计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250400049-7.doi: 10.11896/jsjkx.250400049
吴晓晓1, 吴兴隆1,2
WU Xiaoxiao1, WU Xinglong1,2
摘要： 胎儿小脑发育不良(CH)是一种严重的中枢神经系统发育异常,其早期诊断对胎儿健康至关重要。对此,提出一种基于脑解剖结构的网络(Brain Anatomy-Based Network,BAB-Net),用于胎儿CH的产前诊断,旨在提高超声图像的诊断准确性。BAB-Net以超声图像和脑部解剖结构特征作为输入,利用解剖结构约束网络进行特征提取与融合。收集了2019年9月至2023年9月期间某三甲医院的超声图像数据,共纳入301例CH胎儿病例和547例正常胎儿病例。病例中小脑、延髓池及颅骨边界均由经验丰富的超声医师标定。在训练集上完成模型训练后,BAB-Net在两个独立测试集上的分类准确率分别达到0.977 8和0.922 2,显著优于多个主流分类网络。在孕周小于30周的病例中,BAB-Net表现出较高的诊断准确性。进一步分析发现,胎儿小脑及延髓池的解剖结构对网络性能的影响大于颅骨结构。通过融合解剖结构约束,BAB-Net有效提升了胎儿CH的诊断精度,为CH的产前筛查提供了新的技术手段,并为临床医生在孕期管理和精准干预方面提供了重要参考。
中图分类号:
| [1] WANG Y,JIA J,GAO N N,et al.Developmental characteristics of width of cerebellomedullary cistern in fetus and its differences in gender[J].Journal of Clinical Medicine in Practice,2023,27(2):117-121. [2] XU Z H,HUANG B Q,CHEN S H,et al.The clinical prognosis and significance of fetus with cisterna magna dilatation [J].Chinese Journal of Ultrasound Medicine,2012,28(12):1120-1122. [3] TONG T,XIONG Y.New advances in ultrasound diagnosis of fetal posterior cranial fossa malformations [J].Chinese Journal of Prenatal Diagnosis(Electronic Version),2015,7(2):48-53. [4] KONKEL L.The brain before birth:using fMRI to explore the secrets of fetal neurodevelopment [J].Environmental Health Perspectives,2019,127(1/2):014002. [5] VOLPE J J.Brain injury in premature infants:a complex amalgam of destructive and developmental disturbances [J].The Lancet Neurology,2009,8(1):110-124. [6] XIAO S S,ZHANG J M,ZHU Y,et al.Application and progress of artificial intelligence in fetal ultrasound[J].Journal of Clinical Medicine,2023,12(9):3298. [7] WÜEST A,SURBEK D,WIEST R,et al.Enlarged posteriorfossa on prenatal imaging:differential diagnosis,associated anomalies and postnatal outcome[J].Acta Obstetricia et Gynecologica Scandinavica,2017,96(7):837-843. [8] JIANG M X,ZHAO T Z,WU Z J.Application of convolutional neural network based on target shape in ships and warships classification [J].Journal of Wuhan Institute of Technology,2020,42(2):213-217. [9] ZHENG Y,WANG L,YANG Z W.Monocular depth estimation based on adaptive fusion of multi-scale depth maps[J].Journal of Wuhan Institute of Technology,2024,46(1):85-90. [10] GOFER S,HAIK O,BARDIN R,et al.Machine learning algorithms for classification of first-trimester fetal brain ultrasound images [J].Journal of Ultrasound in Medicine,2022,41:1773-1779. [11] YE H,FENG K P,XIE H N.Fetal brain ultrasonic image segmentation algorithm based on fully convolution network[J].Modern Computer,2019(17):51-54. [12] BURGOS-ARTIZZU X P,CORONADO-GUTIÉRREZ D,VAL-ENZUELA-ALCARAZ B,et al.Evaluation of deep convolu-tional neural networks for automatic classification of common maternal fetal ultrasound planes [J].Scientific Reports,2020,10(1):10200. [13] LIN Q,ZHOU Y L,SHI S Y,et al.How much can AI see in early pregnancy:a multi-center study of fetus head characterization in week 10-14 in ultrasound using deep learning [J].Computer Methods and Programs in Biomedicine,2022,226:107170. [14] ZHANG Y,ZHANG X,LUO Y K,et al.The role of artificial intelligence aided teaching for young ultrasound doctors [J].Translational Medicine Journal,2020,9(5):305-307. [15] LI S L,WEN H X,TIAN X X.Fetal cranial ultrasound examination:diagnostic thinking [J].Chinese Journal of Medical Ultrasound(Electronic Edition),2015,12(8):590-598. [16] SALOMON L J,ALFIREVIC Z,BERGHELLA V,et al.Practice guidelines for performance of the routine mid-trimester fetal ultrasound scan [J].Ultrasound in Obstetrics & Gynecology,2011,37:116-126. [17] LI Y Y.Facemask detection using inception V3 model and effect on accuracy of data preprocessing methods[J].Journal of Phy-sics:Conference Series,2021,2010(1):012052. [18] CAI J,DU J C,WANG Q,et al.Facial emotion recognitionbased on VGG16 network[J].Application of Electronic Technique,2022,48(1):67-70,75. [19] LI Y H,YAO T,PAN Y W,et al.Contextual Transformer networks for visual recognition[J].IEEE Transactions on Pattern analysis and Machine Intelligence,2022,45(2):1489-1500. [20] LIU Z,LIN Y T,CAO Y,et al.Swin Transformer:hierarchical vision Transformer using shifted windows [C]//2021 IEEE/CVF International Conference on Computer Vision.New York:IEEE,2021:10012-10022. [21] NAMBIAR V B,RAMAMURTHY B,VEERESHA P.Gender determination from periocular images using deep learning based EfficientNet architecture[J].International Journal of Mathematics and Computer in Engineering,2024,2(1):59-70. [22] LI Y,ZHU X L,PAN M Z,et al.Are fetal gender and gesta-tional age related to the size of cisterna magna? [J].The Journal of Maternal-Fetal & Neonatal Medicine,2022,35(22):4312-4317. [23] LIN Y J,CHIU N C,CHEN H J,et al.Cranial ultrasonographic screening findings among healthy neonates and their association with neurodevelopmental outcomes [J].Pediatrics & Neonato-logy,2021,62(2):158-164. [24] ENACHE I A,IOVOAICA-RĂMESCU C,CIOBANU G,et al.Artificial intelligence in obstetric anomaly scan:heart and brain [J].Life,2024,14(2):166. [25] ZHOU F Y,JIN L P,DONG J.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251. [26] HAN K,WANG Y H,CHEN H T,et al.A survey on vision Transformer[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(1):87-110. [27] PANWAR H,GUPTA P K,SIDDIQUI M K,et al.A deeplearning and grad-CAM based color visualization approach for fast detection of COVID-19 cases using chest X-ray and CT-Scan images[J].Chaos,Solitons & Fractals,2020,140:110190. [28] LI G,XIAO L X,JIAN N,et al.MRI study of growth and development of fetal cerebellum in the second and third trimester of pregnancy and its relationship with simple posterior fossa pool broadening [J].Chinese Journal of Clinical Anatomy,2018,36(5):495-499. |
|
||
首页