AI recognises hip misalignments
Hip malalignment in newborns can make many operations necessary. SCS has developed an AI that recognises such malpositions.
-
Problem definition
Hip malalignment in newborns can necessitate various operations in the course of their lives. To prevent this, it is sufficient to determine the alpha and beta angles according to Graf using an ultrasound scan of the hip and, if necessary, apply a corset for a few months for treatment. Doctors' current diagnoses are subjective and prone to error.
-
Solution
Using training data consisting of ultrasound images, SCS was able to train various algorithms that allow the angles to be determined objectively. In this way, the doctor can be supported.
-
Added value
Doctors are alerted to an error in image acquisition at an early stage and can rely on an objective and accurate angle determination. This can prevent many hip operations in the future.
Project insights
Around 3 per cent of all newborns are born with a hip deformity. If this deformity is recognised by ultrasound examinations, it can be easily treated by fitting the baby with a corset for several weeks (Fig. 1). If undetected, this so-called hip dysplasia leads to hip surgery before the age of sixty in 30% of those affected.
The criterion for determining whether a deformity develops is the alpha angle according to Graf’s method. An evaluation of the angles determined by doctors shows a strong tendency towards “borderline cases” at 60 degrees that do not require treatment.

As an alternative, machine learning and data analytics – especially thanks to developments in image processing with the help of deep learning – offer a way to objectively and reproducibly determine the angle. To confirm this expectation, SCS has conducted a study together with various experts and paediatricians [1], which can confirm this thesis [2].

To do this, the data basis was first cleaned by removing the overhang of borderline cases. A deep neural network was then trained to predict the lines on ultrasound images that determine the hip angle, analogous to the medical procedure. A comparison with data from paediatricians that was not used for training shows that our algorithm can reliably determine the angle.
[1] Partners: Dr Stefan Essig, Dr Thomas Baumann
[2] Reference: Oelen D, Kaiser P, Baumann T, et al. Accuracy of Trained Physicians is Inferior to Deep Learning-Based Algorithm for Determining Angles in Ultrasound of the Newborn Hip Ultrasound Med. 2020;10.1055/a-1177-0480. doi:10.1055/a-1177-0480
