1.Assessment and discussion of quality monitoring data for red blood cell preparations
Yun QING ; Huayou DAI ; Junhong YANG ; Qian XU ; Siqi WU ; Yunbo TIAN ; Xia HUANG
Chinese Journal of Blood Transfusion 2025;38(2):227-232
[Objective] To assess the data characteristics of quality monitoring indicators for red blood cell (RBC) preparations, so as to provide reference for continuous improvement of blood quality. [Methods] The quality inspection data of 6 types of RBC preparations from Chongqing blood center from 2019 to 2023 were summarized. For the same indicators, the numerical range of quality indicators was monitored by comparing different types of preparations with the national standard GB18469. The loss and/or damage to RBCs caused by different preparation process were compared, and the impact of different preparation processes on the quality of RBCs was discussed. [Results] The appearance and sterility test compliance rates of the six types of RBC preparations were both 100%, while the compliance rates of other items were all ≥75%. The compliance rate of hematocrit for suspended RBCs was the lowest at 75%, with a median of 0.52, which was close to the lower limit of GB18469, while the medians of hematocrit for the other types were all at the midline level of GB18469. The Hb content for different types of RBCs was significantly higher than the corresponding requirements of GB18469 (P<0.05). The hemolysis rate at the end of storage for different types of RBCs was significantly lower than the requirements of GB18469 (P<0.05). The 1 U leukoreduction process resulted in a hemoglobin content loss of about 5% and had a significant impact on the hemolysis rate at the end of storage (P<0.05). The washing process resulted in a hemoglobin content loss of <3% and had no significant impact on the hemolysis rate at the end of storage (P>0.05). The concentration process resulted in a hemoglobin content loss of <3% and had a significant impact on the hemolysis rate at the end of storage (P<0.05). [Conclusion] The impact of different processes on RBC preparations is within a controllable range and meets the requirements of GB18469. The quality monitoring data can provide a reference for clinical blood selection, effectiveness evaluation and revision of related standards.
2.Disparities in unexpected antibody distribution and clinical features by frequency of cross-matching incompatibility
Danli CUI ; Bujin LIU ; Haiman ZOU ; Pengwei YIN ; Yun QING ; Huayou DAI ; Siqi WU ; Junhong YANG ; Xia HUANG
Chinese Journal of Blood Transfusion 2025;38(8):1063-1070
Objective: To investigate the clinical characteristics, the types of unexpected antibodies, and their impacts on immunological risks among patients with different frequencies of cross-matching incompatibility, so as to propose corresponding solutions. Methods: Data of cross-matching incompatibility samples from 92 medical institutions during 2022 to 2024 were collected and divided into three groups based on the frequency of cross-matching. Statistical analysis was performed on disease types, distribution of hematologic diseases, alloantibody detection rates, and proportions of alloantibody types. Results: The 858 patients were divided into three groups based on the frequency of blood cross-matching incompatibility: ≥5 times (8.28%, 71/858), 2 to 4 times (28.21%, 242/858); 1 time (63.52%, 545/858). There was a clustered distribution of disease types in the ≥5 cross-matchings group, with 71.83% (51/71) of patients having tumors or hematologic and hematopoietic diseases. In contrast, the disease types in the 2 to 4 cross-matchings and 1 cross-matching groups were more diverse. An analysis of 249 patients with hematologic diseases found that multiple myeloma was the most common disease in all three groups, accounting for 31.43% (11/35), 35.37% (29/82), and 37.88% (50/132) respectively. In the ≥5 cross-matchings group, myelodysplastic syndrome (14.29%, 5/35) and thalassemia (14.29%, 5/35) were the second most common diseases. In contrast, in the 2 to 4 cross-matchings group and 1 cross-matching group, autoimmune hemolytic anemia was the second most common disease, with prevalence rates of 20.73% (17/82) and 24.24% (32/132), respectively. Alloantibodies were detected in 54.66% of the patients, with antibodies against Rh blood group being most frequent (>50%) in all three groups. The detection rates of alloantibodies/alloantibodies with coexisting autoantibodies decreased across groups: the ≥5 cross-matchings group (70.42%, 50/71) > the 2 to 4 cross-matchings group (54.96%, 133/242) > the 1 cross-matching group (52.48%, 286/545). Conclusion: The risk of alloantibody production increases in patients with multiple cross-matching incompatibilities, especially in those with tumors or hematologic diseases. For handling of cross-matching incompatibility cases, it is recommended to optimize the cross-matching process, implement individualized transfusion plans, and enhance the technical capabilities of clinical transfusion departments and blood group reference laboratories to ensure the safety and effectiveness of transfusions.
3. Network pharmacology-based study on mechanism of Zhi-Huang-Zhi-Tong powder in rheumatoid arthritis treatment
Xiao-Yun TIAN ; Ying-Jie YANG ; Wan-Ting ZHENG ; Ming-Qing HUANG ; Li-Hong NAN ; Jian-Yu CHEN ; Hai-Yu ZHAO
Chinese Pharmacological Bulletin 2024;40(2):381-389
Aim To discover the potential active compounds and possible mechanisms in rheumatoid arthritis (RA) treatment with Zhi-Huang-Zhi-Tong powder (ZHZTP) by using network pharmacology and in vitro study. Methods The active ingredient targets and disease targets of Zhihuang Zhitong Powder were searched and screened by database; they intersected to get a common target; and the "drug-component-target" relationship network diagram was constructed for GO and KEGG enrichment analysis of the overlapping genes; then the core components were docked with the core targets. Finally, based on the inflammation model of HUVECs in vitro, the efficacy and mechanism of Zhihuang Zhitong powder were verified by MTT method, plate scratch test and Western blot. Results Active compounds involved in RA treatment were screened in the present study, and the top two were ursolic acid and emodin, all playing crucial roles in RA treatment with ZHZTP. Additionally, the key target was AKTA, TNF and IL-6. GO and KEGG enrichment analysis revealed that ZHZTP regulated BP, MF and CC, and also focused on regulating AKTA, TNF and IL-6 signaling pathway. Molecular docking showed that interactions between key active compounds and key targets were stable. In vitro ZHZTP significantly inhibited cell viability and migration of TNF-a-stimulated HUVECs, and the involved mechanism may be associated with PI3K/AKT/m-TOR signaling. Conclusions The present study reveals that the potential active compounds of ZHZTP are ursolic acid and emodin, and moreover, the involved mechanisms of ZHZTP for RA treatment are associated with PI3 K/AKT/m-TOR signaling.
4.A Bibliometric and Visual Analysis of the Current Status and Trends of Foren-sic Mixed Stain Research
Qing-Wei FAN ; Ling LI ; Hui-Ling YANG ; Ting-Ting DENG ; Dong-Dong XU ; Yun WANG ; Bing DU ; Jiang-Wei YAN
Journal of Forensic Medicine 2024;40(1):20-29
Objective To explore the context and hotspot changes of forensic mixed stain research through bibliometric approach.Methods The literature of forensic mixed stain included in the core col-lection of Web of Science database from 2011 to 2022 were collected as the study object,and the an-nual publication number,countrie(region),institution,journal,keywords,etc.were bibliometrically and visually analyzed using the R-based Bibliometrix 1.1.6 package and VOSviewer 1.6.18 software.Re-sults A total of 732 articles on forensic mixed stain were included from 2011 to 2022,with the an-nual number of articles published and the annual citation frequency showing a steady increase year by year.Among the 59 countries(regions)with the most published articles,the United States ranked first with 246 articles,followed by China with 153 articles.The literature came from 104 journals,and the total number of articles published in the top 10 journals was 633.FORENSIC SCI INT GENET ranked first with 307 articles.Visual analysis using VOSviewer software showed that keywords could be divided into four research clusters,namely the genetic marker development group(blue),the mixed stain typing analysis theory group(red),the sequencing analysis group(yellow),and the case sample research group(green).It can be divided into four development stages in terms of different time peri-ods:early development(2011-2013),middle development(2014-2016),rapid development(2017-2020)and latest development(2021-2022).Conclusion The number of publications by domestic and foreign scholars in the study of mixed stain in forensic science is showing a relatively stable trend.Machine learning,next generation sequencing and other research have been the hottest topics that have attracted the most attention in recent years,which is expected to further develop the theory of mixed stain typing and sequencing analysis in forensic mixed stain research.
5.PRMT7 Regulates Adipogenic Differentiation of hBMSCs by Modulating IGF-1 Signaling
Qian GUO ; Jia QING ; Da-Zhuang LU ; Xu WANG ; Yang LI ; Hui ZHANG ; Ying-Fei ZHANG ; Yun-Song LIU ; Yong-Sheng ZHOU ; Ping ZHANG
Progress in Biochemistry and Biophysics 2024;51(6):1406-1417
ObjectiveProtein arginine methyltransferases (PRMTs) play pivotal roles in numerous cellular biological processes. However, the precise regulatory effects of PRMTs on the fate determination of mesenchymal stromal/stem cells (MSCs) remain elusive. Our previous studies have shed light on the regulatory role and molecular mechanism of PRMT5 in MSC osteogenic differentiation. This study aims to clarify the role and corresponding regulatory mechanism of PRMT7 during the adipogenic differentiation of bone marrow-derived mesenchymal stem cells (BMSCs). Methods(1) Human bone marrow-derived mesenchymal stem cells (hBMSCs) were cultured in a medium that induces adipogenesis. We used qRT-PCR and Western blot to monitor changes in PRMT7 expression during adipogenic differentiation. (2) We created a cell line with PRMT7 knocked down and assessed changes in PRMT7 expression and adipogenic capacity using Oil Red O staining, qRT-PCR and Western blot. (3) We implanted hBMSCs cell lines mixed with a collagen membrane subcutaneously into nude mice and performed Oil Red O staining to observe ectopic lipogenesis in vivo. (4) A cell line overexpressing PRMT7 was generated, and we examined changes in PRMT7 expression using qRT-PCR and Western blot. We also performed Oil Red O staining and quantitative analysis after inducing the cells in lipogenic medium. Additionally, we assessed changes in PPARγ expression. (5) We investigated changes in insulin-like growth factor 1 (IGF-1) expression in both PRMT7 knockdown and overexpressing cell lines using qRT-PCR and Western blot, to understand PRMT7’s regulatory effect on IGF-1 expression. siIGF-1 was transfected into the PRMT7 knockdown cell line to inhibit IGF-1 expression, and knockdown efficiency was confirmed. Then, we induced cells from the control and knockdown groups transfected with siIGF-1 in lipogenic medium and performed Oil Red O staining and quantitative analysis. Finally, we assessed PPARγ expression to explore IGF-1’s involvement in PRMT7’s regulation of adipogenic differentiation in hBMSCs. Results(1) During the adipogenesis process of hBMSCs, the expression level of PRMT7 was significantly reduced (P<0.01). (2) The adipogenic differentiation ability of PRMT7 knockdown group was significantly stronger than that of control group (P<0.001). (3) The ectopic adipogenic differentiation ability of PRMT7 knockdown group was significantly stronger than that of control group. (4) The adipogenic differentiation ability of the PRMT7 overexpression group was significantly weaker than that of the control group (P<0.01). (5) The expression level of IGF-1 increased after PRMT7 knockdown (P<0.000 1). The expression level of IGF-1 decreased after PRMT7 overexpression (P<0.000 1), indicating that PRMT7 regulates the expression of IGF-1. After siIGF-1 transfection, the expression level of IGF-1 in all cell lines decreased significantly (P<0.001). The ability of adipogenic differentiation of knockdown group transfected with siIGF-1 was significantly reduced (P<0.01), indicating that IGF-1 affects the regulation of PRMT7 on adipogenic differentiation of hBMSCs. ConclusionIn this investigation, our findings elucidate the inhibitory role of PRMT7 in the adipogenic differentiation of hBMSCs, as demonstrated through both in vitro cell-level experiments and in vivo subcutaneous transplantation experiments conducted in nude mice. Mechanistic exploration revealed that PRMT7’s regulatory effect on the adipogenic differentiation of hBMSCs operates via modulation of IGF-1 signaling pathway. These collective findings underscore PRMT7 as a potential therapeutic target for fatty metabolic disorders, thereby offering a novel avenue for leveraging PRMT7 and hBMSCs in the therapeutic landscape of relevant diseases.
6.Correlation between tear inflammatory factors and postoperative complications in glaucoma patients after trabeculectomy
Yu ZHANG ; Qing-Yi ZHAO ; Ning-Ling WU ; Yun-Yun YANG ; Xin-Yue ZHANG
Journal of Regional Anatomy and Operative Surgery 2024;33(3):247-251
Objective To investigate the application value of tear inflammatory factors in early evaluation of the occurrence of complications after trabeculectomy in glaucoma patients.Methods A total of 150 eyes of 150 patients with angle-closure glaucoma who underwent trabeculectomy in Department of Ophthalmology of Zhangjiagang Hospital of Traditional Chinese Medicine were included.The clinical data of all patients were collected,and the levels of inflammatory factors(including G-CSF,GM-CSF,IFN-γ,MCP-1,TNF-α,IL-12,IL-13,IL-15,IL-1β,IL-4,IL-7,IL-10,IL-17,IL-5)in tears were detected before surgery.According to the occurrence of postoperative complications,the patients were divided into the complication group and the control group,and the clinical data of the two groups were compared.Logistic regression was used to analyze the related factors of the occurrence of complications after trabeculectomy in glaucoma patients.The predictive model of tear inflammatory factors was established by Logistic regression,and the receiver operating characteristic(ROC)curve was drawn to analyze the value of this model in early evaluation of the occurrence of complications after trabeculectomy in glaucoma patients.Results Complications occurred in 35(23.33%)of 150 glaucoma patients after trabeculectomy,including 24 cases of shallow anterior chamber,5 cases of ciliary body or choroid detachment,3 cases of iridocyclitis,2 cases of hyphema,and 1 case of retinal detachment.IFN-γ,GM-CSF and IL-5 levels of the patients in the complication group were lower than those in the control group(P<0.05).There was no significant difference in the levels of other inflammatory factors of patients between the two groups(P>0.05).Multivariate Logistic regression analysis showed that increased levels of IFN-γ(OR=0.999),GM-CSF(OR=0.988)and IL-5(OR=0.996)were independent protective factors for complications after trabeculectomy in glaucoma patients(P<0.05).ROC curve analysis showed that the sensitivity and specificity of the model were 94.29%and 83.84%in early evaluation of the occurrence of complications after trabeculectomy in glaucoma patients,and the AUC was 0.906,which was higher than that predicted by IFN-γ(AUC=0.642),GM-CSF(AUC=0.721)and IL-5(AUC=0.666)alone.Conclusion Preoperative analysis of tear inflammatory factors in glaucoma patients can early evaluate the occurrence of postoperative complications,especially the combined detection of IFN-γ,GM-CSF and IL-5 levels is of great significance for predicting the occurrence of complications.
7.Electroacupuncture Promotes Functional Recovery after Facial Nerve Injury in Rats by Regulating Autophagy via GDNF and PI3K/mTOR Signaling Pathway.
Jun-Peng YAO ; Xiu-Mei FENG ; Lu WANG ; Yan-Qiu LI ; Zi-Yue ZHU ; Xiang-Yun YAN ; Yu-Qing YANG ; Ying LI ; Wei ZHANG
Chinese journal of integrative medicine 2024;30(3):251-259
OBJECTIVE:
To explore the mechanism of electroacupuncture (EA) in promoting recovery of the facial function with the involvement of autophagy, glial cell line-derived neurotrophic factor (GDNF), and phosphatidylinositol-3-kinase (PI3K)/mammalian target of rapamycin (mTOR) signaling pathway.
METHODS:
Seventy-two male Sprague-Dawley rats were randomly allocated into the control, sham-operated, facial nerve injury (FNI), EA, EA+3-methyladenine (3-MA), and EA+GDNF antagonist groups using a random number table, with 12 rats in each group. An FNI rat model was established with facial nerve crushing method. EA intervention was conducted at Dicang (ST 4), Jiache (ST 6), Yifeng (SJ 17), and Hegu (LI 4) acupoints for 2 weeks. The Simone's 10-Point Scale was utilized to monitor the recovery of facial function. The histopathological evaluation of facial nerves was performed using hematoxylin-eosin (HE) staining. The levels of Beclin-1, light chain 3 (LC3), and P62 were detected by immunohistochemistry (IHC), immunofluorescence, and reverse transcription-polymerase chain reaction, respectively. Additionally, IHC was also used to detect the levels of GDNF, Rai, PI3K, and mTOR.
RESULTS:
The facial functional scores were significantly increased in the EA group than the FNI group (P<0.05 or P<0.01). HE staining showed nerve axons and myelin sheaths, which were destroyed immediately after the injury, were recovered with EA treatment. The expressions of Beclin-1 and LC3 were significantly elevated and the expression of P62 was markedly reduced in FNI rats (P<0.01); however, EA treatment reversed these abnormal changes (P<0.01). Meanwhile, EA stimulation significantly increased the levels of GDNF, Rai, PI3K, and mTOR (P<0.01). After exogenous administration with autophagy inhibitor 3-MA or GDNF antagonist, the repair effect of EA on facial function was attenuated (P<0.05 or P<0.01).
CONCLUSIONS
EA could promote the recovery of facial function and repair the facial nerve damages in a rat model of FNI. EA may exert this neuroreparative effect through mediating the release of GDNF, activating the PI3K/mTOR signaling pathway, and further regulating the autophagy of facial nerves.
Rats
;
Male
;
Animals
;
Rats, Sprague-Dawley
;
Electroacupuncture
;
Phosphatidylinositol 3-Kinase/metabolism*
;
Facial Nerve Injuries/therapy*
;
Phosphatidylinositol 3-Kinases/metabolism*
;
Beclin-1
;
Glial Cell Line-Derived Neurotrophic Factor
;
Signal Transduction
;
TOR Serine-Threonine Kinases/metabolism*
;
Autophagy
;
Mammals/metabolism*
8.Effect of Simo decoction on the regulation of NLRP3/Caspase-1/GSDMD signal pathway on duodenal microinflammation in rats with functional dyspepsia
Qin LIU ; Xiao-Yuan LIN ; Ling-Feng YANG ; Qian LUO ; Yun-Zong HAN ; Si-Qing CHEN ; Hai-Yue ZHANG ; Shu ZHOU ; Sai-Nan ZHOU
The Chinese Journal of Clinical Pharmacology 2024;40(1):67-71
Objective To investigate the effects of Simo decoction on duodenal microinflammation and NOD-like receptor thermal protein domain associated protein 3(NLRP3)/cysteinyl aspartate-specific proteinase-1(Caspase-1)/gasdermin D(GSDMD)signaling pathway in rats with functional dyspepsia(FD).Methods The FD model was established by multifactorial method.SD rats were randomly divided into normal group,model group(FD model),positive control group(gavage administration of 0.305 mg·kg-1 mosapride injection)and experimental-H,-M,-L groups(gavage administration of 5.62,2.81,1.40 g·kg-1 Simo decoction).Small intestinal advancement rate and gastric emptying rate was determined;the levels of interleukin(IL)-1 β and IL-18 in serum were determined by enzyme linked immunosorbent assay(ELISA);the protein expression of NLRP3 and GSDMD in duodenal tissue was detected by Western blotting.Results The gastric emptying rates of normal,model,positive control and experimental-H,-M,-Lgroupswere(58.34±5.72)%,(29.16±8.37)%,(48.77±6.10)%,(48.35±6.04)%,(48.20±3.49)%and(39.24±4.20)%;the small intestinal propulsion rates were(82.01±7.55)%,(41.95±9.53)%,(64.61±10.18)%,(75.04±9.76)%,(60.58±7.13)%and(45.89±7.40)%;serum IL-1 β expression were(12.86±0.88),(43.73±4.60),(18.84±0.86),(24.61±1.57),(19.14±0.77)and(29.04±0.72)pg·mL-1;IL-18 expressions were(95.00±3.74),(170.60±8.78),(108.50±3.05),(118.90±3.45),(99.90±8.70)and(141.00±3.71)pg·mL-1;the relative expression levels of NLRP3 proteins were 0.32±0.02,0.84±0.05,0.42±0.03,0.48±0.02,0.61±0.04 and 0.62±0.05;the relative expression levels of GSDMD proteins were 0.34±0.05,0.93±0.06,0.35±0.03,0.52±0.02,0.53±0.06 and 0.55±0.05,respectively.Compared with the normal group,the above indexes in the model group have statistical significance;compared with the model group,the above indexes in the experimental-H group and the positive control group also have statistical significance(P<0.01 or P<0.05).Conclusion Simo decoction can effectively improve the general condition and duodenal microinflammation in FD rats,and the mechanism may be related to the inhibition of duodenal NLRP3/Caspase-1/GSDMD signaling pathway.
9.Clinical trial of brexpiprazole in the treatment of adults with acute schizophrenia
Shu-Zhe ZHOU ; Liang LI ; Dong YANG ; Jin-Guo ZHAI ; Tao JIANG ; Yu-Zhong SHI ; Bin WU ; Xiang-Ping WU ; Ke-Qing LI ; Tie-Bang LIU ; Jie LI ; Shi-You TANG ; Li-Li WANG ; Xue-Yi WANG ; Yun-Long TAN ; Qi LIU ; Uki MOTOMICHI ; Ming-Ji XIAN ; Hong-Yan ZHANG
The Chinese Journal of Clinical Pharmacology 2024;40(5):654-658
Objective To evaluate the efficacy and safety of brexpiprazole in treating acute schizophrenia.Methods Patients with schizophrenia were randomly divided into treatment group and control group.The treatment group was given brexpiprozole 2-4 mg·d-1 orally and the control group was given aripiprazole 10-20 mg·d-1orally,both were treated for 6 weeks.Clinical efficacy of the two groups,the response rate at endpoint,the changes from baseline to endpoint of Positive and Negative Syndrome Scale(PANSS),Clinical Global Impression-Improvement(CGI-S),Personal and Social Performance scale(PSP),PANSS Positive syndrome subscale,PANSS negative syndrome subscale were compared.The incidence of treatment-related adverse events in two groups were compared.Results There were 184 patients in treatment group and 186 patients in control group.After treatment,the response rates of treatment group and control group were 79.50%(140 cases/184 cases)and 82.40%(150 cases/186 cases),the scores of CGI-I of treatment group and control group were(2.00±1.20)and(1.90±1.01),with no significant difference(all P>0.05).From baseline to Week 6,the mean change of PANSS total score wese(-30.70±16.96)points in treatment group and(-32.20±17.00)points in control group,with no significant difference(P>0.05).The changes of CGI-S scores in treatment group and control group were(-2.00±1.27)and(-1.90±1.22)points,PSP scores were(18.80±14.77)and(19.20±14.55)points,PANSS positive syndrome scores were(-10.30±5.93)and(-10.80±5.81)points,PANSS negative syndrome scores were(-6.80±5.98)and(-7.30±5.15)points,with no significant difference(P>0.05).There was no significant difference in the incidence of treatment-related adverse events between the two group(69.00%vs.64.50%,P>0.05).Conclusion The non-inferiority of Brexpiprazole to aripiprazole was established,with comparable efficacy and acceptability.
10.Near Infrared Spectral Analysis Based on Data Augmentation Strategy and Convolutional Neural Network
Yun ZHENG ; Si-Yu YANG ; Tao WANG ; Zhuo-Wen DENG ; Wei-Jie LAN ; Yong-Huan YUN ; Lei-Qing PAN
Chinese Journal of Analytical Chemistry 2024;52(9):1266-1276
Near infrared spectroscopy(NIRS)technology combined with chemometrics algorithms has been widely used in quantitative and qualitative analysis of food and medicine.However,traditional chemometrics methods,especially linear classification methods,often yield unsatisfactory results when addressing multi-class classification problems.Convolutional neural network(CNN)is adept at extracting deep-level features from data and suitable for handling non-linear relationships.The modeling performance of CNN depends on the size and diversity of sample,while the collection and preprocessing of NIRS sample data is often time-consuming and labor-intensive.This study proposed a NIRS qualitative analysis method based on data augmentation strategies and CNN.The data augmentation strategy included two steps.Firstly,applying Bootstrap resampling and generative adversarial network(GAN)methods to augment three NIRS datasets(Medicine,coffee and grape).Secondly,combining the original samples(Y)with the Bootstrap augmented samples(B)and GAN augmented samples(G)to obtain three augmented datasets(Y-B,Y-G and Y-B-G).Based on this,a CNN model structure suitable for these datasets was designed,consisting of 2 one-dimensional convolutional layers,1 max-pooling layer,and 1 fully connected layer.The results showed that compared to the optimal models of partial least squares discriminant analysis(PLS-DA),support vector machine(SVM),and back propagation neural network(BP),the CNN model based on Y-B dataset achieved average accuracy improvements of 3.998%,9.364%,and 4.689%for medicine(Binary classification);the CNN model based on the Y-B-G dataset achieved average accuracy improvements of 6.001%,2.004%,and 7.523%for coffee(7-class classification);and the CNN model based on the Y-B dataset achieved average accuracy improvements of 33.408%,51.994%,and 34.378%for grapes(20-class classification).It was evident that the models established based on data augmentation strategies and CNN demonstrated better classification accuracy and generalization performance with different datasets and classification categories.

Result Analysis
Print
Save
E-mail