Eye Movement Abnormalities in Glaucoma Patients: A Review.

Glaucoma is a common condition that relies on careful clinical assessment to diagnose and determine disease progression. There is growing evidence that glaucoma is associated not only with loss of retinal ganglion cells but also with degeneration of cortical and subcortical brain structures associated with vision and eye movements. The effect of glaucoma pathophysiology on eye movements is not well understood. In this review, we examine the evidence surrounding altered eye movements in glaucoma patients compared to healthy controls, with a focus on quantitative eye tracking studies measuring saccades, fixation, and optokinetic nystagmus in a range of visual tasks. The evidence suggests that glaucoma patients have alterations in several eye movement domains. Patients exhibit longer saccade latencies, which worsen with increasing glaucoma severity. Other saccadic abnormalities include lower saccade amplitude and velocity, and difficulty inhibiting reflexive saccades. Fixation is pathologically altered in glaucoma with reduced stability. Optokinetic nystagmus measures have also been shown to be abnormal. Complex visual tasks (eg reading, driving, and navigating obstacles), integrate these eye movements and result in behavioral adaptations. The review concludes with a summary of the evidence and recommendations for future research in this emerging field.

Development of an artificial intelligence system to classify pathology and clinical features on retinal fundus images.

IMPORTANCE: Artificial intelligence (AI) algorithms are under development for use in diabetic retinopathy photo screening pathways. To be clinically acceptable, such systems must also be able to classify other fundus abnormalities and clinical features at the point of care. BACKGROUND: We aimed to develop an AI system that can detect several fundus pathologies and report relevant clinical features. DESIGN: Convolutional neural network training with retrospective data set. PARTICIPANTS: Colour fundus photos were obtained from publicly available fundus image databases. METHODS: Images were uploaded to a web-based AI platform for training and validation of AI classifiers. Separate classifiers were created for each fundus pathology and clinical feature. MAIN OUTCOME MEASURES: Accuracy, sensitivity, specificity and area under receiver operating characteristic curve (AUC) for each classifier. RESULTS: We obtained 4435 images from publicly available fundus image databases. AI classifiers were developed for each disease state above. Although statistical performance was limited by the small sample size, average accuracy was 89%, average sensitivity was 75%, average specificity was 89% and average AUC was 0.58. CONCLUSION AND RELEVANCE: This study is a proof-of-concept AI system that could be implemented within a diabetic photo-screening pathway. Performance was promising but not yet at the level that would be required for clinical application. We have shown that it is possible for clinicians to develop AI classifiers with no previous programming or AI knowledge, using standard laptop computers.

A novel pathway for investigation of colorectal symptoms with colonoscopy or computed tomography colonography.

BACKGROUND: Colorectal cancer (CRC) is a common problem in New Zealand and there is significant pressure on colonoscopy resources. Lower gastrointestinal symptoms are common in the community hence the appropriate selection of patients for colonoscopy is problematic. The Canterbury District Health Board recently developed the Canterbury Colorectal Symptom Pathway (CCrSP) to attempt to improve prioritisation using a combination of presenting clinical features integrated into a scoring tool. This study describes that pathway and its outcomes over a 6-month period. METHOD: Following implementation of the CCrSP, all outpatient referrals receiving colonoscopy or Computerised Tomography Colonography (CTC) over a 6-month period were audited. The clinical features included in the referral, waiting time and outcome of investigation were recorded. Using the scoring tool, a score was calculated for all referrals and compared with outcome. RESULTS: Some 1,369 procedures were performed during the study period. Of the symptomatic patients, 38 CRCs were diagnosed from 633 colonoscopies and 253 CTCs. Individual factors predictive for CRC were rectal bleeding (OR 2.1, 95%CI 1.1-4.2), iron deficiency anaemia (OR 3.2, 95%CI 1.6-6.3) and positive faecal occult blood test (OR 6.1, 95%CI 2.1-16.3). No CRCs were diagnosed in the group scoring below the pre-set threshold for investigation. Multiple logistic regression analysis demonstrated a 1 unit increase in score increased the likelihood of CRC by 7.2% (95%CI 4.4%-10.1%, p<0.001). Of the 11 CRCs suggested by CTC, there was one false positive. The follow up colonoscopy rate after CTC was 11.5% and further radiology was recommended in 7.9%. CONCLUSION: The CCrSP pathway was accurate for predicting CRC and offers a reliable triage tool. The scoring tool was both sensitive for CRC and predictive of the risk of CRC in patients who received colonoscopy or CTC.