Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Infant Development Dysplasia of the Hip (DDH) is a disorder where the hip joint does not form properly. The Graf method and The Femoral Head Coverage (FHC) method are ultrasound-based screening techniques which use anatomical landmarks to classify disease severity and guide treatment. Deep learning has been used for either Graf Classification or for FHC, yet there has been only minimal investigation into combining the two. No work to-date has detected FHC using landmarks alone. This paper develops a model which predicts both Graf Classification and FHC from landmarks only. In this method, Recall (Precision) improved when combining methods compared to FHC and Graf methods alone. Two external datasets were used to evaluate model performance under domain shift. Improvements are needed to generalise the model to new datasets. Since the model encompasses both techniques, it gains a clinical understanding of automated methods for DDH screening and improves clinical use.

Type

Conference paper

Publisher

Springer

Publication Date

11/09/2024

Keywords

landmark detection, ultrasound, screening, DDH