Machine learning application in visual field loss for Dominant Optic Atrophy

Abstract

Purpose : Identification of the characteristic visual field loss patterns of dominant optic atrophy (DOA) employing the Archetypal Analysis (AA) machine learning algorithm.

Methods : For 66 patients affected by molecularly confirmed DOA with OPA1 heterozygous mutation, binocular visual field (VF) tests performed by SITA standard 30-2 or 24-2 Humphrey VF analyser (Carl Zeiss Meditec, Dublin, CA, USA) were collected. For 38 patients that had multiple VF tests, 2 or 3 were collected to enlarge the dataset. Resorting to Python 3.8 (Anaconda) an AA model is being developed to determine the DOA underlying patterns of visual loss.

Results : Considering 139 VF test, preliminary results using AA detected archetypes (AT) for the characterisation of visual loss in DOA. Total loss and central AT revealed to be the most significant ones, matching the common characteristic central or ceco-central scotoma for DOA patients, followed by the quadrantanopia and hemianopia ATs. The primary implementation of this algorithm focused on the DOA disease evidenced the potential to help the distinction of the typical visual loss patterns, which is in line with other works focused on diseases as glaucoma and idiopathic intracranial hypertension.

Conclusions : Archetypal analysis shows to be a potential powerful tool to support clinicians in the identification and distinction of visual loss patterns in DOA. Moreover, the analysis can be useful in disease classification and prognosis for future clinical and therapeutical trials.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.