Cansu Yuksel Elgin

Dr. Cansu Yuksel Elgin

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 December 2024 in BMC Medical Ethics
Verified
Journal Article
Open access
PDF file
CECansu Yuksel Elgin
Cansu Yuksel Elgin
CECeyhun Elgin
Ceyhun Elgin
Background: Artificial intelligence-driven Clinical Decision Support Systems (AI-CDSS) are increasingly being integrated into healthcare for various purposes, including resource allocation. While these systems promise improved efficiency and decision-making, they also raise significant ethical concerns. This study aims to explore healthcare professionals’ perspectives on the ethical implications of using AI-CDSS for healthcare resource allocation. Methods: We conducted semi-structured qualitative interviews with 23 healthcare professionals, including physicians, nurses, administrators, and medical ethicists in Turkey. Interviews focused on participants’ views regarding the use of AI-CDSS in resource allocation, potential ethical challenges, and recommendations for responsible implementation. Data were analyzed using thematic analysis. Results: Participant responses are clustered around five pre-determined thematic areas: (1) balancing efficiency and equity in resource allocation, (2) the importance of transparency and explicability in AI-CDSS, (3) shifting roles and responsibilities in clinical decision-making, (4) ethical considerations in data usage and algorithm development, and (5) balancing cost-effectiveness and patient-centered care. Participants acknowledged the potential of AI-CDSS to optimize resource allocation but expressed concerns about exacerbating healthcare disparities, the need for interpretable AI models, changing professional roles, data privacy, and maintaining individualized care. Conclusions: The integration of AI-CDSS into healthcare resource allocation presents both opportunities and significant ethical challenges. Our findings underscore the need for robust ethical frameworks, enhanced AI literacy among healthcare professionals, interdisciplinary collaboration, and rigorous monitoring and evaluation processes. Addressing these challenges proactively is crucial for harnessing the potential of AI-CDSS while preserving the fundamental values of equity, transparency, and patient-centered care in healthcare delivery.

20 January 2022 in Clinical & Experimental Ophthalmology
Verified
Journal Article
Closed access
PDF file
CECansu Yuksel Elgin
Cansu Yuksel Elgin
DCDinah Chen
Dinah Chen
LALama A. Al‐Aswad
Lama A. Al‐Aswad

16 June 2015 in Current Eye Research
Verified
Journal Article
Closed access
PDF file
CECansu Yuksel Elgin
Cansu Yuksel Elgin
GIGüzin İsKeleli
Güzin İsKeleli
STSerap Talaz
Serap Talaz
SASibel Akyol
Sibel Akyol
Purpose: To compare tear films levels of various inflammatory cytokines in asymptomatic contact lens (CL) users. CL users of rigid gas-permeable CLs (RGPCL) (group 1) or silicone hydrogel CLs (SiHCL) (group 2) were compared with non-CL-using healthy subjects (group 3). Materials and Methods: Tear samples were collected from subjects in each group after ensuring that there were no complications secondary to CL wear in the CL-wearing participants. Tear-film levels of interleukins (ILs)-1β, -6, and -8; granulocyte-macrophage colony-stimulating factor (GM-CSF) (using the Luminex method); and leukotriene B4 (LTB4) (using the ELISA method) were determined. Cytokine levels were compared among the three groups using analysis-of-variance (ANOVA) and Kruskall–Wallis tests. Results: There were significant differences in concentrations of IL-1β, GM-CSF and LTB4 among the three groups (p = 0.002, p = 0.021 and p = 0.009, respectively), as shown by the Kruskall–Wallis test comparing all three groups for the three cytokines. There were no significant differences for IL-6 and IL-8 (p = 0.079 and 0.094, respectively) when all three groups were compared. Conclusions: There were substantial statistically significant differences between RGPCL users, SiHCL users and control subjects in levels of tear film cytokines. Although CL users were asymptomatic, changes in tear-film levels of several important inflammatory mediators revealed that a chronic inflammatory process occurs during CL wear.

18 January 2016 in International Journal of Ophthalmology
Verified
Journal Article
Open access
CECansu Yuksel Elgin
Cansu Yuksel Elgin
CECeyhun Elgin
Ceyhun Elgin
To report on intraocular pressure (IOP) after intravitreal injections of triamcinolone acetonide. Systematic literature review of studies that investigated the effects of an injection of triamcinolone intravitreal triamcinolone acetonide on IOP was conducted according to the Cochrane Collaboration methodology and the reported effects have been analyzed with Meta-analysis. We found that the IOP follows an inverted-U shape pattern over time starting with an average value of 14.81±1.22 mm Hg before the injection, rising to a maximum of 19.48±2.15 mm Hg after one month of injection and falling down to 16.16±1.92 mm Hg after 6mo. Moreover, country of study, age, previous history of glaucoma and gender compositions matter for cross-study were different in reported IOP changes. Our findings may be helpful in determining pressure elevation risk of intravitreal triamcinolone acetonide therapy as well as comparing it with those of more recent therapies such as the anti-vascular endothelial growth factor agents.

1 June 2018 in Contact Lens and Anterior Eye
Verified
Journal Article
Closed access
CECansu Yuksel Elgin
Cansu Yuksel Elgin
GIGuzin Iskeleli
Guzin Iskeleli
OAOvgu Aydin
Ovgu Aydin

1 January 2025 in Journal of Ophthalmology
Verified
Journal Article
Open access
CECansu Yuksel Elgin
Cansu Yuksel Elgin
Purpose: To propose a conceptual framework for an AI‐driven personalized glaucoma progression prediction system that integrates diverse data sources to enhance clinical decision‐making and improve patient outcomes. This framework is specifically designed to address healthcare disparities in glaucoma care through scalable AI technology that can function across diverse resource settings, from tertiary care centers to remote clinics. The system aims to democratize access to expert‐level glaucoma care while addressing challenges of bias, equity, and accessibility. Methods: The paper outlines a comprehensive framework consisting of four main components: (1) data integration and preprocessing, (2) AI model architecture and training, (3) personalized prediction generation, and (4) a clinical decision support interface. The framework leverages multimodal neural networks to analyze structural imaging data, functional test results, clinical measurements, and patient demographics. Results: The proposed framework addresses current limitations in glaucoma progression prediction by capturing complex interactions between various risk factors. Potential benefits include early detection of rapid progressors, optimized treatment strategies, improved patient counseling, and support for clinical research. Implementation challenges such as data quality, model interpretability, workflow integration, regulatory approval, and ethical considerations are discussed along with strategies to address them. Conclusions: The AI‐driven framework for glaucoma progression prediction represents a significant advancement in personalized glaucoma management. While challenges remain, the potential benefits in terms of preserved vision, improved quality of life, and more efficient healthcare delivery are substantial. Future research directions include incorporating genetic data, advanced imaging modalities, and federated learning techniques to further enhance the system’s capabilities and impact.