收稿日期: 2026-04-15
修回日期: 2026-06-04
录用日期: 2026-06-08
网络出版日期: 2026-08-13
基金资助
国家自然科学基金项目(82572220);广东省自然科学基金项目(2024A1515011462)
Development of a disease activity assessment model for rheumatoid arthritis using ultrasound radiomics and deep learning
Received date: 2026-04-15
Revised date: 2026-06-04
Accepted date: 2026-06-08
Online published: 2026-08-13
目的 构建并初步评价融合影像组学和深度学习特征的超声整合模型在类风湿关节炎(RA)疾病活动度客观评估的应用价值。 方法 纳入92例RA患者的184个受累关节,按8∶2分为训练集和测试集。经感兴趣区域勾画、特征提取筛选及关节切面声像图裁剪后,构建超声评分、影像组学、深度学习3个单一模型,及超声-影像组学、超声-影像组学-深度学习2个联合模型,应用受试者工作特征(ROC)曲线、曲线下面积(AUC)及临床决策(DCA)曲线评估模型性能。 结果 测试集中,相比单一特征的超声评分模型(AUC = 0.832,95%CI:0.709 ~ 0.954)、影像组学模型(AUC = 0.824,95%CI:0.701 ~ 0.947)或深度学习特征模型(AUC = 0.788,95%CI:0.642 ~ 0.933),超声-影像组学-深度学习联合模型(AUC = 0.901,95%CI:0.815 ~ 0.988)的性能最佳,拟合度较好(拟合度检验P > 0.05),DCA曲线分析表明在一定阈值内具有更大临床净收益。 结论 该多模态融合量化模型对RA疾病活动度评估效果较好,为未来RA的分层治疗策略制定与动态疗效监测奠定方法学基础。
朱欣 , 梁杨标 , 肖雄 , 谢佳霖 , 朱俊燕 , 陈彦 . 基于超声影像组学与深度学习的类风湿关节炎疾病活动度评估模型构建[J]. 实用医学杂志, 2026 , 42(15) : 2786 -2797 . DOI: 10.3969/j.issn.1006-5725.2026.15.015
Objective To develop and preliminarily evaluate the value of an integrated ultrasound model that combines radiomics and deep learning features for the objective assessment of disease activity in rheumatoid arthritis (RA). Methods A total of 184 affected joints from 92 RA patients were included and then divided into training and testing sets at an 8:2 ratio. After the delineation of the region-of-interest (ROI), the extraction and screening of features, and the cropping of joint cross-sectional images, three standalone models, namely ultrasound scoring, bioinformatics, and deep learning, were constructed. Additionally, two combined models, ultrasound-bioinformatics and ultrasound-bioinformatics-deep learning, were also built. The performance of the models was evaluated using receiver operating characteristic (ROC) curves, the area under the curve (AUC), and decision-making analysis (DCA) curves. Results In the test set, when compared with the ultrasound scoring model based on single features (AUC = 0.832, 95% CI: 0.709 - 0.954), the radiomics model (AUC = 0.824, 95% CI: 0.701 - 0.947), or the deep learning feature model (AUC = 0.788, 95% CI: 0.642 - 0.933), the combined ultrasound-radiomics-deep learning model (AUC = 0.901, 95% CI: 0.815 - 0.988) exhibited the best performance. The Hosmer-Lemeshow goodness-of-fit test for the combined ultrasound-radiomics-deep learning model yielded a p-value greater than 0.05, indicating adequate model calibration. Moreover, DCA showed that the combined model provides a higher net clinical benefit within a defined range of threshold probabilities. Conclusion This multimodal fusion quantification model exhibits excellent efficacy in evaluating the activity of RA disease, establishing a methodological basis for the formulation of future stratified treatment strategies and dynamic efficacy monitoring for RA.
Key words: rheumatoid arthritis; ultrasound; radiomics; deep learning
| [1] | KLARESKOG L, CATRINA A I, PAGET S. Rheumatoid arthritis[J]. Lancet, 2009, 373(9664): 659-672. doi:10.1016/S0140-6736(09)60008-8 . |
| [2] | MCINNES I B, SCHETT G. The pathogenesis of rheumatoid arthritis[J]. N Engl J Med, 2011, 365(23): 2205-2219. doi:10.1056/nejmra1004965 . |
| [3] | FINCKH A, GILBERT B, HODKINSON B, et al. Global epidemiology of rheumatoid arthritis[J]. Nat Rev Rheumatol, 2022, 18(10): 591-602. doi:10.1038/s41584-022-00827-y . |
| [4] | VAN DER WOUDE D, VAN DER HELM-VAN MIL A H M. Update on the epidemiology, risk factors, and disease outcomes of rheumatoid arthritis[J]. Best Pract Res Clin Rheumatol, 2018, 32(2): 174-187. doi:10.1016/j.berh.2018.10.005 . |
| [5] | ALETAHA D, SMOLEN J S. Diagnosis and management of rheumatoid arthritis: A review[J]. JAMA, 2018, 320(13): 1360-1372. doi:10.1001/jama.2018.13103 . |
| [6] | LI Z, WANG X Q. Clinical effect and biological mechanism of exercise for rheumatoid arthritis: A mini review[J]. Front Immunol, 2023, 13: 1089621. doi:10.3389/fimmu.2022.1089621 . |
| [7] | FRAENKEL L, BATHON J M, ENGLAND B R, et al. 2021 American college of rheumatology guideline for the treatment of rheumatoid arthritis[J]. Arthritis Rheumatol, 2021, 73(7): 1108-1123. doi:10.1002/art.41752 . |
| [8] | BROWN P, PRATT A G, HYRICH K L. Therapeutic advances in rheumatoid arthritis[J]. Bmj, 2024, 384: e070856. doi:10.1136/bmj-2022-070856 . |
| [9] | SMOLEN J S, BREEDVELD F C, BURMESTER G R, et al. Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force[J]. Ann Rheum Dis, 2016, 75(1): 3-15. doi:10.1136/annrheumdis-2015-207524 . |
| [10] | ANDERSON J K, ZIMMERMAN L, CAPLAN L, et al. Measures of rheumatoid arthritis disease activity: Patient (PtGA) and Provider (PrGA) Global Assessment of Disease Activity, Disease Activity Score (DAS) and Disease Activity Score With 28-Joint Counts (DAS28), Simplified Disease Activity Index (SDAI), Clinical Disease Activity Index (CDAI), Patient Activity Score (PAS) and Patient Activity Score-II (PASII), Routine Assessment of Patient Index Data (RAPID), Rheumatoid Arthritis Disease Activity Index (RADAI) and Rheumatoid Arthritis Disease Activity Index-5 (RADAI-5), Chronic Arthritis Systemic Index (CASI), Patient-Based Disease Activity Score With ESR (PDAS1) and Patient-Based Disease Activity Score Without ESR (PDAS2), and Mean Overall Index for Rheumatoid Arthritis (MOI-RA)[J]. Arthritis Care Res, 2011, 63(S11): S14-S36. doi:10.1002/acr.20621 . |
| [11] | ENGLAND B R, TIONG B K, BERGMAN M J, et al. 2019 update of the American college of rheumatology recommended rheumatoid arthritis disease activity measures[J]. Arthritis Care Res, 2019, 71(12): 1540-1555. doi:10.1002/acr.24042 . |
| [12] | OHRNDORF S, BOER A C, BOETERS D M, et al. Do musculoskeletal ultrasound and magnetic resonance imaging identify synovitis and tenosynovitis at the same joints and tendons? A comparative study in early inflammatory arthritis and clinically suspect arthralgia[J]. Arthritis Res Ther, 2019, 21(1): 59. doi:10.1186/s13075-019-1824-z . |
| [13] | KELLNER D A, MORRIS N T, LEE S M, et al. Clinical utility of ultrasound and MRI in rheumatoid arthritis: An expert review[J]. Best Pract Res Clin Rheumatol, 2025, 39(3): 102072. doi:10.1016/j.berh.2025.102072 . |
| [14] | 耿研, 张卓莉. 影像学技术在类风湿关节炎诊治中的应用[J/OL]. 中国医学前沿杂志(电子版), 2020, 12(11): 13-18. doi:10.12037/YXQY.2020.11-04 . |
| [15] | D’AGOSTINO M A, TERSLEV L, WAKEFIELD R, et al. Novel algorithms for the pragmatic use of ultrasound in the management of patients with rheumatoid arthritis: From diagnosis to remission[J]. Ann Rheum Dis, 2016, 75(11): 1902-1908. doi:10.1136/annrheumdis-2016-209646 . |
| [16] | YOSHIDA Y, HAMMER H B, SEXTON J, et al. How to score the wrist joint by ultrasound: Exploration of the most sensitive joint assessment during follow-up of early or established rheumatoid arthritis[J]. Arthritis Res Ther, 2025, 27(1): 200. doi:10.1186/s13075-025-03668-8 . |
| [17] | CATE D F TEN, LUIME J J, SWEN N, et al. Role of ultrasonography in diagnosing early rheumatoid arthritis and remission of rheumatoid arthritis - a systematic review of the literature[J]. Arthritis Res Ther, 2013, 15(1): R4. doi:10.1186/ar4132 . |
| [18] | 马艳, 金莉, 李向培, 等. 高频超声在类风湿关节炎双手关节病变的应用价值[J]. 实用医学杂志, 2016, 32(11): 1810-1812. doi:10.3969/j.issn.1006-5725.2016.11.025 . |
| [19] | COLEBATCH A N, EDWARDS C J, ?STERGAARD M, et al. EULAR recommendations for the use of imaging of the joints in the clinical management of rheumatoid arthritis[J]. Ann Rheum Dis, 2013, 72(6): 804-814. doi:10.1136/annrheumdis-2012-203158 . |
| [20] | M?LLER I, JANTA I, BACKHAUS M, et al. The 2017 EULAR standardised procedures for ultrasound imaging in rheumatology[J]. Ann Rheum Dis, 2017, 76(12): 1974-1979. doi:10.1136/annrheumdis-2017-211585 . |
| [21] | 郭文静, 吕海霞, 彭锦锦, 等. 超声新型评分模型预测类风湿关节炎常规内科治疗短期疗效的临床价值[J]. 临床超声医学杂志, 2025, 27(7): 572-578. doi:10.3969/j.issn.1008-6978.2025. 07.009 . |
| [22] | TERSLEV L, ELLEGAARD K, CHRISTENSEN R, et al. Head-to-head comparison of quantitative and semi-quantitative ultrasound scoring systems for rheumatoid arthritis: Reliability, agreement and construct validity[J]. Rheumatology, 2012, 51(11): 2034-2038. doi:10.1093/rheumatology/kes124 . |
| [23] | D′AGOSTINO M A, TERSLEV L, AEGERTER P, et al. Scoring ultrasound synovitis in rheumatoid arthritis: A EULAR-OMERACT ultrasound taskforce—Part 1: Definition and development of a standardised, consensus-based scoring system[J]. RMD Open, 2017, 3(1): e000428. doi:10.1136/rmdopen-2016-000428 . |
| [24] | TERSLEV L, NAREDO E, AEGERTER P, et al. Scoring ultrasound synovitis in rheumatoid arthritis: A EULAR-OMERACT ultrasound taskforce-Part 2: Reliability and application to multiple joints of a standardised consensus-based scoring system[J]. RMD Open, 2017, 3(1): e000427. doi:10.1136/rmdopen-2016-000427 . |
| [25] | GADEHOLT O. Forward to the past: Ultrasound might be necessary in some patients with rheumatoid arthritis[J]. Ann Rheum Dis, 2019, 78(6): e56. doi:10.1136/annrheumdis-2018-213278 . |
| [26] | LAMBIN P, RIOS-VELAZQUEZ E, LEIJENAAR R, et al. Radiomics: Extracting more information from medical images using advanced feature analysis[J]. Eur J Cancer, 2012, 48(4): 441-446. doi:10.1016/j.ejca.2011.11.036 . |
| [27] | LAMBIN P, LEIJENAAR R T H, DEIST T M, et al. Radiomics: The bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749-762. doi:10.1038/nrclinonc.2017.141 . |
| [28] | GILLIES R J, KINAHAN P E, HRICAK H. Radiomics: Images are more than pictures, they are data[J]. Radiology, 2016, 278(2): 563-577. doi:10.1148/radiol.2015151169 . |
| [29] | TONG Y, ZHANG J, WEI Y, et al. Ultrasound-based radiomics analysis for preoperative prediction of central and lateral cervical lymph node metastasis in papillary thyroid carcinoma: A multi-institutional study[J]. BMC Med Imag, 2022, 22(1): 82. doi:10.1186/s12880-022-00809-2 . |
| [30] | MAO B, ZHANG L, NING P, et al. Preoperative prediction for pathological grade of hepatocellular carcinoma via machine learning-based radiomics[J]. Eur Radiol, 2020, 30(12): 6924-6932. doi:10.1007/s00330-020-07056-5 . |
| [31] | QI Y J, SU G H, YOU C, et al. Radiomics in breast cancer: Current advances and future directions[J]. Cell Rep Med, 2024, 5(9): 101719. doi:10.1016/j.xcrm.2024.101719 . |
| [32] | LITJENS G, KOOI T, BEJNORDI B E, et al. A survey on deep learning in medical image analysis[J]. Med Image Anal, 2017, 42: 60-88. doi:10.1016/j.media.2017.07.005 . |
| [33] | WANG H J, SU C P, LAI C C, et al. Deep learning-based computer-aided diagnosis of rheumatoid arthritis with hand X-ray images conforming to modified total sharp/van der heijde score[J]. Biomedicines, 2022, 10(6): 1355. doi:10.3390/biomedicines10061355 . |
| [34] | ANDERSEN J K H, PEDERSEN J S, LAURSEN M S, et al. Neural networks for automatic scoring of arthritis disease activity on ultrasound images[J]. RMD Open, 2019, 5(1): e000891. doi:10.1136/rmdopen-2018-000891 . |
| [35] | YAN L, XU J, YE X, et al. Development and validation of ultrasound-based radiomics deep learning model to identify bone erosion in rheumatoid arthritis[J]. Clin Rheumatol, 2025, 44(7): 2635-2645. doi:10.1007/s10067-025-07481-1 . |
| [36] | HE X, WANG M, ZHAO C, et al. Deep learning-based automatic scoring models for the disease activity of rheumatoid arthritis based on multimodal ultrasound images[J]. Rheumatology, 2024, 63(3): 866-873. doi:10.1093/rheumatology/kead366 . |
| [37] | VAN VOLLENHOVEN R. Treat-to-target in rheumatoid arthritis—are we there yet?[J]. Nat Rev Rheumatol, 2019, 15(3): 180-186. doi:10.1038/s41584-019-0170-5 . |
| [38] | CHIU Y M, CHEN D Y. Biological ageing: A promising target for prevention and management of rheumatoid arthritis[J]. Lancet Healthy Longev, 2024, 5(1): e6-e7. doi:10.1016/S2666-7568(23)00239-8 . |
| [39] | GBD 2021 Rheumatoid Arthritis Collaborators.. Global, regional, and national burden of rheumatoid arthritis, 1990-2020, and projections to 2050: A systematic analysis of the Global Burden of Disease Study 2021[J]. Lancet Rheumatol, 2023, 5(10): e594-e610. doi:10.1016/S2665-9913(23)00211-4 . |
| [40] | CASTRO A, DIAZ J, KUMAR S. FRI0007?Performance of musculoskeletal ultrasonography (MSUS) and optimal cutoff criteria for patients at risk for rheumatoid arthritis: A meta-analysis[J]. Ann Rheum Dis, 2018, 77: 550-551. doi:10.1136/annrheumdis-2018-eular.1752 . |
| [41] | SILVAGNI E, ZANDONELLA CALLEGHER S, MAURIC E, et al. Musculoskeletal ultrasound for treating rheumatoid arthritis to target-a systematic literature review[J]. Rheumatology, 2022, 61(12): 4590-4602. doi:10.1093/rheumatology/keac261 . |
| [42] | KONDO Y, SUZUKI K, INOUE Y, et al. Significant association between joint ultrasonographic parameters and synovial inflammatory factors in rheumatoid arthritis[J]. Arthritis Res Ther, 2019, 21(1): 14. doi:10.1186/s13075-018-1802-x . |
| [43] | MCINNES I B, BUCKLEY C D, ISAACS J D. Cytokines in rheumatoid arthritis—shaping the immunological landscape[J]. Nat Rev Rheumatol, 2016, 12(1): 63-68. doi:10.1038/nrrheum. 2015.171 . |
| [44] | WRIGHT H L, LYON M, CHAPMAN E A, et al. Rheumatoid arthritis synovial fluid neutrophils drive inflammation through production of chemokines, reactive oxygen species, and neutrophil extracellular traps[J]. Front Immunol, 2021, 11: 584116. doi:10.3389/fimmu.2020.584116 . |
| [45] | BURKE C J, ALIZAI H, BELTRAN L S, et al. MRI of synovitis and joint fluid[J]. J Magn Reson Imag, 2019, 49(6): 1512-1527. doi:10.1002/jmri.26618 . |
| [46] | KISO T, OKADA Y, KAWATA S, et al. Ultrasound-based radiomics and machine learning for enhanced diagnosis of knee osteoarthritis: Evaluation of diagnostic accuracy, sensitivity, specificity, and predictive value[J]. Eur J Radiol Open, 2025, 14: 100649. doi:10.1016/j.ejro.2025.100649 . |
| [47] | LIN M, YAN L, HE M, et al. Development and validation of an ultrasound-based clinical radiomics nomogram for diagnosing gouty arthritis[J]. Ultrasound Med Biol, 2025, 51(4): 650-660. doi:10.1016/j.ultrasmedbio.2024.12.009 . |
| [48] | BOESEN M, KUBASSOVA O, BOUERT R, et al. Correlation between computer-aided dynamic gadolinium-enhanced MRI assessment of inflammation and semi-quantitative synovitis and bone marrow oedema scores of the wrist in patients with rheumatoid arthritis—a cohort study[J]. Rheumatology (Oxford), 2012, 51(1): 134-143. doi:10.1093/rheumatology/ker220 . |
| [49] | YU F H, MIAO S M, LI C Y, et al. Pretreatment ultrasound-based deep learning radiomics model for the early prediction of pathologic response to neoadjuvant chemotherapy in breast cancer[J]. Eur Radiol, 2023, 33(8): 5634-5644. doi:10.1007/s00330-023-09555-7 . |
| [50] | AGYEKUM E A, KONG W, AGYEKUM D N, et al. Ultrasound derived deep learning features for predicting axillary lymph node metastasis in breast cancer using graph convolutional networks in a multicenter study[J]. Sci Rep, 2025, 15: 27796. doi:10.1038/s41598-025-13086-0 . |
| [51] | 罗为尧, 范誉铧, 黎一夫, 等. 基于深度学习超声模型对乳腺癌分子分型与新辅助治疗早期最佳疗效时机的预测[J]. 实用医学杂志, 2026, 42(9): 1501-1510. doi: 10.3969/j.issn.1006-5725.2026.09.003 . |
| [52] | REYES M, MEIER R, PEREIRA S, et al. On the interpretability of artificial intelligence in radiology: Challenges and opportunities[J]. Radiol Artif Intell, 2020, 2(3): e190043. doi:10.1148/ryai.2020190043 . |
| [53] | WU L, LI S, WU C, et al. Ultrasound-based deep learning radiomics nomogram for differentiating mass mastitis from invasive breast cancer[J]. BMC Med Imag, 2024, 24(1): 189. doi:10.1186/s12880-024-01353-x . |
| [54] | LIANG W, YU H, DUAN L, et al. MRI-based 2.5D deep learning radiomics nomogram for the differentiation of benign versus malignant vertebral compression fractures[J/OL]. Frontiers in Oncology, 2025, 15: 1603672[2025-08-11]. doi:10.3389/fonc. 2025.1603672 . |
| [55] | 杜阳春, 郑红雨, 陈海宁, 等. 深度学习超声影像组学列线图模型鉴别Ⅰ和Ⅱ型上皮性卵巢癌[J]. 实用医学杂志, 2025, 41(18): 2920-2927. doi:10.3969/j.issn.1006-5725.2025.18.020 . |
/
| 〈 |
|
〉 |