{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T10:04:18Z","timestamp":1790762658924,"version":"4.1.0"},"reference-count":30,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Endocrinol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>\n                      Polycystic ovary syndrome (PCOS) is a heterogeneous endocrine disease condition that affects 8%\u201313% of the female population worldwide. Its clinical heterogeneity makes the binary approach to phenotyping less effective for the biological significance of such phenotypic groups and the personalized management of PCOS. To overcome this issue, we present the expert knowledge-based weighted\n                      <jats:italic>K<\/jats:italic>\n                      -means (EKWKM) approach.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>The proposed EKWKM is a new, weakly supervised phenotyping approach that uses structured clinical knowledge along with mutual information (MI)-based feature weighting, whereby the feature weighting process is weakly supervised, but clustering is unsupervised.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      The EKWKM was benchmarked using a clinical dataset (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a0541, 41 features) and a large-scale population study (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a0120,000). A grid search was performed for\n                      <jats:italic>K<\/jats:italic>\n                      \u2208 {2, 3, 4} and\n                      <jats:italic>\u03b1<\/jats:italic>\n                      \u00a0=\u00a00.1\u20130.9. The silhouette score (95% bootstrap confidence interval), the adjusted Rand index (ARI), and the Davies\u2013Bouldin (DB) index were used to quantify cluster quality. Biomarker importance was measured using the Kruskal\u2013Wallis test with Benjamini\u2013Hochberg correction and\n                      <jats:italic>\u03f5<\/jats:italic>\n                      <jats:sup>2<\/jats:sup>\n                      effect size.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The most effective clinical model (\n                      <jats:italic>K<\/jats:italic>\n                      \u00a0=\u00a02,\n                      <jats:italic>\u03b1<\/jats:italic>\n                      \u00a0=\u00a00.9) achieved a silhouette score of 0.3606. The model also achieved ARI\u00a0=\u00a00.9780 \u00b1 0.0362 and DB\u00a0=\u00a01.1480 and identified 13 significant biomarkers. The population survey acted as a negative control dataset [random forest area under the curve (AUC) = 0.4974 \u00b1 0.0096] and demonstrated no phenotypic signal. Cross-dataset transfer decreased Cram\u00e9r\u2019s\n                      <jats:italic>V<\/jats:italic>\n                      from 0.497 to 0.006, which confirmed that EKWKM did not add false patterns. The main part of the clustering was due to data-driven MI, as confirmed by the ablation analysis. Expert knowledge improved clustering robustness.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>EKWKM formed clinically meaningful and Rotterdam-based PCOS phenotype clusters. In contrast, the population survey clusters formed symptom-reporting patterns and are not meaningful PCOS phenotypes.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fendo.2026.1871986","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T09:44:25Z","timestamp":1790761465000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Expert knowledge-based weighted K-means clustering for polycystic ovary syndrome phenotype identification"],"prefix":"10.3389","volume":"17","author":[{"given":"Priyanka","family":"D.","sequence":"first","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sumangali","family":"K.","sequence":"additional","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,30]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1093\/humrep\/dew218","article-title":"The prevalence and phenotypic features of polycystic ovary syndrome: a systematic review and meta-analysis","volume":"31","author":"Bozdag","year":"2016","journal-title":"Hum Reprod"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1093\/ejendo\/lvad096","article-title":"Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome","volume":"189","author":"Teede","year":"2023","journal-title":"Eur J Endocrinol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"467","DOI":"10.1210\/er.2015-1104","article-title":"The pathogenesis of polycystic ovary syndrome (PCOS): the hypothesis of PCOS as functional ovarian hyperandrogenism revisited","volume":"37","author":"Rosenfield","year":"2016","journal-title":"Endocr Rev"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-025-03984-1","article-title":"Data-driven subtypes of polycystic ovary syndrome and their association with clinical outcomes","volume":"31","author":"Gao","year":"2025","journal-title":"Nat Med"},{"key":"B5","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1016\/j.fertnstert.2003.10.004","article-title":"Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome","volume":"81","year":"2004","journal-title":"Fertility Sterility"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2023.1106625","article-title":"Application of machine learning and artificial intelligence in the diagnosis and classification of polycystic ovarian syndrome: a systematic review","volume":"14","author":"Barrera","year":"2023","journal-title":"Front Endocrinol"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.3389\/fdgth.2023.1142822","article-title":"Interpretable clinical phenotypes among patients hospitalized with COVID-19 using cluster analysis","volume":"5","author":"Yamga","year":"2023","journal-title":"Front Digital Health"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1210\/clinem\/dgae298","article-title":"Clustering identifies subtypes with different phenotypic characteristics in women with polycystic ovary syndrome","volume":"109","author":"van der Ham","year":"2024","journal-title":"J Clin Endocrinol Metab"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pmed.1003132","article-title":"Distinct subtypes of polycystic ovary syndrome with novel genetic associations: an unsupervised, phenotypic clustering analysis","volume":"17","author":"Dapas","year":"2020","journal-title":"PLoS Med"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1007\/s10115-025-02388-4","article-title":"The significance of Kappa and F-score in clustering ensemble: a comprehensive analysis","volume":"67","author":"Yan","year":"2025","journal-title":"Knowl Inf Syst"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1002\/sam.70061","article-title":"Benchmarking of clustering validity measures revisited","volume":"19","author":"Simpson","year":"2026","journal-title":"Stat Anal Data Min"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1007\/s11547-025-02032-9","article-title":"Artificial intelligence in polycystic ovarian syndrome management: past, present, and future","volume":"130","author":"Wang","year":"2025","journal-title":"La Radiologia Med"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2005.95","article-title":"Automated variable weighting in k-means type clustering","volume":"27","author":"Huang","year":"2005","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2601.17977","article-title":"Domain-expert-guided hybrid mixture-of-experts for medical AI: integrating data-driven learning with clinical priors","author":"Gu","year":"2026","journal-title":"arXiv preprint arXiv:2601.17977"},{"key":"B15","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: a graphical aid to the interpretation and validation of cluster analysis","volume":"20","author":"Rousseeuw","year":"1987","journal-title":"J Comput Appl Math"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.1979.4766909","article-title":"A cluster separation measure","author":"Davies","year":"1979","journal-title":"EEE Trans Pattern Anal Mach Intell"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1198\/016214502760047131","article-title":"Model-based clustering, discriminant analysis, and density estimation","volume":"97","author":"Fraley","year":"2002","journal-title":"J Am Stat Assoc"},{"key":"B18","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1080\/01621459.1952.10483441","article-title":"Use of ranks in one-criterion variance analysis","volume":"47","author":"Kruskal","year":"1952","journal-title":"J Am Stat Assoc"},{"key":"B19","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","article-title":"Controlling the false discovery rate: a practical and powerful approach to multiple testing","volume":"57","author":"Benjamini","year":"1995","journal-title":"J R Stat Soc"},{"key":"B20","article-title":"The need to report effect size estimates revisited. An overview of some recommended measures of effect size","volume":"21","author":"Tomczak","year":"2014","journal-title":"Trends Sport Sci."},{"key":"B21","doi-asserted-by":"publisher","first-page":"42649","DOI":"10.1038\/s41598-025-26718-2","article-title":"Insulin resistance in polycystic ovary syndrome phenotypes and the vicious cycle model in its etiology","volume":"15","author":"Szkodziak","year":"2025","journal-title":"Sci Rep"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-013-1368-0","article-title":"A review of feature selection methods based on mutual information","volume":"24","author":"Vergara","year":"2014","journal-title":"Neural Comput Appl"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.119532","article-title":"Early identification of PCOS with commonly known diseases: obesity, diabetes, high blood pressure and heart disease using machine learning techniques","volume":"217","author":"Aggarwal","year":"2023","journal-title":"Expert Syst Appl"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.3390\/app12199467","article-title":"Clustering analysis for classifying student academic performance in higher education","volume":"12","author":"Mohamed Nafuri","year":"2022","journal-title":"Appl Sci"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-30176-1","article-title":"Enhancing classification accuracy in medical datasets using a hybrid distance and cluster refinement-based K-means clustering method","author":"Al-Khamees","year":"2026","journal-title":"Sci Rep"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.3389\/fbinf.2022.927312","article-title":"A review of feature selection methods for machine learning-based disease risk prediction","volume":"2","author":"Pudjihartono","year":"2022","journal-title":"Front Bioinf"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.3309","article-title":"The silhouette coefficient and the davies-bouldin index are more informative than dunn index, calinski-harabasz index, shannon entropy, and gap statistic for unsupervised clustering internal evaluation of two convex clusters","volume":"11","author":"Chicco","year":"2025","journal-title":"PeerJ Comput Sci"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.3389\/fphys.2025.1435036","article-title":"A stacked learning framework for accurate classification of polycystic ovary syndrome with advanced data balancing and feature selection techniques","volume":"16","author":"Emara","year":"2025","journal-title":"Front Physiol"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.3389\/fcell.2024.1358755","article-title":"PCOS stratification for precision diagnostics and treatment","volume":"12","author":"Joshi","year":"2024","journal-title":"Front Cell Dev Biol"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.3934\/math.2024402","article-title":"Breaking new ground in cardiovascular heart disease Diagnosis K-RFC: An integrated learning approach with K-means clustering and Random Forest classifier","volume":"9","author":"Osman","year":"2024","journal-title":"AIMS Math"}],"container-title":["Frontiers in Endocrinology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1871986\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T09:44:26Z","timestamp":1790761466000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1871986\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,30]]},"references-count":30,"alternative-id":["10.3389\/fendo.2026.1871986"],"URL":"https:\/\/doi.org\/10.3389\/fendo.2026.1871986","relation":{},"ISSN":["1664-2392"],"issn-type":[{"value":"1664-2392","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,30]]},"article-number":"1871986"}}