切换至 "中华医学电子期刊资源库"

中华老年骨科与康复电子杂志 ›› 2026, Vol. 12 ›› Issue (03) : 143 -149. doi: 10.3877/cma.j.issn.2096-0263.2026.03.003

康复医学

基于卷积神经网络的X线在椎体压缩性骨折康复中的应用研究
李鑫1, 宋晓宇2, 李光磊3,()   
  1. 1150040 哈尔滨市第五医院放射科
    2150040 哈尔滨市第五医院核磁科
    3150040 哈尔滨,黑龙江省医院康复科
  • 收稿日期:2025-03-27 出版日期:2026-06-05
  • 通信作者: 李光磊
  • 基金资助:
    黑龙江省卫生健康委科研项目(20240909010376)

Application of convolutional neural network based X-ray in rehabilitation of vertebral compression fractures

Xin Li1, Xiaoyu Song2, Guanglei Li3,()   

  1. 1Department of Radiology, Harbin Fifth Hospital, Harbin 150040, China
    2Department of MRI, Harbin Fifth Hospital, Harbin 150040, China
    3Rehabilitation Department of Heilongjiang Provincial Hospital, Harbin 150040, China
  • Received:2025-03-27 Published:2026-06-05
  • Corresponding author: Guanglei Li
引用本文:

李鑫, 宋晓宇, 李光磊. 基于卷积神经网络的X线在椎体压缩性骨折康复中的应用研究[J/OL]. 中华老年骨科与康复电子杂志, 2026, 12(03): 143-149.

Xin Li, Xiaoyu Song, Guanglei Li. Application of convolutional neural network based X-ray in rehabilitation of vertebral compression fractures[J/OL]. Chinese Journal of Geriatric Orthopaedics and Rehabilitation(Electronic Edition), 2026, 12(03): 143-149.

目的

探讨基于卷积神经网络的X线在椎体压缩性骨折康复中的应用效果。

方法

前瞻性选取本院2022年1月至2023年6月就诊的130例椎体压缩性骨折患者,随机将其分为常规X线辅助PVP治疗(A)组,基于卷积神经网络X线辅助PVP治疗(B)组,每组65例,观察两组康复效果。

结果

130例患者均获得完整随访,随访率100.00%,随访时间1~12个月,平均(7.33±1.95)个月。其中A组随访时间1~12个月,平均(7.48±1.58)个月;B组随访时间1~12个月,平均(7.35±1.68)个月。与A组对比,B组X线曝光次数、骨水泥注入量、术中失血量、手术时间、住院天数均降低(P<0.05);治疗前与A组相比,B组伤椎前缘高度、伤椎后凸Cobb角对比差异无统计学意义(P>0.05),治疗后,两组伤椎前缘高度升高,伤椎后凸Cobb角降低,但B组与A组相比差异无统计学意义(P>0.05);术前与A组相比,B组VAS评分对比差异无统计学意义(P>0.05),术后24 h,两组VAS评分升高(P<0.05),且A组升高明显(P<0.05),术后12个月,两组VAS评分降低(P<0.05),且B组比A组降低明显(P<0.05);术前与A组相比,B组ODI评分对比差异无统计学意义(P>0.05),术后24 h及术后12个月,两组ODI评分均降低(P<0.05),且B组与A组相比降低明显(P<0.05);与A组比较,B组骨水泥渗漏、伤椎再骨折、邻近椎体骨折等并发症人数降低,B组总不良反应率(6.15%)低于A组(20.00%),组间差异有统计学意义(P<0.05)。

结论

基于卷积神经网络的X线对椎体压缩性骨折具有显著疗效,可显著提高患者椎体成形术后康复效果,值得临床应用。

Objective

To explore the application effect of X-ray based on convolutional neural network in the rehabilitation of vertebral compression fractures.

Methods

From January 2022 to June 2023, 130 patients with vertebral compression fracture were selected from our hospital and randomly divided into routine X-ray-assisted PVP treatment (A) group and convolutional neural network X-ray-assisted PVP treatment (B) group, with 65 cases in each group, to observe the rehabilitation effects of the two groups.

Results

All 130 patients were followed up completely, with a follow-up rate of 100.00%. The follow-up period ranged from 1 to 12 months, with an average of (7.33±1.95) months. Among them, the follow-up time of group A ranged from 1 to 12 months, with an average of (7.48±1.58) months. The follow-up period of Group B ranged from 1 to 12 months, with an average of (7.35±1.68) months. Compared with group A, the number of X-ray exposure, bone cement injection amount, intraoperative blood loss, operative time and hospitalization days in group B were significantly decreased (P<0.05). Compared with group A before treatment, there were no significant differences in anterior vertebra height and kyphotic Cobb angle between group A and group B (P>0.05). After treatment, the anterior vertebra height and kyphotic Cobb Angle of the two groups were significantly increased, and the Cobb Angle of the injured vertebra was significantly decreased, but there was no significant difference between group B and group A (P>0.05). Before surgery, there was no significant difference in the VAS score of group B compared with group A (P>0.05); 24 h after surgery, the VAS score of the two groups was significantly increased (P<0.05), and the VAS score of group A was significantly increased (P<0.05); 12 months after surgery, the VAS score of the two groups was significantly decreased (P<0.05), and the VAS score of group B was significantly lower than that of group A (P<0.05); Before surgery, there was no significant difference in ODI score between group B and group A (P>0.05). 24 h and 12 months after surgery, ODI score of both groups was significantly decreased (P<0.05), and that of group B was significantly decreased compared with group A (P<0.05). Compared with group A, the number of complications such as bone cement leakage, refracture of injured vertebra and fracture of adjacent vertebra in group B was significantly reduced, and the total adverse reaction rate in group B (6.15%) was significantly lower than that in group A (20.00%), with significant differences among groups (P<0.05).

Conclusion

X-ray based on convolutional neural network has a significant effect on vertebral compression fractures, and can significantly improve the rehabilitation effect of patients after vertebraplasty, which is worthy of clinical application.

表1 两组椎体压缩性骨折患者临床资料对比
表2 两组椎体压缩性骨折患者X线曝光次数、骨水泥注入量、术中失血量、手术时间、住院天数比较(±s
表3 两组椎体压缩性骨折患者伤椎前缘高度、伤椎后凸Cobb角比较(±s
表4 两组椎体压缩性骨折患者VAS评分比较(分,±s
表5 两组椎体压缩性骨折患者ODI评分比较(分,±s
图1 女,45岁,DR检查腰椎正侧位,胸12椎体变扁,提示椎体压缩性骨折伴椎体后凸畸形,胸腰段退行性变(图1A-B),腰椎正侧位,胸12椎体成形术后,骨水泥注入术后改变,胸腰段后凸畸形未改善(图1C-D)  图2 男,52岁,CT多平面重建腰1椎体压缩骨折,椎体中部横行致密带  图3 男,63岁,CT三维重建腰1椎体成形术后,骨水泥注入术后改变
表6 两组椎体压缩性骨折患者术后并发症发生情况比较[例(%)]
1
Dibs K, Facer B, Mageswaran P, et al. Vertebral compression fracture after spine stereotactic body radiotherapy: The Role of Vertebral Endplate Disruption [J]. Neurosurgery, 2024, 94(4): 797-804.
2
El-Ghandour NMF. Commentary: Vertebral Compression Fracture After Spine Stereotactic Body Radiotherapy: The Role of Vertebral Endplate Disruption [J]. Neurosurgery, 2024, 94(4): e50-e51.
3
Haibier A, Yusufu A, Lin H, Kayierhan A, Abudukelimu Y, Aximu A, Abudurexiti T. Effect of different cement distribution in bilateral and unilateral Percutaneous vertebro plasty on the clinical efficacy of vertebral compression fractures [J]. BMC Musculoskelet Disord, 2023, 24(1): 908.
4
Qi Z, Zhao S, Li H, et al. A study on vertebral refracture and scoliosis after percutaneous kyphoplasty in patients with osteoporotic vertebral compression fractures [J]. J Orthop Surg Res, 2024, 19(1): 302.
5
Sharif A, Nathani KR, Nguyen R, et al. Percutaneous curved vertebroplasty versus unipedicular vertebroplasty for osteoporotic vertebral compression fractures: a systematic review and meta-analysis [J]. Neurosurg Rev, 2025, 48(1): 410.
6
Ge J, Chen K, Xu P, et al. Percutaneous vertebroplasty by two-step fluoroscopy: a treatment for osteoporotic compression fractures of thoracic vertebrae in older adults [J]. BMC Musculoskelet Disord, 2025, 26(1): 135.
7
Monchka BA, Schousboe JT, Davidson MJ, et al. Development of a manufacturer-independent convolutional neural network for the automated identification of vertebral compression fractures in vertebral fracture assessment images using active learning [J]. Bone, 2022, 161: 116427.
8
Del Lama RS, Candido RM, Chiari-Correia NS, et al. Computer-Aided Diagnosis of Vertebral Compression Fractures Using Convolutional Neural Networks and Radiomics [J]. J Digit Imaging, 2022, 35(3): 446-458.
9
Bozkurt M, Kahilogullari G, Ozdemir M, et al. Comparative analysis of vertebroplasty and kyphoplasty for osteoporotic vertebral compression fractures [J]. Asian Spine J, 2014, 8(1): 27-34.
10
Al Taha K, Lauper N, Bauer DE, et al. Multidisciplinary and Coordinated Management of Osteoporotic Vertebral Compression Fractures: Current State of the Art [J]. J Clin Med, 2024, 13(4): 930.
11
Cao DH, Gu WB, Zhao HY, et al. Advantages of unilateral percutaneous kyphoplasty for osteoporotic vertebral compression fractures-a systematic review and meta-analysis [J]. Arch Osteoporos, 2024, 19(1): 38.
12
Feng ST, Yang Y, Li X, et al. Risk Factors of New Symptomatic Fractures after Vertebroplasty: A Retrospective Cohort Study of 268 Patients with Painful Osteoporotic Vertebral Compression Fracture [J]. World Neurosurg, 2024, 9: S1878-8750(24)00760-5.
13
Wang Z, Li L, Zhang T, et al. Evaluation of predictive performance for new vertebral compression fracture between Hounsfield units value and vertebral bone quality score following percutaneous vertebroplasty or kyphoplasty [J]. Acad Radiol, 2025, 32(3): 1562-1573.
14
Haibier A, Jie Y, Yusufu A, et al. Effect of different cement distribution on the clinical efficacy of vertebral compression fractures in unilateral percutaneous vertebroplasty [J]. Eur Spine J, 2025, 34(5): 1673-1684.
15
Monchka BA, Kimelman D, Lix LM, et al. Feasibility of a generalized convolutional neural network for automated identification of vertebral compression fractures: The Manitoba Bone Mineral Density Registry [J]. Bone, 2021, 150: 116017.
16
Yoda T, Maki S, Furuya T, et al. Automated Differentiation Between Osteoporotic Vertebral Fracture and Malignant Vertebral Fracture on MRI Using a Deep Convolutional Neural Network [J]. Spine, 2022, 47(8): E347-E352.
17
Petraikin AV, Belaya ZE, Kiseleva AN, et al. [Artificial intelligence for diagnosis of vertebral compression fractures using a morphometric analysis model, based on convolutional neural networks] [J]. Probl Endokrinol (Mosk), 2020, 66(5): 48-60.
18
Meng X, Zhou C, Liao Y, et al. Biomechanical Effects of Different Spacing Distributions Between the Cemented Superior Boundary and Surgical Vertebral Superior Endplates After Percutaneous Vertebroplasty for Osteoporotic Vertebral Compression Fractures: A Three-Dimensional Finite Element Analysis [J]. Orthop Surg, 2025, 17(2): 373-392.
19
徐鉴,马园.术后X线片骨折复位满意经皮椎体强化治疗胸腰椎骨质疏松性压缩性骨折的疗效观察[J].宁夏医学杂志, 2019, 41(03): 265-267.
20
Kim C, Kang M, Yuh WT, et al. Comparative efficacy of anteroposterior and lateral X-ray based deep learning in the detection of osteoporotic vertebral compression fracture [J]. Sci Rep, 2024, 14(1): 28388.
[1] 胡庆中, 赵嘉文, 李瑞, 刘文权, 黄白妙, 牟志伟. 肌肉能量技术在反式全肩关节置换术后早期的作用[J/OL]. 中华关节外科杂志(电子版), 2026, 20(03): 261-265.
[2] 黄颖琳, 潘婉诗, 彭琪琪, 林皓健, 刘昌盛, 李桂元, 何颖, 邓宇, 陆丽明, 童娟. 膝骨关节炎常用评分与影像学指标的相关性研究[J/OL]. 中华关节外科杂志(电子版), 2026, 20(03): 307-313.
[3] 杨碧涵, 赵丹, 夏京花, 贾云洋, 鲁雪梅. 髋部骨折患者术后康复效果的评估工具[J/OL]. 中华损伤与修复杂志(电子版), 2026, 21(03): 191-198.
[4] 高嘉敏, 韩俊毅. 老年胃肠癌患者多模式预康复:异质性挑战与分类管理策略[J/OL]. 中华普通外科学文献(电子版), 2026, 20(03): 192-199.
[5] 葛杨洋, 姚尚龙. 多模式镇痛在加速康复外科中的应用现状[J/OL]. 中华普通外科学文献(电子版), 2026, 20(03): 212-216.
[6] 李佳颖, 刘溪, 陈宇航, 桂程鹏, 习明, 陈炜, 罗俊航. 卷积神经网络在预测肾透明细胞癌全视野数字切片肿瘤坏死中的应用研究[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2026, 20(04): 389-398.
[7] 张璐, 王颖, 钱航, 刘雪萍, 代芬. 心肺交互呼吸康复训练用于主动脉夹层合并慢性阻塞性肺疾病患者术后康复随机对照研究[J/OL]. 中华肺部疾病杂志(电子版), 2026, 19(03): 386-391.
[8] 张双双, 杨秀华, 黄山, 贾蓉, 吴迪, 丁静, 余皓月. 痰液诱导联合主动呼吸循环技术对改善慢性阻塞性肺疾病患者肺功能与康复依从性的96例临床研究[J/OL]. 中华肺部疾病杂志(电子版), 2026, 19(03): 470-475.
[9] 林婉琪, 任芳菲, 钟伟航, 杨卿, 马玲, 陈海丽, 向莎, 陈孟娇, 朱丽萍, 何心念, 张献玲. 基于多维度评估的前瞻性干预策略在肝癌肝移植围手术期的应用及效果[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(04): 612-618.
[10] 杨健良, 张爱玲, 王晓明, 代天星, 袁峰, 朱灿华, 蓝岚, 杨汉宇, 汪国营. 早期拔除气管插管快通道麻醉在肝移植中应用[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(03): 398-404.
[11] 章妙玉, 李远美, 黄嫦玉. 基于脑-肠轴非药物康复干预在便秘型肠易激综合征的应用进展[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(03): 284-288.
[12] 孔琴, 岳慧玲, 侯卫华, 郑智, 何阿祥, 赵松, 陆群峰. 基于互联网技术赋能的医护康协作模式在前交叉韧带重建术后患者康复中的应用与评价[J/OL]. 中华临床医师杂志(电子版), 2026, 20(04): 265-274.
[13] 郭格, 谢亚娜, 唐冲, 关振鹏, 李晓. 本体感觉治疗在膝骨关节炎康复中的研究进展[J/OL]. 中华临床医师杂志(电子版), 2026, 20(03): 228-233.
[14] 刘诗馨, 宋晓微, 武剑. 临床视角下人工智能赋能脑卒中康复的范式重塑[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(03): 231-239.
[15] 宋文迪, 孙蔚, 宋海庆. 脑卒中后平衡障碍的发生机制及治疗的研究进展[J/OL]. 中华脑血管病杂志(电子版), 2026, 20(02): 111-118.
阅读次数
全文


摘要


AI


AI小编
你好!我是《中华医学电子期刊资源库》AI小编,有什么可以帮您的吗?