et al. [2], Song et al. [18],
and Zhang and Wang [19]. e sample information is
summarized in Table 1. All samples were kept below −20°C
and housed in Hubei University of Chinese Medicine,
Wuhan, China.
2.3. Content Determination. After being scanned with a
near-infrared spectrometer, the nitrogen content of 50 mg
powder of each sample was determined with the semimicro
quantitative nitrogen determination method referring to the
guideline of ChP 2015. e samples were digested using the
DK 20 Heating Digester with a program as follows: 200°C for
5 min, then up to 260°C sustaining for 5 min, 340°C for
5 min, and 420°C for 40 min, and at last cooled down to
200°C. e sample solution was measured using the UDK
149 Automatic Distillation Unit with a program as follows:
50 ml H
2
O and 20 ml 40% NaOH were added to the digested
solution, 20 ml 2% H
3
PO
4
was used for receiving, the steam
quantity was 50%, the distillation time was 4 min, and then
titration was done with 0.025 mol/L H
2
SO
4
standard solu-
tion (Metrological Testing Technology Research Institute of
Shanghai; Batch number 150901).
2.4. Spectra Acquisition. After samples were smashed and
dried at 55°C for 24 h, the powder of 2 g of individuals was
scanned using the MPA FT-NIR spectrometer with a diffuse
reflection integral sphere. e spectra were obtained in a
range of 12000∼4000 cm
−1
by the coaddition of 32 scans at a
resolution of 8 cm
−1
. Each sample was scanned three times,
and the average of three spectra was used for analysis. e
spectra diagram is shown in Figure 1.
2.5. Spectral Pretreatment Method. Usually, the raw spec-
trum includes a lot of irrelevant information or noise, which
would lead to baseline drift and instability. erefore,
spectrum pretreatment is a critical step in spectral analysis.
ere are many pretreatment methods, and each has ad-
vantages to improve model performance. For instance,
vector normalization (VN) can be used to eliminate in-
fluences of the optical path change on the spectrum. e
derivative methods including the first derivative (FD) and
second derivative (SD) are always employed to eliminate
spectral difference from baseline [20], while multiple scat-
tering correction (MSC) is commonly performed to process
diffuse reflection spectra [21]. In this study, methods such as
VN, FD, SD, and MSC or combined pretreatments were
employed by OPUS to optimize model performance.
2.6. Spectral Data Compression Method
2.6.1. Principal Component Analysis (PCA) Method. PCA is
a commonly used method for data compression. It performs
dimensionality reduction of a high-dimensional dataset,
while retaining its variation as much as possible. is
method can transform a number of possibly correlated
variables (the original data matrix) into one or a few im-
portant variables (principal components (PCs)) to reveal the
internal structure. Each PC is a linear combination of the
original data. e new variables are not related to each other,
which can eliminate the overlapped part of information.
Moreover, these new variables include the most informative
dimensions of the original variables without losing too much
information. Commonly, the number of PCs is determined
by the contribution rate to original variables. When the
cumulative contribution rate is more than 85%, the main
components can represent most of the information provided
by the original variable [22]. In our identification research,
the PCA was used to reduce the dimension of original
spectral data.
2Journal of Spectroscopy

Table 1: Sample information of medicinal centipedes.
Number Species Batch no. Nitrogen content (%) Origin
1S. mutilans L. Koch WG 002-1 10.09 Suizhou, Hubei
2S. mutilans L. Koch WG 003-1 11.36 Suizhou, Hubei
3S. mutilans L. Koch WG 004-1 11.82 Jingmen, Hubei
4S. mutilans L. Koch WG 004-2 10.60 Jingmen, Hubei
5S. mutilans L. Koch WG 005-1 10.77 Xiangyang, Hubei
6S. mutilans L. Koch WG 005-2 10.49 Xiangyang, Hubei
7S. mutilans L. Koch WG 006-1 9.47 Yichang, Hubei
8S. mutilans L. Koch WG 013-1 10.43 Suizhou, Hubei
9S. mutilans L. Koch WG 014-1 9.25 Jinshan, Hubei
10 S. mutilans L. Koch WG 014-2 10.22 Jinshan, Hubei
11 S. mutilans L. Koch WG 016-1 10.10 Suizhou, Hubei
12 S. mutilans L. Koch WG 016-2 9.83 Suizhou, Hubei
13 S. mutilans L. Koch WG 017-1 8.20 Anlu, Hubei
14 S. mutilans L. Koch WG 017-2 10.15 Anlu, Hubei
15 S. mutilans L. Koch WG 018-1 11.02 Yichang, Hubei
16 S. mutilans L. Koch WG 019-1 10.05 Nanzhang, Hubei
17 S. mutilans L. Koch WG 019-2 10.16 Nanzhang, Hubei
18 S. mutilans L. Koch WG 020-1 9.06 Anhui
19 S. mutilans L. Koch WG 020-2 11.74 Anhui
20 S. mutilans L. Koch WG 027-1 10.55 Henan
21 S. mutilans L. Koch WG 027-2 11.10 Henan
22 S. mutilans L. Koch WG 032 -1 9.96 Machang, Hubei
23 S. mutilans L. Koch WG 032-2 10.01 Machang, Hubei
24 S. mutilans L. Koch WG 045-1 9.68 Machang, Hubei
25 S. mutilans L. Koch WG 045-2 9.68 Machang, Hubei
26 S. multidens Newport WG 012-1 11.27 Yulin, Guangxi
27 S. multidens Newport WG 012-2 10.83 Yulin, Guangxi
28 S. multidens Newport WG 012-3 10.54 Yulin, Guangxi
29 S. multidens Newport WG 012-4 11.74 Yulin, Guangxi
30 S. multidens Newport WG 012-5 11.33 Yulin, Guangxi
31 S. multidens Newport WG 012-6 9.77 Yulin, Guangxi
32 S. multidens Newport WG 012-7 11.27 Yulin, Guangxi
33 S. multidens Newport WG 012-8 11.94 Yulin, Guangxi
34 S. multidens Newport WG 021-1 9.85 Guangxi
35 S. multidens Newport Wg039-1 11.92 Guangxi
36 S. multidens Newport Wg039-2 10.98 Guangxi
37 S. multidens Newport Wg040-1 10.25 Mengzi, Yunnan
38 S. multidens Newport Wg040-2 9.81 Mengzi, Yunnan
39 S. dehaani Brandt WG 028-1 12.31 Yunnan
40 S. dehaani Brandt WG 038-1 12.62 Yunnan
41 S. dehaani Brandt WG 038-2 12.17 Yunnan
42 S. dehaani Brandt WG 038-3 12.58 Yunnan
43 S. dehaani Brandt WG 038-4 12.30 Yunnan
44 S. dehaani Brandt WG 038-5 12.36 Yunnan
45 S. dehaani Brandt WG 038-6 11.67 Yunnan
46 S. dehaani Brandt WG 038-7 12.46 Yunnan
47 S. dehaani Brandt WG 038-8 11.93 Yunnan
48 S. mojiangica Zhang et Chi WG 007-1 8.37 Mojiang, Yunnan
49 S. mojiangica Zhang et Chi WG 007-2 8.04 Mojiang, Yunnan
50 S. mojiangica Zhang et Chi WG 007-3 8.16 Mojiang, Yunnan
51 S. mojiangica Zhang et Chi WG 007-4 7.47 Mojiang, Yunnan
52 S. mojiangica Zhang et Chi WG 007-5 8.18 Mojiang, Yunnan
53 S. mojiangica Zhang et Chi WG 008-1 8.92 Mojiang, Yunnan
54 S. mojiangica Zhang et Chi WG 008-2 9.06 Mojiang, Yunnan
55 S. mojiangica Zhang et Chi WG 041-2 8.54 Bixi, Yunnan
56 S. mojiangica Zhang et Chi WG 041-3 8.37 Bixi, Yunnan
57 S. negrocapitis Zhang et Wang WG 022-1 10.48 Suizhou, Hubei
58 S. negrocapitis Zhang et Wang WG 022-2 10.07 Suizhou, Hubei
59 S. negrocapitis Zhang et Wang WG 022-3 10.52 Suizhou, Hubei
60 S. negrocapitis Zhang et Wang WG 022-4 10.65 Suizhou, Hubei
Journal of Spectroscopy 3

2.6.2. Partial Least-Squares (PLS) Method. e PLS is a new
multivariate statistical analysis method. It attempts to
recombine the original variables (mainly continuous vari-
ables) into a group of new independent comprehensive
variables and extracts a few comprehensive variables to
reflect the information on the original variables as much as
possible. e extracted new variables have good in-
terpretation ability for the dependent variables. During
modeling, it not only considers factors of the independent
variable matrix (spectral matrix) but also takes the “re-
sponse” matrix (content matrix) into account. e principal
component scores extracted by dimension reduction are
used as input variables to avoid multicollinearity, improve
stability, and simplify the model. erefore, the PLS has the
ability to simplify the model and characteristics of quick
calculation and strong prediction ability, and as one of the
most classical data processing tools in multiple correlation
regression, it is widely applied in NIR spectroscopy quan-
titative analysis [23–25].
2.7. Support Vector Machine (SVM) Algorithm. SVM is a
powerful supervised learning algorithm that was first pro-
posed by Vapnik [26] and successfully extended by re-
searchers in recent years. It is based on the principle of
minimization of structural risk in constructing an optimally
separating hyperplane that separates different classes of data.
In the process, input vectors are mapped to a newly con-
structed high-dimensional space, and then parallel hyper-
planes are constructed to maximize the interplane distance
which separates the data. e SVM can solve nonlinear
problems in a higher-dimensional space based on radical
basis function (RBF) to construct a linear function. Usually,
the SVM algorithm includes both support vector machine
for classification (SVC) and support vector machine for
regression (SVR); the former is used to solve problems of
classification, while the latter is used for regression analysis.
RBF is a commonly used kernel function in the SVM
algorithm. It has a strong ability to deal with nonlinear
problems. It can be expressed as follows:
K(x, y) � exp −g|x−y|2
􏼐 􏼑, g >0.(1)
RBF has two important parameters in the SVM algorithm,
i.e., penalty factor “C” and kernel function parameter “g,”
which have a great influence on model prediction ability, and
the values should be determined during the model optimi-
zation process. Commonly, the optimization methods include
the grid search (GS), particle swarm optimization (PSO), and
genetic algorithm (GA). e PSO algorithm simulates the
flight foraging behaviors of bird clusters through collabora-
tion among birds to achieve the best objective. e GA is an
operation based on biological natural selection and genetic
mechanism to realize the optimal result, while the GS iterates
through every intersection in the grid to find the best
combination of parameters (C,g) and makes cross-validation
most accurate. e RMSECV was used to guide the opti-
mization of internal parameters.
During modeling of the SVM algorithm, the input data
need to be mapped to a higher-dimensional space to realize
Table 1: Continued.
Number Species Batch no. Nitrogen content (%) Origin
61 S. negrocapitis Zhang et Wang WG 022-5 11.45 Suizhou, Hubei
62 S. negrocapitis Zhang et Wang WG 015-1 10.86 Chaohu, Anhui
63 S. negrocapitis Zhang et Wang WG 015-2 11.71 Chaohu, Anhui
64 S. negrocapitis Zhang et Wang WG 015-3 10.82 Chaohu, Anhui
4000500060007000800090001000011000
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Absorbance
Wavenumber (cm–1)
Figure 1: NIR spectra diagram of samples.
4Journal of Spectroscopy

dimension reduction and regression fitting. So, the data
should be firstly pretreated and compressed.
2.8. Model Validation and Evaluation
2.8.1. Qualitative Model. In the PCA-SVC qualitative
model, the model performance was evaluated by 3-fold
cross-validation (3-CV) of the calibration set. e internal
parameters Cand gwere optimized with the GS method.
When the accuracy of 3-CV reached maximum values, the
optimal Cand gwere determined. After models were
established, the calibration set and prediction set were input
to the model, and the prediction accuracies were used as
indexes to evaluate the prediction ability.
2.8.2. Quantitative Model. During the process of modeling,
the calibration set was used for internal cross-validation to
validate model performance, the internal cross-validation
adopted 6-fold cross-validation, and the root mean square
error of internal cross-validation (RMSECV), coefficient of
determination (R
2
), and ratio of performance to deviation
(RPD) were taken to guide the model optimization process.
e predication set was used for external validation to
evaluate the model, with the root mean square error of
prediction (RMSEP), R
2
, and RPD taken as indexes to
evaluate prediction ability. Generally, the smaller the
RMSECV and the larger the R
2
are, the better the model
performance would be; the smaller the RMSEP and the
greater the R
2
are, the stronger the model prediction ability
is. Moreover, when RPD >2, it indicates that the model has
excellent reliability. After the model was established, the
calibration set used for full cross-validation was input to the
model again, and the root mean square error of evaluation
(RMSEE) was used as an index to further evaluate the
reasonability of the model. eoretically, the RMSECV value
was higher than the RMSEE value, which indicated that the
modeling process was reasonable and feasible.
3. Results and Discussion
3.1. Determination of Nitrogen Content. e content of 64
samples was measured. Samples used in the analysis are as
follows: 25 specimens of S. mutilans, 13 of S. multidens, 9 of
S. mojiangica, 8 of S. negrocapitis, and 9 of S. dehaani. e
nitrogen content of species was between 8.19% and 12.27%,
the total mean was 10.0%, and the mean value of each species
was 10.25%, 11.04%, 8.19%, 10.58%, and 12.27%, re-
spectively. e results are shown in Table 1.
3.2. Analysis of NIR Spectra. e NIR spectra of samples
were scanned in the range of 12000–4000 cm
−1
; the spectra
diagram is shown in Figure 1. It indicated that the char-
acteristic wavenumber was mainly in the range of 9000 to
4000 cm
−1
, while the spectral characteristics showed high
similarity among samples that it was difficult to distinguish
species from peak data. Hence, the chemometric method was
needed for spectral pretreatment and characteristic in-
formation extraction in qualitative and quantitative analysis.
3.3. Qualitative Model Based on PCA-SVC Algorithm
3.3.1. Sample Classification. e spectra of 64 samples were
randomly classified into calibration and prediction sets in a
proportion of approximately 2 : 1. Finally, 42 samples of the
calibration set were used for model establishment, and 22
samples of the prediction set were used for model evaluation.
e species were represented with category label numbers 1
to 5. e classification information is shown in Table 2.
3.3.2. Optimization of Pretreatment. In this qualitative
analysis, the three methods VN, FD, and SD were used to
pretreat the raw spectra. e PCA method was used to
reduce dimensions of raw and three pretreated spectra. e
accumulative contribution rates of PCs were calculated. e
result showed that the contribution rates of the first two PCs
(PC1 and PC2) were more than 85%, which can represent
most of the spectrum information [22]. Hence, the PC1-PC2
correlation diagram was attempted for a preliminary in-
vestigation to differentiate the samples. However, most
species overlapped together in space and cannot be dis-
criminated, except S. mojiangica in the FD and SD.
To further investigate the influence of different pre-
treatments, a group of PCA-SVC models was established
using the scores of the first 2 PCs as input variables and
category labels as output variables. e model performance
was evaluated by 3-fold cross-validation (3-CV) of the cali-
bration set. e internal parameters of the SVC algorithm
were optimized with the GS method. e values of best Cand
gwere obtained based on the initial optimization in a range of
log 2 c∈[0,50]and log 2 g∈[0,50]and in steps of 5 and
then the second fine optimization in an adjusted narrow
scope. When the 3-CV accuracy reached the maximum, the
optimal values of Cand gwere determined. After models
were established, the calibration set and prediction set were
input to models and predicted, and the prediction accuracies
were used to evaluate the prediction ability. As shown in
Table 3, it can be seen that “overfitting” and mismatching
existed in the raw spectrum model for high accuracy (90.48%)
in the calibration set and low accuracy (59.09%) in the
prediction set. In contrast, the other three models VN, FD,
and SD had nearer accuracies between calibration and pre-
diction sets to possess a relatively rational structure, although
the values were not even high. Also, the model of the SD had
the highest accuracy among all pretreatments, whether in the
calibration or prediction set. erefore, the SD was regarded
as the best pretreatment for its better prediction ability.
3.3.3. Optimization of the Number of Principal Components
(NPC). Although the SD was determined as the optimal
pretreatment in a preliminary investigation, the accuracy in
the model with scores of the first 2 PCs as input variables was
just about 70%, which did not meet the requirement of
discrimination. Hence, the best NPC still needs to be opti-
mized. In light of the modeling and SD pretreatment method
mentioned above, 10 PCA-SVC models (SVC-5 to SVC-14)
were established using the scores of the first 1, 2, 3, ..., 10 PCs
of the calibration set as input variables. As shown in Table 4,
Journal of Spectroscopy 5

the accuracy improved with the increase of the NPC. When
the NPC was 8, the accuracy in the calibration set was 100%
and in the prediction set was 81.82%. When the NPC was
higher than 8, the accuracy in the calibration set was 100%,
while the accuracy in the prediction set did not increase or
even decreased. erefore, number 8 was considered the best
NPC, and model SVC-12 was the optimal qualitative model.
3.3.4. Validation and Evaluation of PCA-SVC Model.
According to the research above, SVC-12 was determined as
the best qualitative analysis model. After the full spectrum
was pretreated with the SD and the dimension was reduced
with PCA, the model was established using the scores of the
first 8 PCs as input variables and category labels as output
variables. e internal parameters of best Cand gwere
optimized with the GS. e initial search was in a range of
log 2 c∈[0,50]and log 2 g∈[0,50]and in steps of 5, and
then the fine search was in a range of log 2 c∈[20, 23] and
log 2 g∈[12, 16] and in steps of 0.5; when C�5.93164 ∗10
6
and g�5792.62, the accuracy of 3-CV was 83.33%. When
the model was predicted with the calibration set and
prediction set, the accuracy was 100% (42/42) and 81.82%
(18/22), respectively, which might be accepted for rapid
identification. e optimizations of internal parameters are
shown in Figure 2, and the predictive results are shown in
Figure 3.
3.4. Quantitative Model Based on PLSR and PLS-SVR
Algorithms
3.4.1. Partition of Sample Set. In this quantitative analysis,
the Kennard–Stone (K-S) algorithm was used to divide 64-
sample spectra into the calibration set and prediction set in a
proportion of 2 : 1 in the MATLAB R2014a software; 42
samples of the calibration set were used for validation, while
22 samples of the prediction set were used for prediction.
3.4.2. PLSR Model. e partial least-squares regression
(PLSR) model is one of the multiple linear regression (MLR)
models; it can easily realize the ideal linear relationship
between input variables (spectral information) and output
variables (ingredient contents) after high dimensions are
compressed by PLS. PLSR has the desirable property to
analyze data that are strongly collinear (correlated), noisy,
and independent variables and also simultaneously model
several response variables; now, it has been developed as a
standard tool in chemometrics [27].
e full spectral data (12000∼4000 cm
−1
) were used for
modeling. To eliminate noise and other factors, they need to
be firstly pretreated. e pretreatments including Raw, VN,
FD, FD + VN, MSC, and FD + MSC were applied. After the
dimensions were reduced with PLS, the treated spectral data
were used as input variables and nitrogen content was the
Table 3: Different spectral pretreatments of PCA-SVC models.
Model number Pretreatment NPC CgAccuracy rate (%)
3-fold cross-validation Calibration set Prediction set
SVC-1 Raw 2 524288 0.03125 54.7619 90.4762 59.0909
SVC-2 VN 2 6.71089 ∗10
7
0.0078125 64.2857 66.6667 63.6364
SVC-3 FD 2 16 32768 59.5238 64.2857 63.6364
SVC-4 SD 2 3.35544 ∗10
7
32768 66.6667 71.4286 68.1818
Table 2: Classified information of the qualitative model of medicinal centipedes.
Sample set S. multidens S. dehaani S. negrocapitis S. mojiangica S. mutilans Total
Calibration set 8 6 5 6 17 42
Prediction set 5 3 3 3 8 22
Label value 1 2 3 4 5 —
Table 4: Comparison on PCA-SVC models established with different NPCs.
Model number NPC CgAccuracy rate (%)
3-fold cross-validation Calibration set Prediction set
SVC-5 1 64 4.29497 ∗10
9
59.5238 80.9524 63.6364
SVC-6 2 3.35544 ∗10
7
32768 66.6667 71.4286 68.1818
SVC-7 3 4.1943 ∗10
6
262144 66.6667 85.7143 81.8182
SVC-8 4 2521.38 1.27148 ∗10
7
71.4286 90.4762 77.2727
SVC-9 5 23170.5 3.65135 ∗10
6
73.8095 97.6190 72.7273
SVC-10 6 26615.9 794672 73.8095 90.4762 77.2727
SVC-11 7 26615.9 1.2045 ∗10
6
78.5714 97.6190 81.8182
SVC-12 8 5.93164 ∗10
6
5792.62 83.3333 100 81.8182
SVC-13 9 131072 262144 80.9524 100 81.8182
SVC-14 10 1.04858 ∗10
6
65536 80.9524 100 77.2727
6Journal of Spectroscopy

output reference, and a series of PLSR models were estab-
lished with the Unscrambler 9.7 software.
During the process, the model was validated and eval-
uated. As shown in Table 5, the RPDs were all over 2 to
possess model reliability. All the values of RMSECV were
higher than those of RMSEE, which indicated the feasibility
of the models. Models PLSR-1, PLSR-4, and PLSR-6 had
lower RMSECV and higher R
2
to present good performance,
while PLSR-1 had great RMSEP, minimum R
2
, in external
validation, and the largest NPC, and its structure is un-
reasonable. ere was no significant difference in perfor-
mance and prediction ability between PLSR-4 and PLSR-6,
but considering the rationality of pretreatment, RMSECV,
and RMSEE, the PLSR-6 model was considered the best
model.
e optimization of the NPC is an important step during
modeling. It can be obtained from the RMSECV-NPC
diagram. For instance, in the PLSR-6 model, with the change
of the NPC, the RMSECV had different values; when the
NPC was 5, the RMSECV had a minimum value, and the
model had the best performance. erefore, the optimal
NPC was determined as 5. e optimization is shown in
Figure 4. e regression equation of the PLSR-6 model
between the principal component scores (SPL
i
,i�1, 2, ..., 5)
and the nitrogen content is expressed as follows:
Y�803.21 SPL1+661.59 SPL2+589.71 SPL3
+595.33 SPL4+476.86 SPL5+10.26, R2�90.95%.
(2)
As described above, the best PLSR model was finally
determined, the optimized pretreatment was determined as
FD + MSC, and the NPC was 5. During modeling, 6-fold
cross-validation was used as internal validation to validate
0 5 10 15 20 25 30 35 40 45 50
0
5
10
15
20
25
30
35
40
45
50
50
50
50 50
50
55
55
606060
60
60
60
65
65
65
65
65
70
70
70
70
70
7575
75
75
75
80
log2 g
log2 C
(a)
20 20.5 21 21.5 22 22.5 23
12
12.5
13
13.5
14
14.5
15
15.5
16
68
71
71
74
74
74
77
77
77
77
80
80
80
80
80
80
80
80
80
80
83
log2 g
log2 C
(b)
Figure 2: Optimization of internal parameters with the grid search of the PCA-SVC model. (a) Initial grid search. (b) Fine search.
0 5 10 15 20 25 30 35 40
1
2
3
4
5
Sample
Label
(a)
0 5 10 15 20
1
2
3
4
5
Sample
Label
(b)
Figure 3: Validation results of the PCA-SVC model for medicinal centipedes: (a) calibration set; (b) prediction set. e red points represent
validation results, and the blue points represent reference values.
Journal of Spectroscopy 7

the performance, and the predictive ability was evaluated
with external validation using the prediction set. e pre-
dictive results are shown in Figure 5. e average relative
deviation between the predictive value and the reference was
2.71%, and the average recovery was 98.77%.
3.4.3. PLS-SVR Model. Besides ingredient information, the
NIR spectroscopy also contains much other information,
such as physical and chemical information, which often
causes spectral bands seriously overlapped. Actually, in most
cases, it shows nonlinear relationship between sample
spectra and content. With the development of application of
chemometrics, modern intelligent algorithms have attached
more attention to NIR spectroscopy analysis for its strong
nonlinear fitting ability and obtained preliminary explora-
tion and application. e SVM algorithm is based on sta-
tistics to allow obtain a good fitting effect and stable
structure. As a result, it becomes a commonly used nonlinear
regression algorithm. Compared with the ANN algorithm
which is suitable for solving problems of complex mapping
and large sample size [28], the SVR model has undergone
much application to become a relatively mature model, and
it is suitable for small sample size.
In this study, an SVR algorithm combined with di-
mensions reduced by the PLS was used to establish a nonlinear
regression model. When the parameters determined in the
PLSR model (the pretreatment was FD + MSC, and di-
mensions reduced with PLS and NPC were 5) were introduced
into the SVM algorithm, the SVR models were performed in
the MATLAB R2014a software. e GS and GA were adopted
to optimize the internal parameters (C,g). e model per-
formance was validated with 6-fold cross-validation using the
calibration set, and the prediction ability was evaluated with
external validation using the prediction set. As shown in Table 6,
the RMSECV in PLS-SVR-2 was 0.34, which is less than 0.4 in
PLS-SVR-1, and the R
2
in PLS-SVR-2 was 93.29, which is larger
than 91.54 in PLS-SVR-1. erefore, the PLS-SVR-2 model with
internal parameters (C,g) optimized with the GS had relatively
excellent performance, and it was regarded as the suitable SVR
model. e optimization of internal parameters (C,g) and
predictive results are shown in Figure 6.
3.4.4. Analysis and Evaluation of Quantitative Models. In
this study, the linear regression model of PLSR and nonlinear
regression model of PLS-SVR were successfully established. As
shown in Tables 5 and 6, the prediction ability had no sig-
nificant difference between the two models. e relative de-
viations between the predictive value and the reference were
2.71% and 2.69%, respectively, and the average recoveries were
98.77% and 99.02%. Totally, the two optimized models had a
reasonable structure and good prediction ability. Both of them
could meet the requirements of accuracy and precision of
quantitative analysis and could be used for nitrogen content
analysis and quality evaluation of medicinal centipedes.
However, the PLSR model was built based on a linear
regression algorithm to have characteristics of fast fitting and
simple calculation, when the analysis requirements were not
too high, and it would be widely used. In contrast, the SVR
model was established based on the nonlinear regression
algorithm, and it had the strong nonlinear fitting ability. It
0.95
0.85
0.75
0.65
0.55
0.45
0.35
12345678910
RMSECV (%)
NPC
Figure 4: RMSECV-NPC diagram of the PLSR-6 model.
Table 5: Validation and predictive results of PLSR models with different pretreatment methods.
Model number Pretreatment 6-fold cross-validation External validation RMSEE (%) NPC
RMSECV (%) R
2
(%) RPD RMSEP (%) R
2
(%) RPD
PLSR-1 Raw 0.42 90.50 2.85 0.51 80.78 2.30 0.27 9
PLSR-2 VN 0.47 87.22 2.51 0.43 84.3 2.51 0.34 5
PLSR-3 FD 0.46 88.14 2.53 0.46 84.41 2.39 0.31 6
PLSR-4 FD + VN 0.41 90.71 3.04 0.43 85.84 2.53 0.32 5
PLSR-5 MSC 0.44 89.63 2.69 0.50 81.72 2.39 0.26 8
PLSR-6 FD + MSC 0.40 90.95 2.96 0.44 85.61 2.51 0.31 5
8Journal of Spectroscopy

was shown from Tables 5 and 6 that the values of RMSECV
and RMSEP of SVR models were generally less than those of
PLSR models, and the values of R
2
of SVR models were
commonly higher than those of PLSR models. It was indicated
that the PLS-SVR model had perfect performance and better
prediction effect than the PLSR model. For this reason, the
PLS-SVR model now becomes the most widely used re-
gression model in NIR spectroscopy analysis. erefore, the
PLS-SVR model with internal parameters (C,g) optimized
with the GS was considered the most suitable NIR spec-
troscopy quantitative model for nitrogen content analysis of
medicinal centipedes, and the PLSR model can act as sup-
plement and verification for the analysis.
4. Conclusions
is study was carried out to explore the feasibility of using
the NIR spectroscopy method to rapidly differentiate species
Table 6: Validation and evaluation results of SVR models.
Model number Optimization method Cg6-fold cross-validation External validation RMSEE (%)
RMSECV (%) R
2
(%) RPD RMSEP (%) R
2
(%) RPD
PLS-SVR-1 GA 99.99 997.03 0.4 91.54 2.91 0.41 85.89 2.55 0.34
PLS-SVR-2 GS 512 1024 0.34 93.29 3.72 0.43 85.5 2.54 0.32
10
10 –10
–10
5
50
log
2
g
log2 C
0
0
0.4
MSE
0.8
–5
–5
(a)
0
6
80510
Sample of prediction set
Reference value
Predictive value
Sample of calibration set
15 20
9
10
11
12
13
8
10
12
14
10
Nitrogen contents (%)
20 30 40
(b)
Figure 6: Parameter optimization (a) and predictive results (b) of the PLS-SVR-2 model.
13
12
11
10
9
8
71312111098
Reference value
Predictive value
Calibration set
(a)
1312111098
Reference value
13
12
11
10
9
8
7
Predictive value
Prediction set
(b)
Figure 5: Predictive results in the calibration set (a) and prediction set (b) of the PLSR-6 model.
Journal of Spectroscopy 9

and evaluate the quality of Chinese medicinal centipedes.
In the qualitative analysis, after spectra were pretreated
with the SD, dimensions were reduced with PCA, and
internal parameters were optimized with the GS algo-
rithm, a PCA-SVC model was set up using the scores of
the first 8 PCs as input variables and category labels as
output variables. e optimal model (SVC-12) was vali-
dated and evaluated, which could identify five species of
medicinal centipedes with an accuracy of 100% (42/42) in
the calibration set and 81.82% (18/22) in the prediction
set. It could be accepted as an objective, rapid, and
auxiliary method for identifying the species of medicinal
centipedes. rough the spectra pretreated with
FD + MSC, data dimension reduced with PLS, and NPC
determined as 5, two best quantitative models of PLSR and
PLS-SVR were also successfully determined. During the
process of modeling, the RMSECV, R
2
, and RPD of 6-fold
internal cross-validation in the calibration set indicated
the better performance and stronger modeling capacity.
e RMSEP, R
2
, and RPD of external validation in the
prediction set proved stronger prediction ability. In ad-
dition, to investigate the reasonability of the model, the
calibration set used for full cross-validation was input to
the models again, and the RMSEE was used as an index.
Comparing the characteristics and advantages of two
different regression algorithms, the PLS-SVR-2 model had
excellent performance and strong prediction capacity, and
it was finally considered the most suitable quantitative
model of NIR spectroscopy for nitrogen content analysis
of medicinal centipedes.
Meanwhile, the pretreatment methods were also opti-
mized in this paper; although the SD was determined in the
qualitative model, MSC or its combined methods were
applied to pretreat the spectra in quantitative models. e
MSC had advantages of weakening or eliminating in-
terference caused by the uneven grain size of solid powder in
the diffuse reflection spectrum [29]. In this research, all
samples were smashed into powder, and the NIR spectra
were obtained with a diffuse reflection spectrum, so the
application of FD + MSC was proved to be reasonable.
is study indicated that NIR spectroscopy combined
with chemometric algorithms could be successfully used to
differentiate species and evaluate the quality of medicinal
centipedes in China, which was characterized with rapid,
nondestructive, and environmentally friendly properties.
However, this study just represented preliminary explor-
atory research; although 28 batch samples and 64 in-
dividuals were conducted, the sample size was still limited.
In the future, more samples will be used to improve the
prediction ability, and other algorithms will also be con-
sidered to simplify the model and improve performance.
is study also provided a reference for rapid identification
and quality analysis of other animal medicinal materials
using NIR spectroscopy.
Data Availability
e data used to support the findings of this study are
available from the corresponding author upon request.
Conflicts of Interest
e authors declare that there are no conflicts of interest
regarding the publication of this paper.
Acknowledgments
is work was supported by a grant from the major drug
discovery projects of the National Ministry of Science and
Technology of China (no. 2014ZX09304307001).
References
[1] W. Siriwut, G. D. Edgecombe, C. Sutcharit, and S. Panha, “e
centipede genus Scolopendra in mainland southeast asia:
molecular phylogenetics, geometric morphometrics and ex-
ternal morphology as tools for species delimitation,” PLoS
One, vol. 10, no. 8, Article ID e0135355, 2015.
[2] S. H. Kang, H. Y. Deng, Z. Y. Jiang, Y. M. Liu, J. Li, and
K. L. Chen, “Taxonomy and distribution of Chinese medicinal
centipedes,” Journal of Chinese Medicinal Materials, vol. 39,
pp. 727–731, 2016.
[3] S. Yang, Y. Xiao, D. Kang et al., “Discovery of a selective
NaV1.7 inhibitor from centipede venom with analgesic effi-
cacy exceeding morphine in rodent pain models,” Proceedings
of the National Academy of Sciences, vol. 110, no. 43,
pp. 17534–17539, 2013.
[4] Y. Kong, “Cytotoxic and anticoagulant peptide from Scolo-
pendra subspinipes mutilans venom,” African Journal of
Pharmacy and Pharmacology, vol. 7, no. 31, pp. 2238–2245,
2013.
[5] I.-J. Jo, G. S. Bae, K. C. Park et al., “Scolopendra subspinipes
mutilans protected the cerulein-induced acute pancreatitis by
inhibiting high-mobility group box protein-1,” World Journal
of Gastroenterology, vol. 19, no. 10, pp. 1551–1562, 2013.
[6] W. Ma, D. Zhang, L. Zheng, Y. Zhan, and Y. Zhang, “Potential
roles of Centipede Scolopendra extracts as a strategy against
EGFR-dependent cancers,” American Journal of Translational
Research, vol. 7, no. 1, pp. 39–52, 2015.
[7] W. Ma, R. Liu, J. Qi, and Y. Zhang, “Extracts of centipede
Scolopendra subspinipes mutilans induce cell cycle arrest and
apoptosis in A375 human melanoma cells,” Oncology Letters,
vol. 8, no. 1, pp. 414–420, 2014.
[8] Y. Kong, S.-L. Huang, Y. Shao, S. Li, and J.-F. Wei, “Purifi-
cation and characterization of a novel antithrombotic peptide
from Scolopendra subspinipes mutilans,” Journal of Ethno-
pharmacology, vol. 145, no. 1, pp. 182–186, 2013.
[9] M. Stankiewicz, A. Hamon, R. Benkhalifa et al., “Effects of a
centipede venom fraction on insect nervous system, a native
Xenopus oocyte receptor and on an expressed Drosophila
muscarinic receptor,” Toxicon, vol. 37, no. 10, pp. 1431–1445,
1999.
[10] H. Y. Zhang, J. Chen, J. Jia et al., “Identification of Scolo-
pendra subspinipes mutilans and its adulterants using DNA
barcode,” China Journal of Chinese Materia Medica, vol. 39,
p. 2208, 2014.
[11] H. Fang, F. Deng, and K. Q. Wang, “Chemical analysis of
Scolopendra multidens newport,” Chinese Pharmaceutical
Journal, vol. 32, pp. 202–204, 1997.
[12] H. Fang, F. Deng, Y. C. Yan, and K. Q. Wang, “Chemical
constituents of Scolopendra negrocapitis,” Journal of Chinese
Medicinal Materials, vol. 22, no. 5, pp. 226–228, 1999.
[13] X. Chen, H. M. Wen, R. Liu et al., “Analysis of extracted
proteins of Scolopendra by nanoflow reversed phase liquid
10 Journal of Spectroscopy

chromatography-tandem mass spectrometry,” Chinese Jour-
nal of Analytical Chemistry, vol. 42, pp. 239–243, 2014.
[14] J. Wang, Y. P. Chu, S. S Yang et al., “Influences of sterilization
method on the nitrogen content of Scolopendra,” Asia-Pacific
Traditional Medicine, vol. 12, pp. 27-28, 2016.
[15] Y. He, X. L. Li, and Y. N. Shao, “Discrimination of varieties of
apple using near infrared spectra based on principal component
analysis and artificial neural network model,” Spectroscopy and
Spectral Analysis, vol. 26, no. 5, pp. 850–853, 2006.
[16] C. Xie, N. Xu, Y. Shao, and Y. He, “Using FT-NIR spec-
troscopy technique to determine arginine content in fer-
mented Cordyceps sinensis mycelium,” Spectrochimica Acta
Part A: Molecular and Biomolecular Spectroscopy, vol. 149,
pp. 971–977, 2015.
[17] Y. Sun, L. Chen, B. Huang, and K. Chen, “A rapid identifi-
cation method for calamine using near-infrared spectroscopy
based on multi-reference correlation coefficient method and
back propagation artificial neural network,” Applied Spec-
troscopy, vol. 71, no. 7, pp. 1447–1456, 2017.
[18] Z. S. Song, D. X. Song, and M. S. Zhu, “Systematic classifi-
cation of chilopoda and the order scolopendromorpha
(myriapoda),” Journal of Liaoning Normal University, vol. 27,
no. 1, pp. 69–72, 2004.
[19] C. Z. Zhang and K. Q. Wang, “A new centipede, Scolopendra
negrocapitis sp. nov. from Hubei province, China (Chilopoda:
Scolopendromorpha: Scolopendridae),” Acta Zootaxonomica
Sinica, vol. 24, no. 2, pp. 136-137, 1999.
[20] X. L. Chu, H. F. Yuan, and W. Z. Lu, “Progress and application
of spectral data pretreatment and wavelength selection
methods in NIR analytical technique,” Progress in Chemistry,
vol. 16, no. 4, pp. 528–542, 2004.
[21] Q. Kang, Q. Ru, Y. Liu et al., “On-line monitoring the extract
process of Fu-fang Shuanghua oral solution using near in-
frared spectroscopy and different PLS algorithms,” Spec-
trochimica Acta Part A: Molecular and Biomolecular
Spectroscopy, vol. 152, pp. 431–437, 2016.
[22] W. Z. Lu, Near Infrared Spectroscopy Instrument, Chemical
Industry Press, Beijing, China, 2010.
[23] G. B´az´ar, R. Romv´ari, A. Szab´o, T. Somogyi, V. ´
Eles, and
R. Tsenkova, “NIR detection of honey adulteration reveals
differences in water spectral pattern,” Food Chemistry,
vol. 194, pp. 873–880, 2016.
[24] Z.-S. Wu, L.-W. Zhou, S.-Y. Dai, X.-Y. Shi, and Y.-J. Qiao,
“Evaluation of the value of near infrared (NIR) spec-
tromicroscopy for the analysis of glycyrrizhic acid in licorice,”
Chinese Journal of Natural Medicines, vol. 13, no. 4,
pp. 316–320, 2015.
[25] M. Y. Yuan, B. S. Huang, C. Yu, Y. M. Liu, and K. L. Chen, “A
NIR qualitative and quantitative model of 8 kinds of car-
bonate-containing mineral Chinese medicines,” China Jour-
nal of Chinese Materia Medica, vol. 39, pp. 267–272, 2014.
[26] V. Vapnik, e Nature of Statistical Learning eory, Springer
Berlin Heidelberg, New York, NY, USA, 1995.
[27] S. Wold, M. Sj¨
ostr¨
om, and L. Eriksson, “PLS-regression: a
basic tool of chemometrics,” Chemometrics and Intelligent
Laboratory Systems, vol. 58, no. 2, pp. 109–130, 2001.
[28] L. Chen, J. Wang, Z. Ye et al., “Classification of Chinese honeys
according to their floral origin by near infrared spectroscopy,”
Food Chemistry, vol. 135, no. 2, pp. 338–342, 2012.
[29] W. M. Wang, D. M. Dong, W. G. Zheng, X. D. Zhao, L. Z. Jiao,
and M. F. Wang, “Pretteatment method of near-infrared
diffuse reflection spectra for sugar content prediction of
pears,” Spectroscopy and Spectral Analysis, vol. 33, no. 2,
pp. 359–362, 2013.
Journal of Spectroscopy 11