Aims. The diagnosis of developmental dysplasia of the hip (DDH) is challenging owing to extensive variation in paediatric pelvic anatomy. Artificial intelligence (AI) may represent an effective diagnostic tool for DDH. Here, we aimed to develop an anteroposterior pelvic radiograph deep learning system for diagnosing DDH in children and analyze the feasibility of its application. Methods. In total, 10,219 anteroposterior pelvic radiographs were retrospectively collected from April 2014 to December 2018. Clinicians labelled each radiograph using a uniform standard method. Radiographs were grouped according to age and into ‘dislocation’ (dislocation and subluxation) and ‘non-dislocation’ (normal cases and those with dysplasia of the acetabulum) groups based on clinical diagnosis. The deep learning system was trained and optimized using 9,081 radiographs; 1,138 test radiographs were then used to compare the diagnoses made by deep learning system and clinicians. The accuracy of the deep learning system was determined using a receiver operating characteristic curve, and the consistency of acetabular index measurements was evaluated using Bland-Altman plots. Results. In all, 1,138 patients (242 males; 896 females; mean age 1.5 years (SD 1.79; 0 to 10) were included in this study. The area under the receiver operating characteristic curve, sensitivity, and specificity of the deep learning system for diagnosing hip dislocation were 0.975, 276/289 (95.5%), and 1,978/1,987 (99.5%), respectively. Compared with clinical diagnoses, the Bland-Altman 95% limits of agreement for acetabular index, as determined by the deep learning system from the radiographs of non-dislocated and dislocated hips, were -3.27° - 2.94° and -7.36° - 5.36°, respectively (p < 0.001). Conclusion. The deep learning system was highly consistent, more convenient, and more effective for diagnosing DDH compared with clinician-led diagnoses. Deep learning systems should be considered for analysis of anteroposterior pelvic radiographs when diagnosing DDH. The deep learning system will improve the current artificially complicated screening referral process. Cite this article: Bone Joint J 2020;102-B(11):1574–1581

We studied radiographs of 125 children (105 boys, 20 girls) with unilateral Legg-Calvé-Perthes’ disease to examine the epiphyseal development of the femoral head in the contralateral (unaffected) hip. The epiphyseal height (EH) and width (EW) of the unaffected hip were measured on the initial anteroposterior pelvic radiograph. In 109 of the patients (87.2%) the EH was below the mean for normal Japanese children and a significantly small EH (below −2 . sd. s) was observed in 23 patients (18.4%). By contrast, the EW of most patients (95.2%) lay within ± 2 SDs of normal values except for six with a significantly small EW. A strong positive linear correlation (R = 0.87) was observed in the EH:EW ratio in the patients. A smaller EH than expected for EW in our series indicated epiphyseal flattening of the femoral head in Legg-Calvé-Perthes’ disease. Our findings support the hypothesis that a delay in endochondral ossification in the proximal capital femoral epiphysis may be associated with the onset of Perthes’ disease

Aims

Perthes’ disease (PD) often results in femoral head deformity and leg length discrepancy (LLD). Our objective was to analyze femoral morphology in PD patients at skeletal maturity to assess where the LLD originates, and evaluate the effect of contralateral epiphysiodesis for length equalization on proximal and subtrochanteric femoral lengths.

The practice of regular radiological follow-up of infants with a positive family history of developmental dysplasia of the hip is based on the widespread belief that primary acetabular dysplasia is a genetic disorder which can occur in the absence of frank subluxation or dislocation. We reviewed all infants who were involved in our screening programme for developmental dysplasia of the hip, between November 2002 and January 2004, and who had a normal clinical and ultrasound examination of the hip at six to eight weeks of age, but who, because of a family history of developmental dysplasia of the hip, had undergone further radiography after an interval of 6 to 12 months. The radiographs of 89 infants were analysed for signs of late dysplasia of the hip and assessed independently by three observers to allow for variability of measurement. There were 11 infants (11%) lost to follow-up.

All the patients had normal radiographs at the final follow-up and none required any intervention. We therefore question the need for routine radiological follow-up of infants with a positive family history of developmental dysplasia of the hip, but who are normal on clinical examination and assessment by ultrasound screening when six to eight weeks old.

There is much debate about the nature and extent of deformities in the proximal femur in children with cerebral palsy. Most authorities accept that increased femoral anteversion is common, but its incidence, severity and clinical significance are less clear. Coxa valga is more controversial and many authorities state that it is a radiological artefact rather than a true deformity.

We measured femoral anteversion clinically and the neck-shaft angle radiologically in 292 children with cerebral palsy. This represented 78% of a large, population-based cohort of children with cerebral palsy which included all motor types, topographical distributions and functional levels as determined by the gross motor function classification system.

The mean femoral neck anteversion was 36.5° (11° to 67.5°) and the mean neck-shaft angle 147.5° (130° to 178°). These were both increased compared with values in normally developing children. The mean femoral neck anteversion was 30.4° (11° to 50°) at gross motor function classification system level I, 35.5° (8° to 65°) at level II and then plateaued at approximately 40.0° (25° to 67.5°) at levels III, IV and V. The mean neck-shaft angle increased in a step-wise manner from 135.9° (130° to 145°) at gross motor function classification system level I to 163.0° (151° to 178°) at level V. The migration percentage increased in a similar pattern and was closely related to femoral deformity.

Based on these findings we believe that displacement of the hip in patients with cerebral palsy can be explained mainly by the abnormal shape of the proximal femur, as a result of delayed walking, limited walking or inability to walk. This has clinical implications for the management of hip displacement in children with cerebral palsy.