RESUMEN
BACKGROUND: Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population. METHODS: This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study. RESULTS: The DLS could accurately distinguish between all ODD and papilledema (all severities included): area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%). CONCLUSIONS: A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.
RESUMEN
BACKGROUND: The tortuosity of the optic nerve can be quantified radiologically by measuring the angle of optic nerve deformation (the "optic nerve angle" [ONA]). In patients with idiopathic intracranial hypertension (IIH), lowering the intracranial pressure (ICP) to a normal range by lumbar puncture leads to straightening of the optic nerve and an increase in the measured sagittal ONA on MRI. It is uncertain whether there is any correlation between ONA and cerebrospinal fluid (CSF) opening pressure or visual function. METHODS: Retrospective study of patients with and without IIH who had MRI of the brain followed by lumbar puncture with CSF opening pressure within 24 hours of MRI. Before LP and within 24 hours of MRI of the brain, all patients with IIH had neuro-ophthalmologic assessment including visual acuity, Humphrey Visual Field (HVF), and fundus photography. Sagittal ONA was measured on multiplanar T2-SPACE images on a DICOM viewer. Papilledema on the fundus photographs was graded using the Frisén scale. RESULTS: Fifty-four patients with IIH and 30 unmatched controls were included. The IIH group was 6.3 years younger (95% CI 2.4-10.3, P = 0.002), had 8.7 kg/m2 higher body mass index (4.9-12.5, P < 0.001), and 26.3% more women (P = 0.011) compared with controls. In both eyes, the ONA was significantly smaller in patients with IIH by 12° compared with controls (7°-17°, P < 0.00001). In the IIH group, no correlation between ONA and the CSF opening pressure was present in either eye (right eye r = 0.19, P = 0.15; left eye r = 0.18, P = 0.19) The ONA did not correlate with logarithm of the minimum angle of resolution visual acuity (right eye r = 0.26, P = 0.063; left eye r = 0.15, P = 0.27), HVF mean deviation (right eye r = 0.0059, P = 0.97; left eye r = -0.069, P = 0.63), or Frisén grade (Spearman's rho right eye 0.058, P = 0.67; left eye 0.14, P = 0.30). CONCLUSIONS: The ONA is significantly smaller in patients with IIH compared to controls, but does not correlate with CSF opening pressure, severity of papilledema, or visual function. The ONA may be useful in identifying patients with raised ICP, but not necessarily those with a poor visual prognosis.