Research
Print page Print page
Switch language
Rigshospitalet - a part of Copenhagen University Hospital
Published

Optic Disc Classification by Deep Learning versus Expert Neuro-Ophthalmologists

Research output: Contribution to journalJournal articleResearchpeer-review

DOI

  1. MRI in Neuromuscular Diseases: An Emerging Diagnostic Tool and Biomarker for Prognosis and Efficacy

    Research output: Contribution to journalReviewpeer-review

  2. A New Glycogen Storage Disease Caused by a Dominant PYGM Mutation

    Research output: Contribution to journalJournal articleResearchpeer-review

  3. Neuroinflammatory markers associate with cognitive decline after major surgery: findings of an explorative study

    Research output: Contribution to journalJournal articleResearchpeer-review

  4. Early Intrathecal T Helper 17.1 Cell Activity in Huntington Disease

    Research output: Contribution to journalJournal articleResearchpeer-review

  1. Ipsilateral Recurrence of Optic Disc Drusen-Associated Anterior Ischemic Optic Neuropathy in a 15-Year-Old Boy

    Research output: Contribution to journalJournal articleResearchpeer-review

  2. Optic disc drusen diagnosed by optical coherence tomography in a 3-year-old child

    Research output: Contribution to journalJournal articleResearchpeer-review

  3. Multiplexed optical coherence tomography imaging of optic disc drusen

    Research output: Contribution to journalJournal articleResearchpeer-review

  4. Optic Disc Drusen Associated Anterior Ischemic Optic Neuropathy: Prevalence of Comorbidities and Vascular Risk Factors

    Research output: Contribution to journalJournal articleResearchpeer-review

  • BONSAI (Brain and Optic Nerve Study with Artificial Intelligence) Study Group
  • Anna Isabelle Karlesand (Member of study group)
View graph of relations

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. ANN NEUROL 2020;88:785-795.

Original languageEnglish
JournalAnnals of Neurology
Volume88
Issue number4
Pages (from-to)785-795
Number of pages11
ISSN0364-5134
DOIs
Publication statusPublished - Oct 2020

    Research areas

  • Adult, Aged, Deep Learning, Diagnostic Techniques, Ophthalmological, Female, Humans, Image Interpretation, Computer-Assisted/methods, Male, Middle Aged, Ophthalmologists, Optic Disk, Sensitivity and Specificity

ID: 61987275