MBMarian Blazes

Marian Blazes

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Marian Blazes has a total of 90 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    An LLM-assisted framework for accelerated and verifiable clinical hypothesis testing from electronic health records

    by Nayoon Gim, In Gim, Yu Jiang, Yuka Kihara, Marian Blazes, Yue Wu, Cecilia S Lee, Aaron Y Lee

    Acquiring insights from electronic health records (EHRs) is slowed by manual analytical workflows that limit scalability and reproducibility. We present LATCH (LLM-Assisted Testing of Clinical Hypotheses), an agentic framework that converts natural language clinical hypotheses into fully auditable analyses on structured EHR data. LATCH integrates LLM-assisted semantic layers with deterministic execution pipelines to automate cohort construction, statistical analysis, and result reporting, while isolating patient-level data from LLM-involved steps. Using diabetes as a model disease, LATCH reproduced findings from 20 published studies within 3-15 minutes per study. Beyond replication, LATCH enabled study extensions and new insight generation through simple natural language hypothesis modifications. We demonstrated LATCH across 102 hypothesis tests spanning reproduction, extension, and insight generation. We systematically stress-tested LATCH to characterize its limitations and operational boundaries. LATCH provides a scalable framework for reproducible real-world evidence generation, reducing analytical bottlenecks and improving reliability of AI-assisted biomedical discovery while preserving human oversight.

  • A Data-driven Age-related Macular Degeneration Severity Scoring System Leveraging the AREDS Studies and Clinical Electronic Medical Records

    by Cecilia S. Lee, Yu-Ru Su, Rod L. Walker, Chloe Krakauer, Marian Blazes, Eric A. Johnson, David Cronkite, Will Bowers, Chantelle Hess, David Arterburn, et al.


  • INCIDENCE AND PROGRESSION OF AGE-RELATED MACULAR DEGENERATION AMONG PATIENTS WITH AND WITHOUT OBSTRUCTIVE SLEEP APNEA

    by Marian Blazes, Caitlyn Bs Ngadisastra, Pei-Ru Bs Li, Cecilia S. Lee, Jiahn-Shing Lee, Lai-Chu See, Ken-Kuo Lin

    Purpose: Obstructive sleep apnea (OSA) may increase the risk of age-related macular degeneration (AMD) due to repetitive oxygen deprivation or other mechanisms, though whether OSA increases the risk of AMD progression is unknown. The authors analyzed associations between OSA and AMD risk in the Taiwanese population. Methods: The authors identified patients diagnosed with OSA between 2000 and 2018 in the Taiwan National Health Insurance Research Database and used 1:1 propensity score matching on demographics and comorbidities to create a non-OSA cohort. The authors used Cox proportional hazard modeling to investigate the risk of AMD and the risk of progression from nonexudative to exudative AMD in OSA versus non-OSA patients. Results: A total of 66,869 OSA patients were matched with 66,869 non-OSA patients. The hazard ratio of AMD in the OSA cohort was 1.36 (95% confidence interval, 1.29–1.43, P < 0.0001). The hazard ratio for progression from nonexudative to exudative AMD for the OSA cohort was 0.94 (95% confidence interval, 0.77–1.14, P = 0.5073). Conclusion: Obstructive sleep apnea is associated with a higher risk of developing AMD. However, no increased risk of AMD progression is observed among people with OSA and existing nonexudative AMD.


  • Outer Retinal Thinning is Associated With Brain Atrophy in Early Age-Related Macular Degeneration

    by Yu Jiang, Thomas Swain, Nayoon Gim, Marian Blazes, Christine Mac Donald, Ariel Rokem, Julia P. Owen, Niranjan Balu, Mark E. Clark, Lukas Goerdt, et al.


  • Open Access

    Comparing Telephone Survey Responses to Best-Corrected Visual Acuity to Estimate the Accuracy of Identifying Vision Loss: Validation Study

    by John Wittenborn, Aaron Lee, Elizabeth A Lundeen, Phoebe Lamuda, Jinan Saaddine, Grace L Su, Randy Lu, Aashka Damani, Jonathan S Zawadzki, Colin P Froines, et al.

    Background: Self-reported questions on blindness and vision problems are collected in many national surveys. Recently released surveillance estimates on the prevalence of vision loss used self-reported data to predict variation in the prevalence of objectively measured acuity loss among population groups for whom examination data are not available. However, the validity of self-reported measures to predict prevalence and disparities in visual acuity has not been established. Objective: This study aimed to estimate the diagnostic accuracy of self-reported vision loss measures compared to best-corrected visual acuity (BCVA), inform the design and selection of questions for future data collection, and identify the concordance between self-reported vision and measured acuity at the population level to support ongoing surveillance efforts. Methods: We calculated accuracy and correlation between self-reported visual function versus BCVA at the individual and population level among patients from the University of Washington ophthalmology or optometry clinics with a prior eye examination, randomly oversampled for visual acuity loss or diagnosed eye diseases. Self-reported visual function was collected via telephone survey. BCVA was determined based on retrospective chart review. Diagnostic accuracy of questions at the person level was measured based on the area under the receiver operator curve (AUC), whereas population-level accuracy was determined based on correlation. Results: The survey question, “Are you blind or do you have serious difficulty seeing, even when wearing glasses?” had the highest accuracy for identifying patients with blindness (BCVA ≤20/200; AUC=0.797). The highest accuracy for detecting any vision loss (BCVA <20/40) was achieved by responses of “fair,” “poor,” or “very poor” to the question, “At the present time, would you say your eyesight, with glasses or contact lenses if you wear them, is excellent, good, fair, poor, or very poor” (AUC=0.716). At the population level, the relative relationship between prevalence based on survey questions and BCVA remained stable for most demographic groups, with the only exceptions being groups with small sample sizes, and these differences were generally not significant. Conclusions: Although survey questions are not considered to be sufficiently accurate to be used as a diagnostic test at the individual level, we did find relatively high levels of accuracy for some questions. At the population level, we found that the relative prevalence of the 2 most accurate survey questions were highly correlated with the prevalence of measured visual acuity loss among nearly all demographic groups. The results of this study suggest that self-reported vision questions fielded in national surveys are likely to yield an accurate and stable signal of vision loss across different population groups, although the actual measure of prevalence from these questions is not directly analogous to that of BCVA.


  • Diabetic Retinopathy and Dementia Association, Beyond Diabetes Severity

    by Cecilia S. Lee, Chloe Krakauer, Yu-Ru Su, Rod L. Walker, Marian Blazes, Susan M. McCurry, James D. Bowen, Wayne C. McCormick, Aaron Y. Lee, Edward J. Boyko, et al.


  • Open Access

    Comparing Telephone Survey Responses to Best-Corrected Visual Acuity to Estimate the Accuracy of Identifying Vision Loss: Validation Study (Preprint)

    by John Wittenborn, Aaron Lee, Elizabeth A Lundeen, Phoebe Lamuda, Jinan Saaddine, Grace L Su, Randy Lu, Aashka Damani, Jonathan S Zawadzki, Colin P Froines, et al.

    BACKGROUND Self-reported questions on blindness and vision problems are collected in many national surveys. Recently released surveillance estimates on the prevalence of vision loss used self-reported data to predict variation in the prevalence of objectively measured acuity loss among population groups for whom examination data are not available. However, the validity of self-reported measures to predict prevalence and disparities in visual acuity has not been established. OBJECTIVE This study aimed to estimate the diagnostic accuracy of self-reported vision loss measures compared to best-corrected visual acuity (BCVA), inform the design and selection of questions for future data collection, and identify the concordance between self-reported vision and measured acuity at the population level to support ongoing surveillance efforts. METHODS We calculated accuracy and correlation between self-reported visual function versus BCVA at the individual and population level among patients from the University of Washington ophthalmology or optometry clinics with a prior eye examination, randomly oversampled for visual acuity loss or diagnosed eye diseases. Self-reported visual function was collected via telephone survey. BCVA was determined based on retrospective chart review. Diagnostic accuracy of questions at the person level was measured based on the area under the receiver operator curve (AUC), whereas population-level accuracy was determined based on correlation. RESULTS The survey question, “Are you blind or do you have serious difficulty seeing, even when wearing glasses?” had the highest accuracy for identifying patients with blindness (BCVA ≤20/200; AUC=0.797). The highest accuracy for detecting any vision loss (BCVA <20/40) was achieved by responses of “fair,” “poor,” or “very poor” to the question, “At the present time, would you say your eyesight, with glasses or contact lenses if you wear them, is excellent, good, fair, poor, or very poor” (AUC=0.716). At the population level, the relative relationship between prevalence based on survey questions and BCVA remained stable for most demographic groups, with the only exceptions being groups with small sample sizes, and these differences were generally not significant. CONCLUSIONS Although survey questions are not considered to be sufficiently accurate to be used as a diagnostic test at the individual level, we did find relatively high levels of accuracy for some questions. At the population level, we found that the relative prevalence of the 2 most accurate survey questions were highly correlated with the prevalence of measured visual acuity loss among nearly all demographic groups. The results of this study suggest that self-reported vision questions fielded in national surveys are likely to yield an accurate and stable signal of vision loss across different population groups, although the actual measure of prevalence from these questions is not directly analogous to that of BCVA.

  • Training Deep Learning Models to Work on Multiple Devices by Cross-Domain Learning with No Additional Annotations

    by Yue Wu, Abraham Olvera-Barrios, Ryan Yanagihara, Timothy-Paul H. Kung, Randy Lu, Irene Leung, Amit V. Mishra, Hanan Nussinovitch, Gabriela Grimaldi, Marian Blazes, et al.


  • Endophthalmitis Rate in Immediately Sequential versus Delayed Sequential Bilateral Cataract Surgery within the Intelligent Research in Sight (IRIS®) Registry Data

    by Megan Lacy, Timothy-Paul H. Kung, Julia P. Owen, Ryan T. Yanagihara, Marian Blazes, Suzann Pershing, Leslie G. Hyman, Russell N. Van Gelder, Aaron Y. Lee, Cecilia S. Lee, et al.


  • Open Access

    Understanding the Brain through Aging Eyes

    by Marian Blazes, Cecilia S. Lee

    The eye and brain share common mechanisms of aging and disease, thus the retina is an essential source of accessible information about neurodegenerative processes occurring in the brain. Advances in retinal imaging have led to the discovery of many potential biomarkers of Alzheimer’s disease, although further research is needed to validate these associations. Understanding the mechanisms of retinal disease in the context of aging will extend our knowledge of AD and may enable advancements in diagnosis, monitoring, and treatment.