Establishment and analysis of osteoarthritis diagnosis model based on artificial neural networks

doi:10.12307/2024.294

Abstract

Abstract: BACKGROUND: Rapid developments in the field of bioinformatics have provided new methods for the diagnosis of osteoarthritis. Artificial neural networks have powerful data computing and classification capabilities, which have shown better performance in disease diagnosis.
OBJECTIVE: To establish a new diagnostic predictive model of osteoarthritis based on artificial neural network and to verify the diagnostic value of the model in osteoarthritis with an external dataset.
METHODS: The eligible osteoarthritis-related data sets were downloaded through GEO database search and divided into Train group and Test group. The gene expression matrix of the Train group was analyzed to screen the differentially expressed genes. GO and KEGG enrichment analyses were performed on the differentially expressed genes. Through Lasso regression model, support vector machine model and random forest tree model, the key genes of osteoarthritis were further identified from the differentially expressed genes. The R software “Neuralnet” package was then used to construct the osteoarthritis diagnosis model based on artificial neural network, and the model performance was evaluated by the five-fold cross-validation. Two independent data sets in the Test group were used to verify their diagnostic results.
RESULTS AND CONCLUSION: A total of 90 differentially expressed genes related to osteoarthritis were obtained by differential analysis, of which 33 were down-regulated and 57 were up-regulated. GO enrichment analysis showed that the differentially expressed genes were mainly involved in the following biological processes, including leukocyte-mediated immunity, leukocyte migration in bone marrow and chemokine production. KEGG enrichment analysis showed that these genes were mainly enriched in rheumatoid arthritis, interleukin-17 signaling pathway and osteoclast differentiation pathway. Five key genes for the diagnosis of osteoarthritis, HMGB2, GADD45A, SLC19A2, TPPP3 and FOLR2, were identified by three machine learning methods. The artificial neural network model of five key genes in the Train group showed that the accuracy was 96.36% and the area under the curve was 0.997. The five-fold cross validation of the neural network model showed that the average area under the curve was greater than 0.9 and the model was of robustness. Two independent data sets in the Test group showed its area under the curve was 0.814 and 0.788 respectively. Therefore, the establishment of an artificial neural network model for the diagnosis of osteoarthritis has a certain diagnostic value.

Key words: osteoarthritis, diagnostic model, artificial neural network, machine learning

CLC Number:

Fan Yidong, Qin Gang, Su Guowei, Xiao Shifu, Liu Junliang, Li Weicai, Wu Guangtao. Establishment and analysis of osteoarthritis diagnosis model based on artificial neural networks[J]. Chinese Journal of Tissue Engineering Research, 2024, 28(16): 2550-2554.

Figures/Tables 10

References

[1] JANG S, LEE K, JU J H. Recent updates of diagnosis, pathophysiology, and treatment on osteoarthritis of the knee. Int J Mol Sci. 2021;22(5):2619.
[2] WANG J, SUN Y, LIU J, et al. Roles of long non‑coding RNA in osteoarthritis. Int J Mol Med. 2021;48(1):133.
[3] LIANG Y, LIN F, HUANG Y. Identification of Biomarkers Associated with Diagnosis of Osteoarthritis Patients Based on Bioinformatics and Machine Learning. J Immunol Res. 2022;2022:5600190.
[4] XIE NN, WANG FF, ZHOU J, et al. Establishment and analysis of a combined diagnostic model of polycystic ovary syndrome with random forest and artificial neural network. Biomed Res Int. 2020;2020:2613091.
[5] MAMOSHINA P, VIEIRA A, PUTIN E, et al. Applications of deep learning in biomedicine. Mol Pharm. 2016;13(5):1445-1454.
[6] WEI Q, FANG W, CHEN X, et al. Establishment and validation of a mathematical diagnosis model to distinguish benign pulmonary nodules from early non-small cell lung cancer in Chinese people. Transl Lung Cancer Res. 2020;9(5):1843-1852.
[7] WANG C, ZHAO Y, JIN B, et al. Development and validation of a predictive model for coronary artery disease using machine learning. Front Cardiovasc Med. 2021; 8:614204.
[8] LECUN Y, BENGIO Y, HINTON G. Deep learning. Nature. 2015;521(7553):436-444.
[9] TIAN Y, YANG J, LAN M, et al. Construction and analysis of a joint diagnosis model of random forest and artificial neural network for heart failure. Aging (Albany NY). 2020;12(24):26221-26235.
[10] SUN D, PENG H, WU Z. Establishment and analysis of a combined diagnostic model of alzheimer’s disease with random forest and artificial neural network. Front Aging Neurosci. 2022;14:921906.
[11] SHE J, SU D, DIAO R, et al. A Joint Model of Random Forest and Artificial Neural Network for the Diagnosis of Endometriosis. Front Genet. 2022;13:848116.
[12] LAMBERT C, DUBUC JE, MONTELL E, et al. Gene expression pattern of cells from inflamed and normal areas of osteoarthritis synovial membrane. Arthritis Rheumatol. 2014;66(4):960-968.
[13] WOETZEL D, HUBER R, KUPFER P, et al. Identification of rheumatoid arthritis and osteoarthritis patients by transcriptome-based rule set generation. Arthritis Res Ther. 2014;16(2):R84.
[14] BROPHY RH, ZHANG B, CAI L, et al. Transcriptome comparison of meniscus from patients with and without osteoarthritis. Osteoarthritis Cartilage. 2018;26(3): 422-432.
[15] CHOU CH, WU CC, SONG IW, et al. Genome-wide expression profiles of subchondral bone in osteoarthritis. Arthritis Res Ther. 2013;15(6):R190.
[16] LIU W, ZENG S, WU G, et al. Rice seed purity identification technology using hyperspectral image with LASSO logistic regression model. Sensors (Basel). 2021;21(13):4384.
[17] HUANG ML, HUNG YH, LEE WM, et al. SVM-RFE based feature selection and Taguchi parameters optimization for multiclass SVM classifier. ScientificWorldJournal. 2014;2014:795624.
[18] ACHARJEE A, LARKMAN J, XU Y, et al. A random forest based biomarker discovery and power analysis framework for diagnostics research. BMC Med Genomics. 2020;13(1):178.
[19] XIANG J, HUANG W, HE Y, et al. Construction of artificial neural network diagnostic model and analysis of immune infiltration for periodontitis. Front Genet. 2022; 13:1041524.
[20] WU Y, CHEN H, LI L, et al. Construction of novel gene signature-based predictive model for the diagnosis of acute myocardial infarction by combining random forest with artificial neural network. Front Cardiovasc Med. 2022;9:876543.
[21] YE T, HAOYUAN Z, BEI Z, et al. Exploration of biomarkers in osteoarthritis based on bioinformatics. Medicine (Baltimore). 2021;100(31):e26730.
[22] 陈宗锐,孙洁,姜萍.骨关节炎软骨代谢相关生物标志物的研究进展[J].医学综述,2022,28(12):2351-2356
[23] KUO YY, HUANG ST, CHIU HW. Applying artificial neural network for early detection of sepsis with intentionally preserved highly missing real-world data for simulating clinical situation. BMC Med Inform Decis Mak. 2021;21(1):290.
[24] BAXT WG. Application of artificial neural networks to clinical medicine. The lancet. 1995;346(8983):1135-1138.
[25] TANIGUCHI N, CARAMÉS B, RONFANI L, et al. Aging-related loss of the chromatin protein HMGB2 in articular cartilage is linked to reduced cellularity and osteoarthritis.Proc Natl Acad Sci U S A. 2009;106(4):1181-1186.
[26] 谢春意,刘晓辉,谢获，等.膝骨关节炎患者滑液和血清中HMGB2的研究[J].放射免疫学杂志,2012,25(6):675-677
[27] WARMINK K, SIEBELT M, LOW PS, et al. Folate Receptor Expression by Human Monocyte–Derived Macrophage Subtypes and Effects of Corticosteroids. Cartilage. 2022;13(1):19476035221081469.
[28] YI YS. Folate receptor-targeted diagnostics and therapeutics for inflammatory diseases. Immune Netw. 2016;16(6):337-343.
[29] BLOM AB, VAN LENT PL, HOLTHUYSEN AE, et al. Synovial lining macrophages mediate osteophyte formation during experimental osteoarthritis. Osteoarthritis Cartilage. 2004;12(8):627-635.
[30] LI RN, LIN YZ, PAN YC, et al. GADD45a and GADD45b genes in rheumatoid arthritis and systemic lupus erythematosus patients. J Clin Med. 2019;8(6):801.
[31] ZHAN Q. Gadd45a, a p53-and BRCA1-regulated stress protein, in cellular response to DNA damage. Mutat Res. 2005;569(1-2):133-143.
[32] HASHIMOTO S, NISHIYAMA T, HAYASHI S, et al. Role of p53 in human chondrocyte apoptosis in response to shear strain. Arthritis Rheum. 2009;60(8):2340-2349.
[33] JANG HJ, YANG JH, HONG E, et al. Chelidonine induces apoptosis via GADD45a-p53 regulation in human pancreatic cancer cells. Integr Cancer Ther. 2021;20:15347354211006191.
[34] JIN S, MAZZACURATI L, ZHU X, et al. Gadd45a contributes to p53 stabilization in response to DNA damage. Oncogene. 2003;22(52):8536-8540.
[35] AMR K, PAWLIKOWSKA P, AOUFOUCHI S, et al. Whole exome sequencing identifies a new mutation in the SLC19A2 gene leading to thiamine‐responsive megaloblastic anemia in an Egyptian family. Mol Genet Genomic Med. 2019; 7(7):e00777.
[36] LO PK, CHEN JY, TANG PP, et al. Identification of a mouse thiamine transporter gene as a direct transcriptional target for p53. J Biol Chem. 2001;276(40):37186-37193.
[37] STAVEROSKY JA, PRYCE BA, WATSON SS, et al. Tubulin polymerization‐promoting protein family member 3, Tppp3, is a specific marker of the differentiating tendon sheath and synovial joints. Dev Dyn. 2009;238(3):685-692.