Normal radiographic parameters in paediatric pelvic radiographs from a Chinese population

Kuong, Evelyn E; Gardner, William T.; Koljonen, Paul Aarne; Mahapatra, Sudhir Kumar; Chow, Wang

doi:10.1097/bpb.0000000000000426

Cited by 8 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fortunately, the acetabular index measured by deep learning system in the normal group and the abnormal group, using the Bland–Altman principle, both showed good credibility. In addition, the acetabular index declined more rapidly in the first 3 months of life and then changed more rapidly between the ages of 1 and 5 years ( 20 ). Therefore, we divided the cases into four different age groups, and the results showed that the consistency of measured acetabular index in infants under six months of age was worse than that in the other three groups.…”

Section: Discussionmentioning

confidence: 99%

Application of deep-learning–based artificial intelligence in acetabular index measurement

Ma²,

Sun³

et al. 2023

Front. Pediatr.

View full text Add to dashboard Cite

ObjectiveTo construct an artificial intelligence system to measure acetabular index and evaluate its accuracy in clinical application.MethodsA total of 10,219 standard anteroposterior pelvic radiographs were collected retrospectively from April 2014 to December 2018 in our hospital. Of these, 9,219 radiographs were randomly selected to train and verify the system. The remaining 1,000 radiographs were used to compare the system's and the clinicians' measurement results. All plain pelvic films were labeled by an expert committee through PACS system based on a uniform standard to measure acetabular index. Subsequently, eight other clinicians independently measured the acetabular index from 200 randomly selected radiographs from the test radiographs. Bland–Altman test was used for consistency analysis between the system and clinician measurements.ResultsThe test set included 1,000 cases (2,000 hips). Compared with the expert committee measurement, the 95% limits of agreement (95% LOA) of the system was −4.02° to 3.45° (bias = −0.27°, P < 0.05). The acetabular index measured by the system within all age groups, including normal and abnormal groups, also showed good credibility according to the Bland–Altman principle. Comparison of the measurement evaluations by the system and eight clinicians vs. that of, the expert committee, the 95% LOA of the clinician with the smallest measurement error was −2.76° to 2.56° (bias = −0.10°, P = 0.126). The 95% LOA of the system was −0.93° to 2.86° (bias = −0.03°, P = 0.647). The 95% LOA of the clinician with the largest measurement error was −3.41° to 4.25° (bias = 0.42°, P < 0.05). The measurement error of the system was only greater than that of a senior clinician.ConclusionThe newly constructed artificial intelligence system could quickly and accurately measure the acetabular index of standard anteroposterior pelvic radiographs. There is good data consistency between the system in measuring standard anteroposterior pelvic radiographs. The accuracy of the system is closer to that of senior clinicians.

show abstract

Section: Discussionmentioning

confidence: 99%

Application of deep-learning–based artificial intelligence in acetabular index measurement

Ma²,

Sun³

et al. 2023

Front. Pediatr.

View full text Add to dashboard Cite

show abstract

“…The ultrasound and radiographic recovery criteria were as follow: (a) type-I hip according to Graf classification [7] and (b) well-reduced hip on plain radiographs with acetabular index (AI) no more than one SD above the mean value for normal age matched children, as reported by Shi et al . [15,16]. Patients were then divided into two groups whether they met recovery criteria (ultrasound or radiographic) or not: (a) group A: patients meeting recovery criteria and (b) group B: patients not meeting recovery criteria.…”

Section: Methodsmentioning

confidence: 99%

“…PAD was diagnosed if: (a) AI in patients younger than 4 years was greater than 1 SD than the mean value for normal age matched children [15,16,20], or there was evidence of subluxation; (b) patients older than 4 years had grade III or IV hips according to Severin classification [21]; and (c) no evidence of break in Shenton’s line.…”

Section: Methodsmentioning

confidence: 99%

“…If successful reduction was achieved though acetabular dysplasia did not correct following PH treatment or the patient was older than 8-9 months, PH was discontinued and switched to an abduction brace until the hip met the radiographic recovery criteria. The ultrasound and radiographic recovery criteria were as follow: (a) type-I hip according to Graf classification [7] and (b) well-reduced hip on plain radiographs with acetabular index (AI) no more than one SD above the mean value for normal age matched children, as reported by Shi et al [15,16]. Patients were then divided into two groups whether they met recovery criteria (ultrasound or radiographic) or not: (a) group A: patients meeting recovery criteria and (b) group B: patients not meeting recovery criteria.…”

Section: Pavlik Harness Treatmentmentioning

confidence: 99%

See 1 more Smart Citation

Incidence and risk factors for persistent acetabular dysplasia in patients with developmental dislocation of the hip treated by Pavlik Harness

Canavese

Liu

et al. 2022

Journal of Pediatric Orthopaedics B

View full text Add to dashboard Cite

Some patients with developmental dysplasia of the hip (DDH) before 6 months of age successfully treated by Pavlik Harness (PH) still had persistent acetabular dysplasia (PAD). This study aimed to investigate the incidence and risk factors for PAD in patients with DDH treated by PH. We retrospectively reviewed the data of 89 patients (109 hips; mean age, 3.2 ± 1.5 months) with DDH treated by PH. Prior to treatment, all patients underwent ultrasound examination and classified according to Graf’s method. PH was terminated once the hip achieved imaging recovery criteria. At final follow-up, the acetabular index (AI), center-edge angle (CEA) of Wiberg were measured on radiograph. Overall, 67/109 hips (61.5%) had successful PH treatment. Among these 67 hips, 58 hips (86.6%) achieved satisfactory outcome, nine (13.4%) had PAD. Age of the patients with PAD (4.3 ± 1 months) was significantly higher than those without PAD (2.8 ± 1.5 months) (P = 0.001). Hips with PAD had higher mean Graf grade than those with satisfactory outcome (P = 0.014). Logistic regression-confirmed age and Graf classification were risk factors for PAD. Overall, 55/67 (82.1%) met imaging recovery criteria to stop PH treatment, whereas 12/67 hips (17.9%) did not. The rate of PAD at final follow-up in patients achieving recovery criteria (4/55; 7.3%) was significantly lower than those not achieving it (41.7%) (P = 0.007). In conclusion, age and Graf classification are risk factors for PAD. If hips are not normal at the end of PH treatment, the risk of PAD increases further.

show abstract

“…Radiographic dysplasia was defined as AI values > 2 SD from age, sex, and laterality-specific normal values previously defined by Tönnis 18 and Kuong et al 19 If any side of the hip had dysplasia, it was judged as a patient with dysplasia.…”

Section: Radiographic Measurementsmentioning

confidence: 99%

Radiographic Follow-up After Normal Ultrasound Screening of the Hip in Breech Infants

Kim

Yang²,

Choi³

et al. 2022

Journal of Pediatric Orthopaedics

View full text Add to dashboard Cite

Background: Breech presentation is one of the most important risk factors for developmental dysplasia of the hip, and all breech infants should be screened. The necessity of further follow-up of developmental dysplasia of the hip after normal clinical and sonographic screening is a controversial subject. The purpose of this study to identify the incidence of delayed dysplasia in breech infants after normal ultrasound screening and the necessity of further clinical and radiologic follow-up in these patients. Methods: We included the 292 breech babies (128 boys and 164 girls) who showed normal hip ultrasound screening results. To determine the incidence of delayed radiographic dysplasia, anteroposterior hip radiographs were taken between 12 and 24 months of age to measure the acetabular index (AI). Results: The mean AI values were 22.8 ± 3.4 in boys and 24.9 ± 3.1 in girls. Applying the Tönnis criteria, 29 patients (9.9%) were considered to have delayed radiographic dysplasia (16 boys and 13 girls). No significant difference was found in any demographic variables between babies with and without delayed radiographic dysplasia. None of these 29 infants underwent any treatment for radiographic dysplasia. Applying Kuong's criteria to 292 infants, only 2 patients (0.7%) demonstrated radiographic dysplasia on the hip anteroposterior radiographs taken at 14 months. Conclusions: The incidence of radiographic dysplasia significantly varied depending on which criteria were applied. In order to find out more accurate incidence rates of delayed radiographic dysplasia, large-scale studies of the normative AI data for Korean infants are required.

show abstract

Normal radiographic parameters in paediatric pelvic radiographs from a Chinese population

Abstract: Diagnostic Study Level III - Study of nonconsecutive patients (without consistently applying the reference 'gold' standard).

Cited by 8 publications

References 8 publications

Application of deep-learning–based artificial intelligence in acetabular index measurement

Application of deep-learning–based artificial intelligence in acetabular index measurement

Incidence and risk factors for persistent acetabular dysplasia in patients with developmental dislocation of the hip treated by Pavlik Harness

Radiographic Follow-up After Normal Ultrasound Screening of the Hip in Breech Infants

Contact Info

Product

Resources

About