The Application Value of Accurate Diagnosis of CT Image of Skull Base Fractures based on Convolutional Neural Network
-
摘要: 目的：探讨卷积神经网络（CNN）在颅底骨折CT诊断的应用价值。方法：回顾性搜集3 100例颅底骨折患者及2 467例正常患者的颅骨CT图像数据，经纳排标准筛选，最终选用2 488例颅底骨折及1 628例正常患者的颅底CT图像数据。对CT图像进行骨折标注后，随机分配训练集和测试集后。通过CNN构建颅骨区域识别算法模型和颅骨骨折检测算法模型，随后在测试中以颅底骨折区域识别和头颅骨折、颅底骨折对模型进行验证，验证指标为精准率（precision）、召回率（recall）、平均诊断耗时；与人工组（低年资放射科医师）测试进行诊断效能对比。结果：通过CNN运算获得的稳定模型后进行测试对比，结果显示全颅底区域骨折、前、中、后颅底骨折精准度均<0.5，低于人工组（均>0.63）；召回率>0.89，均优于人工组（均<0.8）；平均诊断时间为（3.12±2.67）s，明显少于人工组诊断时间。分别在颅底骨折区域测试中，精准度率：前颅底>中颅底>后颅底，召回率：中颅底>后颅底>前颅底。结论：基于CNN颅底骨折算法模型对于颅脑外伤患者CT诊断颅底骨折在召回率、诊断耗时均优于人工测试结果，在辅助临床诊断、降低漏诊及诊断耗时方面具有一定的价值。Abstract: Objective: To explore the application value of convolutional neural network (CNN) in CT diagnosis of skull base fractures. Methods: The skull CT image data of 3100 patients with skull base fractures and 2 467 normal patients was collected retrospectively. After the standard nanofiltration and actual model calculation, the skull base CT image data of 2 488 patients with skull base fractures and 1 628 normal patients were selected. The CT images were labeled and randomly assigned into training set and test set. The skull area discrimination algorithm model and skull base fractures detection algorithm model were established by CNN, then we performed verification on the models through skull base area discrimination, skull fractures and skull base fractures in the test. The detection indexes included precision, recall and average diagnosis time consumption. We carried out comparisons of diagnostic efficacy with the artificial group (junior radiologist) test. Results: We carried out test comparisons on the steady models obtained by CNN algorithm, the results showed that the accuracy of the whole skull base fractures (including the anterior, middle and posterior skull base fractures) was less than 0.5, which was lower than that of the artificial group (all higher than 0.63); The recall rate > 0.89 was better than that of the artificial group (all < 0.8); The average diagnosis time was (3.12±67)s, significantly less than that of artificial group. In the area test of skull base fractures, the accuracy rate was anterior skull base > middle skull base > posterior skull base while the recall rate was middle skull base > posterior skull base > anterior skull base. Conclusion: The algorithm model of skull base fractures based on CNN is superior to the artificial test results in recall rate and diagnosis time consumption for CT diagnosis of skull base fractures in patients with craniocerebral trauma, which has certain value in assisting clinical diagnosis, reducing missed diagnosis and diagnosis time consumption.
-
Keywords:
- convolutional neural networks /
- TinyNet /
- skull base fractures /
- CT /
- deep learning
-
-
[1] 朱捷, 田瑞霞, 黄振山. 1121例交通事故致颅脑损伤流行病学调查[J]. 安徽医学, 2012, 33(5):601-604. ZHU J, TIAN R X, HUANG Z S. Research of epidemiology of 1121 cases of craniocerebral traffic injuries[J]. Anhui Medical Journal, 2012, 33(5):601-604. (in Chinese).
[2] Neurology T L. Traumatic brain injury:Time to end the silence[J]. Lancet Neurology, 2010, 9(4):331. DOI: 10.1016/S1474-4422(10)70069-7.
[3] CIOMPI F, CHUNG K, van RIEL S J, et al. Towards automatic pulmonary nodule management in lung cancer screening with deep learning[J]. Scientific Reports, 2017, 7:46878. DOI:10. 1038/srep46878.
[4] WANG H K, ZHOU Z W, LI Y C, et al. Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18F-FDG PET/CT images[J]. European Journal of Nuclear Medicine and Molecular Imaging Research, 2017, 7(1):11. DOI: 10.1186/s13550-017-0260-9.
[5] PAUL R, HAWKINS S H, BALAGURUNATHAN Y, et al. Deep feature transfer learning in combination with traditional features predicts survival among patients with lung adenocarcinoma[J]. Tomography, 2016, 2(4):388-395. DOI: 10.18383/j.tom.2016.00211.
[6] 周清清, 王佳硕, 唐雯, 等. 基于卷积神经网络成人肋骨骨折CT自动检测和分类的应用研究[J]. 影像诊断与介入放射学, 2020, 29(1):27-31. ZHOU Q Q, WANG J S, TANG W, et al. Automatic detection of adult rib fractures on CT using convolutional neural network[J]. Diagnostic Imaging & Interventional Radiology, 2020, 29(1):27-31. (in Chinese).
[7] YANG S, YIN B, CAO W, et al. Diagnostic accuracy of deep learning in orthopaedic fractures:A systematic review and meta-analysis[J]. Clinical Radiology, 2020, 75(9):713. e17-713.e28.
[8] LI L H, SONG X F, GUO Y, et al. Deep convolutional neural networks for automatic detection of orbital blowout fractures[J]. The Journal of Craniofacial Surgery, 2020, 31(2):400-403.
[9] 吴孟超, 吴在德. 黄家驷外科学[M]. 7版. 北京:人民卫生出版社, 2019:833-834. [10] MANSON P N, STANWIX M G, YAREMCHUK M J, et al. Frontobasal fractures:Anatomical classification and clinical significance[J]. Plastic Reconstructive Surgery, 2009, 124(6):2096-2106.
[11] ZIU M, SAVAGE J G, JIMENEZ D F. Diagnosis and treatment of cerebrospinal fluid rhinorrhea following accidental traumatic anterior skull base fractures[J]. Neurosurgical Focus, 2012, 32(6):E3. DOI: 10.3171/2012.4.FOCUS1244.
[12] DAUDIA A, BISWAS D, JONES N S. Risk of meningitis with cerebrospinal fluid rhinorrhea[J]. Annals of Otology Rhinology & Laryngology, 2007, 116(12):902-905.
[13] JOHNSON F, SEMAAN M T, MEGERIAN C A. Temporal bone fracture:Evaluation and management in the modern era[J]. Otolaryngologic Clinics of North America, 2008, 41(3):597-618.
[14] DAHIYA R, KELLER J D, LITOFSKY N S, et al. Temporal bone fractures:Otic capsule sparing versus otic capsule violating clinical and radiographic considerations[J]. The Journal of Trauma, 1999, 47(6):1079-1083. DOI:10.1097/00005373-199912000-00014. PMID:10608536.
[15] LITTLE S C, KESSER B W. Radiographic classifification of temporal bone fractures:Clinical predictability using a new system[J]. Archives of Otolaryngology-Head & Neck Surgery, 2006, 132(12):1300-1304. DOI:10.1001/archotol.132.12.1300. PMID:17178939.
[16] OCHALSKI P G, SPIRO R M, Fabio A, et al. Fractures of the clivus:A contemporary series in the computed tomography era[J]. Neurosurgery, 2009, 65(6):1063-1069.
[17] 李文星, 陈森林, 程云, 等. 颅脑外伤急诊CT检查的误、漏诊分析[J]. 中国急救复苏与灾害医学杂志, 2012, 7(9):878-880. -
期刊类型引用(2)
1. 林煜文，黄冬云，龙琳. 基于RefineNet卷积神经网络对CT颅底骨折精准诊断的应用价值. 医学影像学杂志. 2024(07): 5-8+17 . 
百度学术
2. 刘红光，张行坤，刘诚，李伟凯. 基于深度卷积神经网络的异常肺部CT图像初筛平台的设计与临床实践. 电子设计工程. 2023(23): 184-188+195 . 百度学术
其他类型引用(0)
计量
- 文章访问数: 242
- HTML全文浏览量: 10
- PDF下载量: 19
- 被引次数: 2
下载:
