{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,6]],"date-time":"2026-11-06T23:53:18Z","timestamp":1794009198788,"version":"4.8.0"},"reference-count":52,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec sec-type=\"background\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Chronic kidney disease (CKD) is a prevalent condition with significant global health implications. Early detection and management are critical to prevent disease progression and complications. Deep learning (DL) models using retinal images have emerged as potential noninvasive screening tools for CKD, though their performance may be limited, especially in identifying individuals with proteinuria and in specific subgroups.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"objective\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>We aim to evaluate the efficacy of integrating retinal images and urine dipstick data into DL models for enhanced CKD diagnosis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"methods\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The 3 models were developed and validated: eGFR-RIDL (estimated glomerular filtration rate\u2013retinal image deep learning), eGFR-UDLR (logistic regression using urine dipstick data), and eGFR-MMDL (multimodal deep learning combining retinal images and urine dipstick data). All models were trained to predict an eGFR&lt;60 mL\/min\/1.73 m\u00b2, a key indicator of CKD, calculated using the 2009 CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation. This study used a multicenter dataset of participants aged 20\u201079 years, including a development set (65,082 people) and an external validation set (58,284 people). Wide Residual Networks were used for DL, and saliency maps were used to visualize model attention. Sensitivity analyses assessed the impact of numerical variables.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"results\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      eGFR-MMDL outperformed eGFR-RIDL in both the test and external validation sets, with area under the curves of 0.94 versus 0.90 and 0.88 versus 0.77 (\n                      <jats:italic>P<\/jats:italic>\n                      &lt;.001 for both, DeLong test). eGFR-UDLR outperformed eGFR-RIDL and was comparable to eGFR-MMDL, particularly in the external validation. However, in the subgroup analysis, eGFR-MMDL showed improvement across all subgroups, while eGFR-UDLR demonstrated no such gains. This suggested that the enhanced performance of eGFR-MMDL was not due to urine data alone, but rather from the synergistic integration of both retinal images and urine data. The eGFR-MMDL model demonstrated the best performance in individuals younger than 65 years or those with proteinuria. Age and proteinuria were identified as critical factors influencing model performance. Saliency maps indicated that urine data and retinal images provide complementary information, with urine offering insights into retinal abnormalities and retinal images, particularly the arcade vessels, being key for predicting kidney function.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"conclusions\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The MMDL model integrating retinal images and urine dipstick data show significant promise for noninvasive CKD screening, outperforming the retinal image\u2013only model. However, routine blood tests are still recommended for individuals aged 65 years and older due to the model\u2019s limited performance in this age group.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/55825","type":"journal-article","created":{"date-parts":[[2025,2,9]],"date-time":"2025-02-09T21:23:26Z","timestamp":1739136206000},"page":"e55825-e55825","source":"Crossref","is-referenced-by-count":15,"title":["Diagnosis of Chronic Kidney Disease Using Retinal Imaging and Urine Dipstick Data: Multimodal Deep Learning Approach"],"prefix":"10.2196","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-7466-8786","authenticated-orcid":false,"given":"Youngmin","family":"Bhak","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5231-0551","authenticated-orcid":false,"given":"Yu Ho","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3717-2223","authenticated-orcid":false,"given":"Joonhyung","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2658-4342","authenticated-orcid":false,"given":"Kiwon","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-8129-4548","authenticated-orcid":false,"given":"Daehwan","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-4745-0167","authenticated-orcid":false,"given":"Eun Chan","family":"Jang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4267-5586","authenticated-orcid":false,"given":"Eunjeong","family":"Jang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5054-9470","authenticated-orcid":false,"given":"Christopher Seungkyu","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0364-4675","authenticated-orcid":false,"given":"Eun Seok","family":"Kang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-6020-4446","authenticated-orcid":false,"given":"Sehee","family":"Park","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6918-5694","authenticated-orcid":false,"given":"Hyun Wook","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6903-6333","authenticated-orcid":false,"given":"Sang Min","family":"Nam","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,2,7]]},"reference":[{"issue":"3","key":"R1","doi-asserted-by":"publisher","first-page":"20220070","DOI":"10.1002\/VIW.20220070","article-title":"Retinal image\u2010based artificial intelligence in detecting and predicting kidney diseases: current advances and future perspectives","volume":"4","author":"Wen","journal-title":"VIEW"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1016\/j.kisu.2021.11.003","article-title":"Epidemiology of chronic kidney disease: an update 2022","volume":"12","author":"Kovesdy","journal-title":"Kidney Int Suppl (2011)"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1038\/kisup.2012.64","article-title":"Chapter 1: definition and classification of CKD","volume":"3","journal-title":"Kidney Int Suppl (2011)"},{"issue":"13","key":"R4","doi-asserted-by":"publisher","first-page":"1296","DOI":"10.1056\/NEJMoa041031","article-title":"Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization","volume":"351","author":"Go","journal-title":"N Engl J Med"},{"issue":"12","key":"R5","doi-asserted-by":"publisher","first-page":"1341","DOI":"10.1038\/ki.2010.536","article-title":"Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts","volume":"79","author":"van der Velde","journal-title":"Kidney Int"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1038\/ki.2010.531","article-title":"Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts","volume":"80","author":"Gansevoort","journal-title":"Kidney Int"},{"issue":"9731","key":"R7","doi-asserted-by":"publisher","first-page":"2073","DOI":"10.1016\/S0140-6736(10)60674-5","article-title":"Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis","volume":"375","author":"Chronic Kidney Disease Prognosis Consortium","journal-title":"Lancet"},{"issue":"10075","key":"R8","doi-asserted-by":"publisher","first-page":"1238","DOI":"10.1016\/S0140-6736(16)32064-5","article-title":"Chronic kidney disease","volume":"389","author":"Webster","journal-title":"The Lancet"},{"issue":"6","key":"R9","doi-asserted-by":"publisher","first-page":"1290","DOI":"10.1038\/ki.2013.491","article-title":"Kidney and eye diseases: common risk factors, etiological mechanisms, and pathways","volume":"85","author":"Wong","journal-title":"Kidney Int"},{"issue":"2","key":"R10","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1016\/j.kint.2020.01.039","article-title":"The eye, the kidney, and cardiovascular disease: old concepts, better tools, and new horizons","volume":"98","author":"Farrah","journal-title":"Kidney Int"},{"issue":"5","key":"R11","doi-asserted-by":"publisher","first-page":"867","DOI":"10.2215\/CJN.08271109","article-title":"Prevalence of ocular fundus pathology in patients with chronic kidney disease","volume":"5","author":"Grunwald","journal-title":"Clin J Am Soc Nephrol"},{"issue":"9","key":"R12","doi-asserted-by":"publisher","first-page":"1136","DOI":"10.1001\/archophthalmol.2012.1800","article-title":"Retinopathy and chronic kidney disease in the Chronic Renal Insufficiency Cohort (CRIC) study","volume":"130","author":"Grunwald","journal-title":"Arch Ophthalmol"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"2970","DOI":"10.1167\/iovs.13-11941","article-title":"Structural changes in the retinal microvasculature and renal function","volume":"54","author":"Lim","journal-title":"Invest Ophthalmol Vis Sci"},{"issue":"5","key":"R14","doi-asserted-by":"publisher","first-page":"625","DOI":"10.1093\/aje\/kwn367","article-title":"Retinal microvascular caliber and chronic kidney disease in an Asian population","volume":"169","author":"Sabanayagam","journal-title":"Am J Epidemiol"},{"issue":"6","key":"R15","doi-asserted-by":"publisher","first-page":"e295","DOI":"10.1016\/S2589-7500(20)30063-7","article-title":"A deep learning algorithm to detect chronic kidney disease from retinal photographs in community-based populations","volume":"2","author":"Sabanayagam","journal-title":"Lancet Digit Health"},{"issue":"6","key":"R16","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/s41551-021-00745-6","article-title":"Deep-learning models for the detection and incidence prediction of chronic kidney disease and type 2 diabetes from retinal fundus images","volume":"5","author":"Zhang","journal-title":"Nat Biomed Eng"},{"issue":"6","key":"R17","doi-asserted-by":"publisher","first-page":"1114","DOI":"10.1093\/jamia\/ocad051","article-title":"Machine-learning enhancement of urine dipstick tests for chronic kidney disease detection","volume":"30","author":"Jang","journal-title":"J Am Med Inform Assoc"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1038\/s41551-019-0487-z","article-title":"Detection of anaemia from retinal fundus images via deep learning","volume":"4","author":"Mitani","journal-title":"Nat Biomed Eng"},{"issue":"2","key":"R19","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1681\/ASN.2015121308","article-title":"Renal aging: causes and consequences","volume":"28","author":"O\u2019Sullivan","journal-title":"J Am Soc Nephrol"},{"issue":"10","key":"R20","doi-asserted-by":"publisher","first-page":"1359","DOI":"10.1001\/jamainternmed.2021.4813","article-title":"Accounting for age in the definition of chronic kidney disease","volume":"181","author":"Liu","journal-title":"JAMA Intern Med"},{"issue":"4","key":"R21","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1093\/ckj\/sfaa162","article-title":"Hypertensive nephropathy: a major roadblock hindering the advance of precision nephrology","volume":"13","author":"Carriazo","journal-title":"Clin Kidney J"},{"issue":"5","key":"R22","first-page":"120","volume":"39","author":"Siwek","journal-title":"Am Fam Physician"},{"key":"R23","unstructured":"Pastor E Alfaro L Baralis E . Identifying biased subgroups in ranking and classification. arXiv. Preprint posted online onAug17, 2021. [doi: 10.48550\/arXiv.2108.07450]"},{"issue":"11","key":"R24","doi-asserted-by":"publisher","DOI":"10.2196\/23472","article-title":"Deep learning\u2013based detection of early renal function impairment using retinal fundus images: model development and validation","volume":"8","author":"Kang","journal-title":"JMIR Med Inform"},{"issue":"9","key":"R25","doi-asserted-by":"publisher","first-page":"604","DOI":"10.7326\/0003-4819-150-9-200905050-00006","article-title":"A new equation to estimate glomerular filtration rate","volume":"150","author":"Levey","journal-title":"Ann Intern Med"},{"key":"R26","unstructured":"Zagoruyko S Komodakis N . Wide residual networks. arXiv. Preprint posted online onMay23, 2016. [doi: 10.48550\/arXiv.1605.07146]"},{"key":"R27","doi-asserted-by":"crossref","unstructured":"Heiliger L Sekuboyina A Menze B Egger J Kleesiek J . Beyond medical imaging - a review of multimodal deep learning in radiology. TechRxiv. Preprint posted online onOct30, 2023. [doi: 10.36227\/techrxiv.19103432.v1]","DOI":"10.36227\/techrxiv.19103432"},{"key":"R28","unstructured":"MultimodalCKD. GitHub. URL: https:\/\/github.com\/forustous\/multimodalCKD [Accessed 28-01-2025]"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","DOI":"10.1186\/s12938-023-01110-1","article-title":"Ocular images-based artificial intelligence on systemic diseases","volume":"22","author":"Tan","journal-title":"Biomed Eng Online"},{"key":"R30","doi-asserted-by":"crossref","unstructured":"Wang Y Cao Y Zha ZJ Zhang J Xiong Z . Deep degradation prior for low-quality image classification. Presented at: 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR); Jun 13-19, 2020:11046-11055; Seattle, WA, USA. URL: https:\/\/ieeexplore.ieee.org\/xpl\/mostRecentIssue.jsp?punumber=9142308 [Accessed 18-01-2025] [doi: 10.1109\/CVPR42600.2020.01106]","DOI":"10.1109\/CVPR42600.2020.01106"},{"issue":"2","key":"R31","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1167\/tvst.7.2.10","article-title":"Automated method of grading vitreous haze in patients with uveitis for clinical trials","volume":"7","author":"Passaglia","journal-title":"Transl Vis Sci Technol"},{"issue":"12","key":"R32","doi-asserted-by":"publisher","first-page":"1904","DOI":"10.1093\/jamia\/ocad179","article-title":"Deep learning algorithms to detect diabetic kidney disease from retinal photographs in multiethnic populations with diabetes","volume":"30","author":"Betzler","journal-title":"J Am Med Inform Assoc"},{"issue":"1","key":"R33","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-020-00341-z","article-title":"Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines","volume":"3","author":"Huang","journal-title":"NPJ Digit Med"},{"issue":"4","key":"R34","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1016\/j.healthpol.2018.10.013","article-title":"Factors affecting participation in health checkups: evidence from Japanese survey data","volume":"123","author":"Noguchi","journal-title":"Health Policy"},{"issue":"1","key":"R35","doi-asserted-by":"publisher","first-page":"873","DOI":"10.1038\/s41598-020-79531-4","article-title":"Baseline data report of the China Dialysis Outcomes and Practice Patterns Study (DOPPS)","volume":"11","author":"Zhao","journal-title":"Sci Rep"},{"issue":"9","key":"R36","doi-asserted-by":"publisher","first-page":"2469","DOI":"10.1097\/01.ASN.0000136133.28194.E4","article-title":"Retinal microvascular abnormalities and renal dysfunction: the atherosclerosis risk in communities study","volume":"15","author":"Wong","journal-title":"J Am Soc Nephrol"},{"issue":"1","key":"R37","doi-asserted-by":"publisher","first-page":"179","DOI":"10.2337\/diabetes.53.1.179","article-title":"Retinal vessel diameters and the incidence of gross proteinuria and renal insufficiency in people with type 1 diabetes","volume":"53","author":"Wong","journal-title":"Diabetes"},{"issue":"5","key":"R38","doi-asserted-by":"publisher","first-page":"682","DOI":"10.1053\/j.ajkd.2010.11.025","article-title":"Bidirectional association of retinal vessel diameters and estimated GFR decline: the Beaver Dam CKD Study","volume":"57","author":"Sabanayagam","journal-title":"Am J Kidney Dis"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1053\/j.ajkd.2011.02.382","article-title":"Retinal arteriolar narrowing and subsequent development of CKD Stage 3: the Multi-Ethnic Study of Atherosclerosis (MESA)","volume":"58","author":"Yau","journal-title":"Am J Kidney Dis"},{"issue":"1","key":"R40","doi-asserted-by":"publisher","first-page":"9374","DOI":"10.1038\/s41598-017-09204-2","article-title":"Retinal vascular imaging markers and incident chronic kidney disease: a prospective cohort study","volume":"7","author":"Yip","journal-title":"Sci Rep"},{"issue":"1","key":"R41","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00860-5","article-title":"Non-invasive chronic kidney disease risk stratification tool derived from retina-based deep learning and clinical factors","volume":"6","author":"Joo","journal-title":"NPJ Digit Med"},{"issue":"1","key":"R42","doi-asserted-by":"publisher","DOI":"10.1038\/kisup.2012.73","article-title":"Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease","volume":"3","author":"Levin","journal-title":"Kidney Int Suppl"},{"issue":"3","key":"R43","doi-asserted-by":"publisher","first-page":"461","DOI":"10.1093\/ndt\/11.3.461","article-title":"Proteinuria and blood pressure as causal components of progression to end-stage renal failure. Northern Italian Cooperative Study Group","volume":"11","author":"Locatelli","journal-title":"Nephrol Dial Transplant"},{"issue":"12","key":"R44","doi-asserted-by":"publisher","first-page":"3217","DOI":"10.1097\/01.asn.0000100145.27188.33","article-title":"Management of glomerular proteinuria: a commentary","volume":"14","author":"Wilmer","journal-title":"J Am Soc Nephrol"},{"issue":"6","key":"R45","first-page":"1153","volume":"71","author":"Simerville","journal-title":"Am Fam Physician"},{"issue":"2","key":"R46","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1159\/000330487","article-title":"Low urine pH Is a predictor of chronic kidney disease","volume":"35","author":"Nakanishi","journal-title":"Kidney Blood Press Res"},{"key":"R47","doi-asserted-by":"publisher","DOI":"10.1038\/srep15242","article-title":"Microscopic haematuria and clinical outcomes in patients with stage 3-5 nondiabetic chronic kidney disease","volume":"5","author":"You-Hsien Lin","journal-title":"Sci Rep"},{"issue":"4","key":"R48","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1053\/j.ackd.2010.05.005","article-title":"Aging in kidney disease: key issues and gaps in knowledge","volume":"17","author":"Fried","journal-title":"Adv Chronic Kidney Dis"},{"issue":"7","key":"R49","doi-asserted-by":"publisher","first-page":"1334","DOI":"10.1097\/HJH.0000000000002050","article-title":"The effect of SGLT-2 inhibitors on albuminuria and proteinuria in diabetes mellitus: a systematic review and meta-analysis of randomized controlled trials","volume":"37","author":"Piperidou","journal-title":"J Hypertens"},{"issue":"1","key":"R50","doi-asserted-by":"publisher","first-page":"30","DOI":"10.7326\/0003-4819-148-1-200801010-00190","article-title":"Meta-analysis: effect of monotherapy and combination therapy with inhibitors of the renin angiotensin system on proteinuria in renal disease","volume":"148","author":"Kunz","journal-title":"Ann Intern Med"},{"issue":"10225","key":"R51","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(19)32977-0","article-title":"Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017","volume":"395","author":"GBD Chronic Kidney Disease Collaboration","journal-title":"Lancet"},{"issue":"2","key":"R52","doi-asserted-by":"publisher","DOI":"10.1136\/bmjgh-2016-000256","article-title":"Chronic kidney disease in low-income to middle-income countries: the case for increased screening","volume":"2","author":"George","journal-title":"BMJ Glob Health"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2025\/1\/e55825","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:44:15Z","timestamp":1782841455000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2025\/1\/e55825"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,7]]},"references-count":52,"URL":"https:\/\/doi.org\/10.2196\/55825","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,7]]},"article-number":"v13i3e55825"}}