Diagnosis of High Intracranial Pressure by Non-Optic Nerve Retinal Image Features
Clinical Snapshot
PICO Framework
| P — Population | Adults with elevated intracranial pressure and papilledema (n=32, 70 images) versus control subjects (n=31, 62 images) |
| I — Intervention | Deep learning convolutional neural network model for retinal image classification |
| C — Comparator | Two neuro-ophthalmologist graders and comparison between optic nerve head-masked versus unmasked retinal images |
| O — Outcomes | Classification accuracy, sensitivity, specificity, and area under receiver operating characteristic curve for detecting elevated intracranial pressure |
Bottom Line
This small proof-of-concept study suggests that deep learning models can identify elevated intracranial pressure from retinal images even when optic nerve head features are masked, potentially using subtle peripheral retinal changes. While the reported accuracies (79-83%) and AUC values (0.90-0.94) appear promising, the study has significant limitations including very small sample size, missing key diagnostic accuracy metrics, and unclear methodology for the reference standard. The finding that non-optic nerve retinal features may be diagnostically useful is intriguing and could eventually assist in ICP assessment, particularly in settings with limited neuro-ophthalmology expertise. However, much larger validation studies with proper diagnostic accuracy reporting, clear reference standards, and clinical utility assessment are essential before considering clinical implementation. Australian clinicians should view this as early-stage research requiring substantial further validation.
Key Findings
P Value: P ≥ 0.38 for model vs grader comparisons; P = 0.02 for grader 1 masking effect
Effect Size: DL model accuracy: 83% (unmasked), 79% (masked); Expert grader accuracy: 66-93%
Primary Outcome: Classification accuracy for detecting elevated intracranial pressure from retinal images
Nnt Or Sensitivity: AUC 0.94 (unmasked), 0.90 (masked) - sensitivity and specificity not reported
Confidence Interval: Not reported
Clinical Application
Requires widefield retinal imaging capability and specialized deep learning infrastructure - limited current feasibility Could complement existing neuro-ophthalmology services in major Australian hospitals. Would require TGA approval for clinical use. May be particularly valuable in remote areas with limited specialist access. Patients suspected of having elevated intracranial pressure where papilledema assessment is challenging or unavailable
Abstract
PURPOSE: To assess the performance of an artificial intelligence deep learning (DL) model compared with neuro-ophthalmologists for the classification of subjects with elevated intracranial pressure (ICP) and papilledema versus control subjects, based on retinal images with and without masking the optic nerve heads (ONHs). METHODS: Widefield retinal images were obtained in 32 subjects (70 images) with elevated ICP and papilledema and 31 control subjects (62 images). ONH-unmasked images were generated by cropping the image to a 30° circular region. ONH-masked images were generated by masking the ONH and peripapillary region. Classification was performed using a convolutional neural network model and by two neuro-ophthalmologists (graders). RESULTS: For the ONH-unmasked images, the classification accuracies of the model, grader 1, and grader 2 were 83% (area under the receiver operating characteristic curve [AUC] = 0.94), 93%, and 86%, respectively. For the ONH-masked images, the classification accuracies of the model, grader 1, and grader 2 were 79% (AUC = 0.90), 66%, and 76%, respectively. The classification performance of the DL model and graders did not significantly differ for both datasets (P ≥ 0.38). There was no significant effect of ONH masking on the performance of the DL model (P = 1.0) and grader 2 (P = 0.38), whereas the performance of grader 1 was significantly reduced (P = 0.02). CONCLUSIONS: Both expert graders and the DL models demonstrated excellent performance for classifying retinal images of subjects with elevated and normal ICP using images with or without masking the ONH. TRANSLATIONAL RELEVANCE: Methods for assessment of non-optic nerve retinal image features have the potential to improve diagnosis and monitoring the progression and response to treatment of elevated ICP.
References
- 1.Abdolahi, F., Li, K. Z., Zou, Y., Moss, H. E., & Shahidi, M. (2026). Diagnosis of high intracranial pressure by non-optic nerve retinal image features. Translational Vision Science & Technology, 15(4), 2. https://doi.org/10.1167/tvst.15.4.2
Related Research
Indian journal of ophthalmology
Internet of things in assistive technology for people with visual impairment: A scoping review
1 Aug 2026
BMJ open ophthalmology
Screening for diabetic retinopathy with artificial intelligence in a primary care setting: a comparative cost analysis
1 Aug 2026
Science advances
Multispectral infrared-to-full-color upconversion expanding human vision
1 Aug 2026
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service