Radiomics for diagnostic value of osteoporosis in female patients based on lumbar magnetic resonance imaging

doi:10.3969/j.issn.2095-4344.2846

Abstract

Abstract:

BACKGROUND: Magnetic resonance imaging (MRI) plays an increasingly important role in the diagnosis of osteoporosis. As a new method of image analysis, radiomics has potential clinical significance in the diagnosis of osteoporosis.

OBJECTIVE: To explore the diagnostic value of osteoporosis based on lumbar MRI imaging model.

METHODS: Fifty female patients who underwent both lumbar-spine MRI and dual-energy X-ray absorptiometry in Outpatient Department of First Hospital of Hebei Medical University from February 2017 to October 2018 at the age of 40-84 years were enrolled in this study. Dual-energy X-ray absorptiometry showed 28 osteoporosis patients, and 22 normal bone mineral density. In the sagittal plane T1WI and T2WI images of lumbar MRI, five consecutive layers in the middle of L₂_-₄vertebral body were selected for image segmentation; radiomics features were extracted; diagnostic models were constructed. Combined with clinical risk factors, the clinical-radiomics model was constructed. The area under the receiver operating characteristic curve was used to assess the model performance. The experiment was approved by the Ethics Committee of First Hospital of Hebei Medical University.

RESULTS AND CONCLUSION: (1) The 396 × 3 features were extracted from T1WI, T2WI and combined sequences respectively in 50 subjects. After feature dimensionality reduction, four features of T1WI sequence, six features of T2WI sequence and four features of T1WI and T2WI combined sequence were screened out respectively. After diagnostic model establishment, the areas under the receiver operating characteristic curve of T1WI and T2WI models were 0.810 and 0.820, respectively. The area under the receiver operating characteristic curve of combined T1WI+T2WI model was 0.937, which was higher than that of a independent sequence. (2) Combining the T1WI + T 2WI radiomics features with clinical data, the diagnostic model suitable for female combined with clinical factors was constructed, and the area under the receiver operating characteristic curve of clinical-radiomics model was 0.960. (3) The results showed that the radiomics features based on routine lumbar MRI can distinguish osteoporosis from normal bone mass, and the clinical-radiomics diagnosis model based on clinical and radiomics features can improve the diagnostic efficiency, which is valuable for diagnosis of osteoporosis.

Key words: bone, imaging histology, lumbar spine, magnetic resonance, osteoporosis, bone mineral density, model

CLC Number:

He Li, Liu Zhai, Gao Zhimei, Liu Chunying, Zhao Junlu, Ren Qingyun. Radiomics for diagnostic value of osteoporosis in female patients based on lumbar magnetic resonance imaging[J]. Chinese Journal of Tissue Engineering Research, 2020, 24(30): 4841-4846.

Figures/Tables 7

References

[1] 马远征,王以朋,刘强,等.中国老年骨质疏松症诊疗指南(2018)[J].中华健康管理学杂志,2018,12(6):484-489.

[2] LINK TM. Osteoporosis imaging: state of the art and advanced imaging.Radiology. 2012;263(1):3-17.

[3] HOFBAUER LC, RACHNER TD. More DATA to guide sequential osteoporosis therapy.Lancet.2015;386:1116-1118．

[4] PISANI P, RENNA MD, CONVERSANO F, et al. Major osteoporotic fragility fractures: Risk factor updates and societal impact.World J Orthop.2016;7(3):171-181.

[5] 韩亚军,帖小佳,伊力哈木·托合提.中国中老年人骨质疏松症患病率的Meta分析[J].中国组织工程研究,2014,18(7): 1129-1134.

[6] MILLER PD. Underdiagnosis and Undertreatment of Osteoporosis: The Battle to Be Won. J Clin Endocrinol Metab. 2016;101(3):852-859.

[7] SAAD MM, AHMED TA, MOHAMED KE, et al. Role of lumbar spine signal intensity measurement by MRI in the diagnosis of osteoporosis in post-menopausal women. Egyptian Journal of Radiology and Nuclear Medicine.2019;50(1):35-42.

[8] BANDIRALI M, DI LEO G, PAPINI GD, et al.A new diagnostic score to detect osteoporosis in patients undergoing lumbar spine MRI.Eur Radiol.2015;25(10):2951-2959.

[9] KUMAR V, GU Y, BASU S, et al. Radiomics: the process and the challenges.MagnReson Imaging.2012;30(9):1234-1248.

[10] E L, LU L, LI L, et al. Radiomics for Classification of Lung Cancer Histological Subtypes Based on Nonenhanced Computed Tomography.Acad Radiol.2019;26(9):1245-1252.

[11] HORVAT N, VEERARAGHAVAN H, KHAN M, et al. MR Imaging of Rectal Cancer: Radiomics Analysis to Assess Treatment Response after Neoadjuvant Therapy.Radiology. 2018;287(3):833-843.

[12] 吴亚平,刘博,顾建钦,等.基于影像组学的脑胶质瘤分级方法[J].中华放射学杂志,2017,51(12):902-905.

[13] 袁清玉,江玉明,吕闻冰,等.基于18F-FDG PET/CT图像的影像组学列线图对胃癌术后的预后评估[J].中华核医学与分子影像杂志, 2019,39(1):2-5.

[14] 王超,刘侠,董迪,等.基于影像组学的非小细胞肺癌淋巴结转移预测[J].自动化学报,2019,45(6):1087-1093.

[15] MACKAY JW, MURRAY PJ, KASMAI B, et al. Subchondral bone in osteoarthritis: association between MRI texture analysis and histomorphometry.Osteoarthritis Cartilage. 2017;25(5):700-707.

[16] AREECKAL AS, JAYASHEELAN N, KAMATH J, et al. Early diagnosis of osteoporosis using radiogrammetry and texture analysis from hand and wrist radiographs in Indian population. Osteoporos Int.2018;29(3):665-673.

[17] TABARI A, TORRIANI M, MILLER KK, et al. Anorexia Nervosa: Analysis of Trabecular Texture with CT.Radiology. 2017;283(1):178-185.

[18] KANIS JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group.Osteoporos Int. 1994;4(6): 368-381.

[19] 王平,和建伟,黄刚,等.应用双能CT与定量CT对椎体骨密度测量的对照研究[J].中国骨质疏松杂志,2017,23(2):159-162.

[20] DEYO RA, WEINSTEIN JN. Low back pain. N Engl J Med. 2001;344(5):363-370.

[21] MANENTI G, CAPUANI S, FUSCO A, et al.Osteoporosis detection by 3T diffusion tensor imaging and MRI spectroscopy in women older than 60 years.Aging Clin Exp Res.2013;25 Suppl 1:S31-S34.

[22] LI GW, XU Z, CHEN QW, et al. Quantitative evaluation of vertebral marrow adipose tissue in postmenopausal female using MRI chemical shift-based water-fat separation.Clin Radiol.2014;69(3):254-262.

[23] 何杰,方浩,李晓娜,等.腰椎MR扩散加权成像对骨质疏松的定量诊断价值[J].临床放射学杂志,2015(5):95-99.

[24] GRIFFITH JF, YEUNG DK, ANTONIO GE, et al. Vertebral bone mineral density, marrow perfusion, and fat content in healthy men and men with osteoporosis: dynamic contrast-enhanced MR imaging and MR spectroscopy. Radiology. 2005;236(3):945-951.

[25] 常荣,张红,刘正华,等.MRI m-Dixon-Quant技术快速评估中老年骨质疏松症的可行性研究[J].实用放射学杂志,2018,34(12): 1908-1911.

[26] 郭志,许崇永.IDEAL-IQ技术在2型糖尿病性骨质疏松症诊断中的应用价值[J].中国中西医结合影像学杂志,2019,17(5): 489-492.

[27] SCHWARTZ AV. Marrow fat and bone: review of clinical findings.Front Endocrinol (Lausanne).2015;6:40.

[28] QIU X, FU Y, CHEN J, et al.The Correlation between Osteoporosis and Blood Circulation Function Based on Magnetic Resonance Imaging. J Med Syst.2019;43(4):91.

[29] BURIAN E, SUBBURAJ K, MOOKIAH MRK, et al.Texture analysis of vertebral bone marrow using chemical shift encoding-based water-fat MRI: a feasibility study. Osteoporos Int.2019;30(6):1265-1274.

[30] 翁子敬,黄学菁,俞健力,等.常规MRI信号评价腰椎骨质疏松的意义[J].中国组织工程研究,2018,22(35):5667-5673.

[31] LECLER A, DURON L, BALVAY D, et al.Combining Multiple Magnetic Resonance Imaging Sequences Provides Independent Reproducible Radiomics Features.Sci Rep. 2019;9(1):2068.

[32] 刘伯亮,潘万敏.骨质疏松发生与年龄、性别的关系:5200例分析[J].中国临床康复,2004,8(33):139-140.

[33] BLACK DM, ROSEN CJ. Clinical Practice. Postmenopausal Osteoporosis.N Engl J Med. 2016;374(3):254-262.

[34] FERIZI U, BESSER H, HYSI P, et al.Artificial Intelligence Applied to Osteoporosis: A Performance Comparison of Machine Learning Algorithms in Predicting Fragility Fractures From MRI Data.J MagnReson Imaging. 2019;49(4): 1029-1038.