Cristiano Oliveira’s research while affiliated with Weill Cornell Medicine and other places

Importance Prompt and accurate diagnosis of arteritic anterior ischemic optic neuropathy (AAION) from giant cell arteritis and other systemic vasculitis can contribute to preventing irreversible vision loss from these conditions. Its clinical distinction from nonarteritic anterior ischemic optic neuropathy (NAION) can be challenging, especially when systemic symptoms are lacking or laboratory markers of the disease are not reliable. Objective To develop, train, and test a deep learning system (DLS) to discriminate AAION from NAION on color fundus images during the acute phase. Design, Setting, and Participants This was an international study including color fundus images of 961 eyes of 802 patients with confirmed AAION and NAION. Training was performed using images from 21 expert neuro-ophthalmology centers in 16 countries, while external testing was performed in a cohort from 5 expert neuro-ophthalmology centers in the US and Europe. Data for training and external testing were collected from August 2018 to January 2023. A mix of deidentified images of 2 fields of view (optic disc centered and macula centered) were used. For training and internal validation, images were from 16 fundus camera models with fields of 30° to 55°. For external testing, images were from 5 fundus cameras with fields of 30° to 50°. Data were analyzed from January 2023 to January 2024. Main Outcomes and Measures The performance of the DLS was measured using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. Results In the training and validation sets, 374 (54.9%) of patients were female, 301 (44.2%) were male, and 6 (0.9%) were of unknown sex; the median (range) age was 66 (23-96) years. When tested on the external dataset including 121 patients (35 [28.9%] female, 44 [36.4%] male, and 42 [34.7%] of unknown sex; median [range] age, 69 [37-89] years), the DLS achieved an AUC of 0.97 (95% CI, 0.95-0.99), a sensitivity of 91.1% (95% CI, 85.2-96.9), a specificity of 93.4% (95% CI, 91.1-98.2), and an accuracy of 92.6% (95% CI, 90.5-96.6). The accuracy of the 2 experts for classification of the same dataset was 74.3% (95% CI, 66.7-81.9) and 81.6% (95% CI, 74.8-88.4), respectively. Conclusions and Relevance A DLS showing disease-specific averaged class-activation maps had greater than 90% accuracy at discriminating between acute AAION from NAION on color fundus images, at the eye level, without any clinical or biomarker information. A DLS that identifies AAION could improve clinical decision-making, potentially reducing the risk of misdiagnosis and improving patient outcomes.

Background: The use of artificial intelligence is becoming more prevalence in medicine with numerous successful examples in ophthalmology. However, much of the work has been focused on replicating the works of ophthalmologists. Given the analytical potentials of artificial intelligence, it is plausible that artificial intelligence can detect microfeatures not readily distinguished by humans. In this study, we tested the potential for artificial intelligence to detect early optic coherence tomography changes to predict progression toward papilledema or glaucoma when no significant changes are detected on optical coherence tomography by clinicians. Methods: Prediagnostic optical coherence tomography of patients who developed papilledema (n = 93, eyes = 166) and glaucoma (n = 187, eyes = 327) were collected. Given discrepancy in average cup-to-disc ratios of the experimental groups, control groups for papilledema (n = 254, eyes = 379) and glaucoma (n = 441, eyes = 739) are matched by cup-to-disc ratio. Publicly available Visual Geometry Group-19 model is retrained using each experimental group and its respective control group to predict progression to papilledema or glaucoma. Images used for training include retinal nerve fiber layer thickness map, extracted vertical tomogram, ganglion cell thickness map, and ILM-RPE thickness map. Results: Trained model was able to predict progression to papilledema with a precision of 0.714 and a recall of 0.769 when trained with retinal nerve fiber layer thickness map, but not other image types. However, trained model was able to predict progression to glaucoma with a precision of 0.682 and recall of 0.857 when trained with extracted vertical tomogram, but not other image types. Area under precision-recall curve of 0.826 and 0.785 were achieved for papilledema and glaucoma models, respectively. Conclusions: Computational and analytical power of computers have become an invaluable part of our lives and research endeavors. Our proof-of-concept study showed that artificial intelligence (AI) algorithms have the potential to detect early changes on optical coherence tomography for prediction of progression that is not readily observed by clinicians. Further research may help establish possible AI models that can assist with early diagnosis or risk stratification in ophthalmology.

Background: The novel coronavirus 2019 (COVID-19) pandemic has transformed health care. With the need to limit COVID-19 exposures, telemedicine has become an increasingly important format for clinical care. Compared with other fields, neuro-ophthalmology faces unique challenges, given its dependence on physical examination signs that are difficult to elicit outside the office setting. As such, it is imperative to understand both patient and provider experiences to continue to adapt the technology and tailor its application. The purpose of this study is to analyze both neuro-ophthalmology physician and patient satisfaction with virtual health visits during the time of the COVID-19 pandemic. Methods: Across three institutions (NYU Langone Health, Indiana University Health, and Columbia University Medical Center), telemedicine surveys were administered to 159 patients. Neuro-ophthalmologists completed 157 surveys; each of these were linked to a single patient visit. Patient surveys consisted of 5 questions regarding visit preparation, satisfaction, challenges, and comfort. The physician survey included 4 questions that focused on ability to gather specific clinical information by history and examination. Results: Among 159 patients, 104 (65.4%) reported that they were satisfied with the visit, and 149 (93.7%) indicated that they were comfortable asking questions. Sixty-eight (73.9%) patients found the instructions provided before the visit easy to understand. Potential areas for improvement noted by patients included more detailed preparation instructions and better technology (phone positioning, Internet connection, and software). More than 87% (137/157) of neuro-ophthalmologists surveyed reported having performed an examination that provided enough information for medical decision-making. Some areas of the neuro-ophthalmologic examination were reported to be easy to conduct (range of eye movements, visual acuity, Amsler grids, Ishihara color plates, and pupillary examination). Other components were more difficult (saccades, red desaturation, visual fields, convergence, oscillations, ocular alignment, and smooth pursuit); some were especially challenging (vestibulo-ocular reflex [VOR], VOR suppression, and optokinetic nystagmus). Clinicians noted that virtual health visits were limited by patient preparation, inability to perform certain parts of the examination (funduscopy and pupils), and technological issues. Conclusions: Among virtual neuro-ophthalmology visits evaluated, most offer patients with appointments that satisfy their needs. Most physicians in this cohort obtained adequate clinical information for decision-making. Even better technology and instructions may help improve aspects of virtual health visits.

Background Venous sinus stenting (VSS) is an accepted and minimally invasive treatment for adult idiopathic intracranial hypertension (IIH) associated with lateral sinus stenosis (LSS). The efficacy and safety of venous sinus stenting (VSS) in children with IIH has not been established. Methods This is a retrospective analysis of IIH patients 18 years of age or younger with LSS treated with VSS at our institution. Included patients have fulminant disease course or are refractory or intolerant to medical management. Results Eight patients were identified; 4 males and 4 females. Mean age is 13.4 years (range 4–18). All patients had severe headaches, 5 had blurred vision, 3 had diplopia and 3 had pulsatile tinnitus. Papilledema was present in 4 patients. Three patients had prior surgical procedures. Four patients were intolerant to medical management, 3 were refractory and 1 had fulminant course. Cerebral venography demonstrated severe stenosis of the dominant sinus in 6 patients and of bilateral co-dominant sinuses in 2 patients. Six patients had intrinsic stenosis and 2 had extrinsic stenosis. Venous sinus stenting (VSS) resulted in improvement of symptoms, papilledema and normalization of CSF opening pressure in 7 patients. No immediate complications were observed. Mean follow-up period is 21 months (range 6–42). Two patients required re-stenting; one responded well and the other had persistent symptoms and underwent subsequent surgical procedures of CSF diversion, suboccipital decompression and duraplasty which were also ineffective. Conclusion VSS may provide a viable option for pediatric IIH patients who are intolerant to medication, have failed conservative management or prior surgical interventions, or present with fulminant disease.