2023 , Vol. 20 >Issue 09: 945 - 950
DOI: https://doi.org/10.3877/cma.j.issn.1672-6488.2023.09.009
早孕期胎儿头臀长正中矢状切面超声图像的人工智能质控研究
Copy editor: 吴春凤
收稿日期: 2023-05-10
网络出版日期: 2023-12-11
基金资助
国家重点研发计划(2022YFF0606301)
深圳市科技计划项目(JCYJ20220530155208018)
版权
Artificial intelligence-based quality control of mid-sagittal plane ultrasound images for first trimester fetal crown-rump length
Received date: 2023-05-10
Online published: 2023-12-11
Copyright
探讨人工智能对早孕期胎儿头臀长正中矢状切面超声图像标准程度判断的临床应用价值。
选取深圳市妇幼保健院2022年1月至12月11~13+6周早孕期胎儿头臀长正中矢状切面超声图像1251张为研究对象,以产前超声专家委员会对图像标准程度判断的结果作为金标准,对比智能质控模型、高级超声医师、中级超声医师、初级超声医师对图像标准程度判断的符合率,应用McNemar-Bowker检验和Weighted Kappa分析组间结果的差异性与一致性;以每100张图像为1组,记录每组图像的质控耗时,应用两相关样本Wilcoxon符号秩检验比较四者质控耗时的差异。
智能质控模型对胎儿头臀长正中矢状切面超声图像标准程度判断符合率为90.5%,与金标准结果一致性强(Kappa值=0.83,P<0.001),略低于高级超声医师(91.1%),差异具有统计学意义(χ2=16.40,P<0.001),优于中级超声医师(78.7%)和初级超声医师(68.9%),差异具有统计学意义(χ2=100.25、16.88,P均<0.001)。智能质控模型每组超声图像质控耗时明显少于超声医师[3.57(3.55,3.60)s vs 351(309,384)s vs 363(351,370)s vs 433(407,475)s],差异具有统计学意义(Z=-3.180、-3.181、-3.180、P均<0.001)。
智能质控模型对早孕期胎儿头臀长正中矢状切面超声图像标准程度判断准确且高效。
张梅芳 , 谭莹 , 朱巧珍 , 温昕 , 袁鹰 , 秦越 , 郭洪波 , 侯伶秀 , 黄文兰 , 彭桂艳 , 李胜利 . 早孕期胎儿头臀长正中矢状切面超声图像的人工智能质控研究[J]. 中华医学超声杂志(电子版), 2023 , 20(09) : 945 -950 . DOI: 10.3877/cma.j.issn.1672-6488.2023.09.009
To probe into the clinical application value of artificial intelligence in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester fetal crown-rump length (CRL).
A total of 1251 midsagittal plane ultrasound images of fetuese at 11-13+6 weeks of gestation were selected from the database of Shenzhen Maternity & Child Healthcare Hospital from January to December 2022. Using the unified judging results of the image quality standard by the Expert Committee of Prenatal Ultrasound as the golden standard, the performance of an artificial intelligent based quality control model, senior, middle, and junior sonographers in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester CRL was assessed by calculating their coincidence rates with the golden standard. The coincidence rates were compared using the Mcnemar-Bowker tests, and weighted Kappa values were applied to analyse the difference and consistency among these results. Time of quality control for each set of images was recorded as one group per 100 images. The Wilcoxon's two samples signed-rank test was applied to compare the difference in the time spent among the four groups.
The coincidence rate of the intelligent quality control model with the golden standard was 90.5%, suggesting a strong consistency to the golden standard (Kappa=0.83, P<0.001), which was slightly lower than that of senior sonographers (90.5% vs 91.1%, χ2=16.40, P<0.001), but superior to that of middle and junior sonographers' (90.5% vs 78.7% vs 68.9%, χ2=100.25, 16.88, P<0.001 for all). The time spent by the intelligent model quality control was significantly less than that by ultrasound physicians [3.57 (3.55, 3.60) s vs 351 (309, 384) s vs 363 (351, 370) s vs 433 (407, 475) s; Z=-3.180, -3.181, and -3.180, respectively, P<0.001 for all].
The intelligent quality control model is accurate and efficient in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester CRL.
表1 孕11~13+6周胎儿头臀长正中矢状切面评分标准细则表 |
结构 | 显示情况 | 得分 |
---|---|---|
头部 | 显示 | 3 |
不显示 | 0 | |
臀部 | 显示 | 3 |
不显示 | 0 | |
上颌骨 | 显示 | 4 |
不显示 | 0 | |
间脑 | 显示 | 3 |
不显示 | 0 | |
下颌骨 | 显示 | 3 |
不显示 | 0 | |
鼻尖和鼻前皮肤 | 显示 | 1 |
不显示 | 0 | |
鼻骨 | 显示 | 1 |
不显示 | 0 | |
菱脑 | 显示 | 1 |
不显示 | 0 | |
侧脑室 | 显示 | -0.5 |
不显示 | 0 | |
生殖器 | 显示 | 1 |
不显示 | 0 | |
合计 | 20 |
注:19~20分为标准,17.5~18.5分为基本标准,≤17分为非标准 |
表2 智能质控模型与不同级别超声医师对胎儿头臀长正中矢状切面超声图像的质控结果对比(张) |
质控方式 | 金标准 | 合计 | ||
---|---|---|---|---|
标准 | 基本标准 | 非标准 | ||
智能质控模型 | ||||
标准 | 802 | 33 | 10 | 845 |
基本标准 | 63 | 222 | 6 | 291 |
非标准 | 0 | 7 | 108 | 115 |
高级超声医师 | ||||
标准 | 819 | 54 | 1 | 874 |
基本标准 | 46 | 208 | 10 | 264 |
非标准 | 0 | 0 | 113 | 113 |
中级超声医师 | ||||
标准 | 738 | 83 | 0 | 821 |
基本标准 | 84 | 124 | 2 | 210 |
非标准 | 43 | 55 | 122 | 220 |
初级超声医师 | ||||
标准 | 651 | 139 | 2 | 792 |
基本标准 | 204 | 115 | 26 | 345 |
非标准 | 10 | 8 | 96 | 114 |
合计 | 865 | 262 | 124 | 1251 |
注:高级超声医师、中级超声医师和初级超声医师的人工质控结果与智能质控结果比较,差异均具有统计学意义(χ2=16.40,P<0.001;χ2=100.25,P<0.001;χ2=16.88,P<0.001) |
表3 胎儿头臀长正中矢状切面图像智能质控漏检和误检结构情况统计表(个) |
超声结构 | 漏检 | 误检 |
---|---|---|
鼻骨 | 32 | 6 |
鼻尖和鼻前皮肤 | 26 | 8 |
生殖器 | 25 | 9 |
侧脑室 | 19 | 0 |
菱脑 | 0 | 5 |
间脑 | 2 | 2 |
上颌骨 | 0 | 1 |
下颌骨 | 0 | 0 |
总计 | 104 | 31 |
1 |
|
2 |
|
3 |
|
4 |
International Society of Ultrasound in Obstetrics and Gynecology,
|
5 |
|
6 |
|
7 |
|
8 |
|
9 |
|
10 |
|
11 |
施俊, 汪琳琳, 王珊珊, 等. 深度学习在医学影像中的应用综述 [J]. 中国图象图形学报, 2020, 25(10): 1953-1981.
|
12 |
|
13 |
谭莹, 文华轩, 彭桂艳, 等. 产科超声图像智能质量控制系统的效能 [J]. 中国医学影像技术, 2022, 38(9): 1361-1366.
|
14 |
中华医学会超声医学分会妇产超声学组, 国家卫生健康委妇幼司全国产前诊断专家组医学影像组. 超声产前筛查指南 [J]. 中华超声影像学杂志, 2022, 31(1): 1-12.
|
15 |
李胜利, 文华轩. 11~13~(+6)周早孕期胎儿超声规范化扫查技术及判断标准 [J/CD]. 中华医学超声杂志(电子版), 2014, 11(1): 6-8.
|
16 |
李胜利, 文华轩. 中孕期胎儿系统超声检查切面及临床意义 [J/CD]. 中华医学超声杂志(电子版), 2010, 7(3): 366-381.
|
17 |
|
18 |
|
/
〈 |
|
〉 |