Artificial Intelligence-Assisted Ultrasound Diagnosis on Infant Developmental Dysplasia of the Hip Under Constrained Computational Resources

被引:5
作者
Huang, Bingxuan [1 ]
Xia, Bei [1 ,5 ]
Qian, Jikuan [2 ]
Zhou, Xinrui [3 ]
Zhou, Xu [3 ]
Liu, Shengfeng [3 ]
Chang, Ao [3 ]
Yan, Zhongnuo [3 ]
Tang, Zijian [1 ]
Xu, Na [1 ]
Tao, Hongwei [1 ]
He, Xuezhi [1 ]
Yu, Wei [1 ]
Zhang, Renfu [4 ]
Huang, Ruobing [3 ]
Ni, Dong [3 ]
Yang, Xin [3 ,6 ]
机构
[1] Shantou Univ, Affiliated Shenzhen Childrens Hosp, Coll Med, Ultrasonog Dept, Shenzhen, Peoples R China
[2] Shenzhen RayShape Med Technol Co Ltd, R&D Dept, Shenzhen, Peoples R China
[3] Shenzhen Univ, Hlth Sci Ctr, Sch Biomed Engn, Natl Reg Key Technol Engn Lab Med Ultrasound, Shenzhen, Peoples R China
[4] EDAN Instruments Inc, Ultrasound Dept, Shenzhen, Peoples R China
[5] Shenzhen Childrens Hosp, 7019 Yitian Ave, Shenzhen 518048, Peoples R China
[6] Shenzhen Univ, 1066 Xueyuan Ave, Shenzhen 518000, Peoples R China
基金
中国国家自然科学基金;
关键词
artificial intelligence; diagnosis; developmental dysplasia of the hip; infants; ultrasound; PUBO-FEMORAL DISTANCE; DISLOCATION; SONOGRAPHY; US;
D O I
10.1002/jum.16133
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
ObjectivesUltrasound (US) is important for diagnosing infant developmental dysplasia of the hip (DDH). However, the accuracy of the diagnosis depends heavily on expertise. We aimed to develop a novel automatic system (DDHnet) for accurate, fast, and robust diagnosis of DDH. MethodsAn automatic system, DDHnet, was proposed to diagnose DDH by analyzing static ultrasound images. A five-fold cross-validation experiment was conducted using a dataset containing 881 patients to verify the performance of DDHnet. In addition, a blind test was conducted on 209 patients (158 normal and 51 abnormal cases). The feasibility and performance of DDHnet were investigated by embedding it into ultrasound machines at low computational cost. ResultsDDHnet obtained reliable measurements and accurate diagnosis predictions. It reported an intra-class correlation coefficient (ICC) on alpha angle of 0.96 (95% CI: 0.93-0.97), beta angle of 0.97 (95% CI: 0.95-0.98), FHC of 0.98 (95% CI: 0.96-0.99) and PFD of 0.94 (95% CI: 0.90-0.96) in abnormal cases. DDHnet achieved a sensitivity of 90.56%, specificity of 100%, accuracy of 98.64%, positive predictive value (PPV) of 100%, and negative predictive value (NPV) of 98.44% for the diagnosis of DDH. For the measurement task on the US device, DDHnet took only 1.1 seconds to operate and complete, whereas the experienced senior expert required an average 41.4 seconds. ConclusionsThe proposed DDHnet demonstrate state-of-the-art performance for all four indicators of DDH diagnosis. Fast and highly accurate DDH diagnosis is achievable through DDHnet, and is accessible under constrained computational resources.
引用
收藏
页码:1235 / 1248
页数:14
相关论文
共 27 条
  • [1] Residual dysplasia after successful Pavlik harness treatment - Early ultrasound predictors
    Alexiev, VA
    Harcke, HT
    Kumar, SJ
    [J]. JOURNAL OF PEDIATRIC ORTHOPAEDICS, 2006, 26 (01) : 16 - 23
  • [2] American Institute of Ultrasound in Medicine, 2013, J Ultrasound Med, V32, P1307, DOI 10.7863/ultra.32.7.1307
  • [3] [Anonymous], ONL WHO GROWTH CHART
  • [4] ARONSSON DD, 1994, PEDIATRICS, V94, P201
  • [5] Comparative Evaluation of Hand-Engineered and Deep-Learned Features for Neonatal Hip Bone Segmentation in Ultrasound
    El-Hariri, Houssam
    Mulpuri, Kishore
    Hodgson, Antony
    Garbi, Rafeef
    [J]. MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT II, 2019, 11765 : 12 - 20
  • [6] Flynn JT, 2017, PEDIATRICS, V140, DOI [10.1542/peds.2017-3035, 10.1542/peds.2017-1904]
  • [8] Graf R., 2006, Hip Sonography: Diagnosis and Management of Infant Hip Dysplasia
  • [9] Hip sonography: background; technique and common mistakes; results; debate and politics; challenges
    Graf, Reinhard
    [J]. HIP INTERNATIONAL, 2017, 27 (03) : 215 - 219
  • [10] Hip ultrasound for developmental dysplasia: the 50% rule
    Harcke, H. Theodore
    Pruszczynski, B.
    [J]. PEDIATRIC RADIOLOGY, 2017, 47 (07) : 817 - 821