<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.2d1 20170631//EN" "JATS-journalpublishing1.dtd">
<article xlink="http://www.w3.org/1999/xlink" dtd-version="1.0" article-type="ophthalmology" lang="en">
  <front>
    <journal-meta>
      <journal-id journal-id-type="publisher">JOHS</journal-id>
      <journal-id journal-id-type="nlm-ta">Journ of Health Scien</journal-id>
      <journal-title-group>
        <journal-title>Journal of HealthCare Sciences</journal-title>
        <abbrev-journal-title abbrev-type="pubmed">Journ of Health Scien</abbrev-journal-title>
      </journal-title-group>
      <issn pub-type="ppub">2231-2196</issn>
      <issn pub-type="opub">0975-5241</issn>
      <publisher>
        <publisher-name>Radiance Research Academy</publisher-name>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id">528</article-id>
      <article-id pub-id-type="doi">http://dx.doi.org/10.52533/JOHS.2026.60504</article-id>
      <article-id pub-id-type="doi-url"/>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Ophthalmology</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>A Systematic Review of Diagnostic Accuracy and Clinical Validation Studies Using Artificial Intelligence for Detection of Non-Diabetic Retinal Diseases&#13;
</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Almadani</surname>
            <given-names>Abdulaziz Salman</given-names>
          </name>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Almutairi</surname>
            <given-names>Mohammed Naji</given-names>
          </name>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Kirat</surname>
            <given-names>Elyas Ali Mohammed</given-names>
          </name>
        </contrib>
      </contrib-group>
      <pub-date pub-type="ppub">
        <day>31</day>
        <month>05</month>
        <year>2026</year>
      </pub-date>
      <volume>6</volume>
      <issue>5</issue>
      <fpage>292</fpage>
      <lpage>313</lpage>
      <permissions>
        <copyright-statement>This article is copyright of Popeye Publishing, 2009</copyright-statement>
        <copyright-year>2009</copyright-year>
        <license license-type="open-access" href="http://creativecommons.org/licenses/by/4.0/">
          <license-p>This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0) Licence. You may share and adapt the material, but must give appropriate credit to the source, provide a link to the licence, and indicate if changes were made.</license-p>
        </license>
      </permissions>
      <abstract>
        <p>Background: Retinal diseases are often associated with diabetes; however, non-diabetic retinal diseases can be associated with various ocular conditions. Retinal diseases are diagnosed using imaging, such as fundus photography and optical coherence tomography (OCT). The implementation of artificial intelligence (AI) in ophthalmology improves diagnostic accuracy, enabling early diagnosis, optimizing workflows, and enhancing the overall quality of patient care. However, AI-based detection of retinal diseases and challenges with validation across different patient populations remain. This systematic review aims to examine existing research on the diagnostic accuracy and clinical validation of AI-based detection of non-diabetic retinal diseases.&#13;
&#13;
Methods: A systematic search of studies published in PubMed, Cochrane Library, and Science Direct was completed from inception through February 23, 2026, without geographic restriction. Major outcomes of interest included diagnostic accuracy and clinical validation outcomes of AI models and retinal imaging modalities outcomes. The target population was human participants of any age or sex who are diagnosed with non-diabetic retinal diseases. The QUADAS-2 assessment tool was used to evaluate the methodological quality and risk of bias in the included studies.&#13;
&#13;
Results: 15 studies were included for the systematic review. The included studies showed that the used AI models had high sensitivity, specificity, and diagnostic accuracy for several non-diabetic retinal conditions, with the highest performance being associated with age-related macular degeneration (AMD) and the lowest sensitivity being associated with glaucoma. Additionally, AI models improved efficiency, reduced examination time, and working load. However, the majority of the included models lacked external validation and provided low sensitivity for rare condition detection.&#13;
&#13;
Conclusion: AI-based models demonstrate high diagnostic accuracy and specificity for detecting non-diabetic retinal diseases and could serve as effective tools for screening and triage, particularly in resource limited areas. While promising, prospective studies and careful implementation strategies are essential to ensure efficiency, safety, and improved patient outcomes.&#13;
</p>
      </abstract>
      <kwd-group>
        <kwd>artificial intelligence</kwd>
        <kwd> non-diabetic retinal disease</kwd>
        <kwd> OCT</kwd>
        <kwd> fundus examination</kwd>
        <kwd> deep learning</kwd>
        <kwd> external validation</kwd>
      </kwd-group>
    </article-meta>
  </front>
</article>