and for the BONSAI (Brain and Optic Nerve Study with Artificial Intelligence) Study Group Objective: To compare the diagnostic performance of an artificial intelligence deep learning system with that of expert neuro-ophthalmologists in classifying optic disc appearance. Methods: The deep learning system was previously trained and validated on 14,341 ocular fundus photographs from 19 international centers. The performance of the system was evaluated on 800 new fundus photographs (400 normal optic discs, 201 papilledema [disc edema from elevated intracranial pressure], 199 other optic disc abnormalities) and compared with that of 2 expert neuro-ophthalmologists who independently reviewed the same randomly presented images without clinical information. Area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity were calculated. Results: The system correctly classified 678 of 800 (84.7%) photographs, compared with 675 of 800 (84.4%) for Expert 1 and 641 of 800 (80.1%) for Expert 2. The system yielded areas under the receiver operating characteristic curve of 0.97 (95% confidence interval [CI] = 0.96-0.98), 0.96 (95% CI = 0.94-0.97), and 0.89 (95% CI = 0.87-0.92) for the detection of normal discs, papilledema, and other disc abnormalities, respectively. The accuracy, sensitivity, and specificity of the system's classification of optic discs were similar to or better than the 2 experts. Intergrader agreement at the eye level was 0.71 (95% CI = 0.67-0.76) between Expert 1 and Expert 2, 0.72 (95% CI = 0.68-0.76) between the system and Expert 1, and 0.65 (95% CI = 0.61-0.70) between the system and Expert 2. Interpretation: The performance of this deep learning system at classifying optic disc abnormalities was at least as good as 2 expert neuro-ophthalmologists. Future prospective studies are needed to validate this system as a diagnostic aid in relevant clinical settings.

Atypical sensory behaviours represent a core symptom of autism spectrum disorder (ASD). Investigating early visual processing is crucial to deepen our understanding of higher-level processes. Visual evoked potentials (VEPs) to pattern-reversal checkerboards were recorded in ASD children and age-matched controls. Peak analysis of the P100 component and two types of single-trial analyses were carried out. P100 amplitude was reduced in the ASD group, consistent with previous reports. The analysis of the proportion of trials with a positive activity in the latency range of the P100, measuring inter-trial (in)consistency, allowed identifying two subgroups of ASD participants: the first group, as control children, showed a high inter-trial consistency, whereas the other group showed an inter-trial inconsistency. Analysis of median absolute deviation of single-trial P100 (st-P100) latencies revealed an increased latency variability in the ASD group. Both single-trial analyses revealed increased variability in a subset of children with ASD. To control for this variability, VEPs were reconstructed by including only positive trials or trials with homogeneous st-P100 latencies. These control analyses abolished group differences, confirming that the reduced P100 amplitude results from increased inter-trial variability in ASD. This increased variability in ASD supports the neural noise theory. The existence of subgroups in ASD suggests that the neural response variability is not a genuine characteristic of the entire autistic spectrum, but rather characterized subgroups of children. Exploring the relationship between sensory responsiveness and inter-trial variability could provide more precise bioclinical profiles in children with ASD, and complete the functional diagnostic crucial for the development of individualized therapeutical projects.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.