Deep Learning Model for Static Ocular Torsion Detection Using Synthetically Generated Fundus Images

Abstract

Purpose The objective of the study is to develop deep learning models using synthetic fundus images to assess the direction (intorsion versus extorsion) and amount (physiologic versus pathologic) of static ocular torsion. Static ocular torsion assessment is an important clinical tool for classifying vertical ocular misalignment; however, current methods are time-intensive with steep learning curves for frontline providers. Methods We used a dataset ( n = 276) of right eye fundus images. The disc-foveal angle was calculated using ImageJ to generate synthetic images via image rotation. Using synthetic datasets ( n = 12,740 images per model) and transfer learning (the reuse of a pretrained deep learning model on a new task), we developed a binary classifier (intorsion versus extorsion) and a multiclass classifier (physiologic versus pathologic intorsion and extorsion). Model performance was evaluated on unseen synthetic and nonsynthetic data. Results On the synthetic dataset, the binary classifier had an accuracy and area under the receiver operating characteristic curve (AUROC) of 0.92 and 0.98, respectively, whereas the multiclass classifier had an accuracy and AUROC of 0.77 and 0.94, respectively. The binary classifier generalized well on the nonsynthetic data (accuracy = 0.94; AUROC = 1.00). Conclusions The direction of static ocular torsion can be detected from synthetic fundus images using deep learning methods, which is key to differentiate between vestibular misalignment (skew deviation) and ocular muscle misalignment (superior oblique palsies). Translational Relevance Given the robust performance of our models on real fundus images, similar strategies can be adopted for deep learning research in rare neuro-ophthalmologic diseases with limited datasets.

Ocular Torsion in Children with Horizontal Strabismus or Orthophoria

“…The development of deep learning models can aid in diagnosis and measuring the direction and amount of intorsion and extorsion [ 23 ]. Sensory torsional recalibration was recently described as a critical mechanism to overcome objective torsion [ 1 ].…”

Deep Learning Detection of Subtle Torsional Eye Movements: Preliminary Results

“…During head tilt, dynamic changes in ocular torsion is characterized as torsional nystagmus with slow and fast changes in the torsional eye position (Figure 1). With sustained head tilt, there is a static change in the torsional eye position that represents how the vestibular system can detect and generate an ocular response to the pull of gravity [2] [3]. For example, during a left head tilt, the dynamic torsional response consists of a series of slow-phase changes of ocular torsion to the right side, each followed by a fast-phase correction to the left side.…”