Machine-Aided Self-diagnostic Prediction Models for Polycystic Ovary Syndrome: Observational Study
Open accessPublished version of a preprint
The DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Abstract
Background: Artificial intelligence and digital health care have substantially advanced to improve and enhance medical diagnosis and treatment during the prolonged period of the COVID-19 global pandemic. In this study, we discuss the development of prediction models for the self-diagnosis of polycystic ovary syndrome (PCOS) using machine learning techniques. Objective: We aim to develop self-diagnostic prediction models for PCOS in potential patients and clinical providers. For potential patients, the prediction is based only on noninvasive measures such as anthropomorphic measures, symptoms, age, and other lifestyle factors so that the proposed prediction tool can be conveniently used without any laboratory or ultrasound test results. For clinical providers who can access patients’ medical test results, prediction models using all predictor variables can be adopted to help health providers diagnose patients with PCOS. We compare both prediction models using various error metrics. We call the former model the patient model and the latter, the provider model throughout this paper. Methods: In this retrospective study, a publicly available data set of 541 women’s health information collected from 10 different hospitals in Kerala, India, including PCOS status, was acquired and used for analysis. We adopted the CatBoost method for classification, K-fold cross-validation for estimating the performance of models, and SHAP (Shapley Additive Explanations) values to explain the importance of each variable. In our subgroup study, we used k-means clustering and Principal Component Analysis to split the data set into 2 distinct BMI subgroups and compared the prediction results as well as the feature importance between the 2 subgroups. Results: We achieved 81% to 82.5% prediction accuracy of PCOS status without any invasive measures in the patient models and achieved 87.5% to 90.1% prediction accuracy using both noninvasive and invasive predictor variables in the provider models. Among noninvasive measures, variables including acanthosis nigricans, acne, hirsutism, irregular menstrual cycle, length of menstrual cycle, weight gain, fast food consumption, and age were more important in the models. In medical test results, the numbers of follicles in the right and left ovaries and anti-Müllerian hormone were ranked highly in feature importance. We also reported more detailed results in a subgroup study. Conclusions: The proposed prediction models are ultimately expected to serve as a convenient digital platform with which users can acquire pre- or self-diagnosis and counsel for the risk of PCOS, with or without obtaining medical test results. It will enable women to conveniently access the platform at home without delay before they seek further medical care. Clinical providers can also use the proposed prediction tool to help diagnose PCOS in women.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures


















Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Machine learning using health care records, patients, and health care providers resulted in 80%‐82.5% accuracy [49]”
From Availability and Use of Digital Technology Among Women With Polycystic Ovary Syndrome: Scoping Review (2025) · cites this paper as [49]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“In 9% of studies (2/34), apps that included machine learning were developed and tested to track menstrual characteristics [49] and predict the likelihood of a PCOS diagnosis [35]”
From Availability and Use of Digital Technology Among Women With Polycystic Ovary Syndrome: Scoping Review (2025) · cites this paper as [49]
The cited source says what the claim attributes to it.
In the citing paper
“Nine included studies mention the use of machine learning [63,64,65,66,67,76,87,88,89] (Supplementary File S3).”
From Polycystic Ovary Syndrome and the Internet of Things: A Scoping Review (2024) · cites this paper as [89]
The cited source says what the claim attributes to it.
In the citing paper
“Most of them were published in 2023 (n = 5) [63,64,65,76,87], while two were published in 2022 [88,89], and one each in 2021 [66] and 2020 [67].”
From Polycystic Ovary Syndrome and the Internet of Things: A Scoping Review (2024) · cites this paper as [89]
The cited source says what the claim attributes to it.
In the citing paper
“The majority of these studies originated from India (n = 5) [63,64,65,66,88], while three originated from the USA (n = 3) [67,87,89], and one from New Zealand [76].”
From Polycystic Ovary Syndrome and the Internet of Things: A Scoping Review (2024) · cites this paper as [89]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
An Introduction to Statistical Learning
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013) · Springer Texts in Statistics · doi:10.1007/978-1-4614-7138-7 · Cited by 12,106
Principal component analysis: A review and recent developments
Jolliffe, I. T., & Cadima, J. (2016) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2015.0202 · Cited by 10,117 · cited 1×
“It increases interpretability to a certain extent and minimizes data loss by combining correlated variables together to create a new set of uncorrelated yet more representative variables [25,26].”
BackgroundThe Rotterdam ESHRE/ASRM-sponsored PCOS consensus workshop group (2004) · Human Reproduction · doi:10.1093/humrep/deh098 · Cited by 6,138
Polycystic Ovary Syndrome: Definition, Aetiology, Diagnosis and Treatment
Escobar-Morreale, H. F. (2018) · Nature Reviews Endocrinology · doi:10.1038/nrendo.2018.24 · Cited by 2,083 · cited 2×
“The prevalence of PCOS ranges from 6% to 20%, depending on the population and the diagnostic criteria reported in previous studies [4-6].”
BackgroundTelehealth for global emergencies: Implications for coronavirus disease 2019 (COVID-19)
Smith, A. C., Thomas, E., Snoswell, C. L., Haydon, H., Mehrotra, A., Clemensen, J., & Caffery, L. J. (2020) · Journal of Telemedicine and Telecare · doi:10.1177/1357633x20916567 · Cited by 1,828 · cited 1×
“Telehealth using remote technologies between medical providers and patients is another emerging trend during the pandemic, and many conditions are diagnosed and managed through telehealth, including polycystic ovary syndrome (PCOS) [2,3].”
BackgroundTeede, H. J., Misso, M. L., Costello, M. F., Dokras, A., Laven, J., Moran, L., Piltonen, T., Norman, R. J., & the International PCOS Network (2018) · Clinical Endocrinology · doi:10.1111/cen.13795 · Cited by 1,823
Digital technology and COVID-19
Ting, D. S. W., Carin, L., Dzau, V., & Wong, T. Y. (2020) · Nature Medicine · doi:10.1038/s41591-020-0824-5 · Cited by 1,413 · cited 1×
“Background There has been substantial advancement in artificial intelligence and machine learning technologies in health care owing to the prolonged COVID-19 pandemic, resulting in improvements and enhancements in medical diagnosis and treatment that were previously impossible or unavailable [1].”
BackgroundSyakur, M. A., Khotimah, B. K., Rochman, E. M. S., & Satoto, B. D. (2018) · IOP Conference Series: Materials Science and Engineering · doi:10.1088/1757-899x/336/1/012017 · Cited by 1,181
Deep learning-based multi-omics integration robustly predicts survival in liver cancer
Chaudhary, K., Poirion, O. B., Lu, L., & Garmire, L. X. (2018) · Clinical Cancer Research · doi:10.1158/1078-0432.ccr-17-0853 · Cited by 1,091 · cited 1×
“Objectives Machine learning and deep learning techniques have been widely used to analyze health data and improve diagnostic accuracy and precision, disease treatment, and prevention [16,17].”
MethodDiagnosis of Polycystic Ovarian Syndrome: The Rotterdam Criteria Are Premature
Azziz, R. (2006) · The Journal of Clinical Endocrinology & Metabolism · doi:10.1210/jc.2005-2153 · Cited by 432
Gibson-Helm, M., Teede, H., Dunaif, A., & Dokras, A. (2016) · The Journal of Clinical Endocrinology & Metabolism · doi:10.1210/jc.2016-2963 · Cited by 428 · cited 3×
“Gibson-Helm et al [37,38] reported that many women experience a prolonged and frustrating diagnosis of PCOS because it requires multiple health care providers to JMIR Form Res 2022 | vol. 6 | iss.”
BackgroundSaber, A., Sakr, M., Abo-Seida, O. M., Keshk, A., & Chen, H. (2021) · IEEE Access · doi:10.1109/access.2021.3079204 · Cited by 419 · cited 1×
“Objectives Machine learning and deep learning techniques have been widely used to analyze health data and improve diagnostic accuracy and precision, disease treatment, and prevention [16,17].”
MethodPrevalence and Characteristics of the Polycystic Ovary Syndrome in Overweight and Obese Women
Álvarez-Blasco, F., Botella-Carretero, J. I., San Millán, J. L., & Escobar-Morreale, H. F. (2006) · Archives of Internal Medicine · doi:10.1001/archinte.166.19.2081 · Cited by 348
Dapas, M., Lin, F. T. J., Nadkarni, G. N., Sisk, R., Legro, R. S., Urbanek, M., Hayes, M. G., & Dunaif, A. (2020) · PLOS Medicine · doi:10.1371/journal.pmed.1003132 · Cited by 297
Diagnosis of Polycystic Ovarian Syndrome: In Defense of the Rotterdam Criteria
Franks, S. (2006) · The Journal of Clinical Endocrinology & Metabolism · doi:10.1210/jc.2005-2501 · Cited by 277
Women with polycystic ovary syndrome are often depressed or anxious—A case control study
Månsson, M., Holte, J., Landin-Wilhelmsen, K., Dahlgren, E., Johansson, A., & Landén, M. (2008) · Psychoneuroendocrinology · doi:10.1016/j.psyneuen.2008.06.003 · Cited by 260 · cited 1×
“In addition, there is abundant evidence that PCOS substantially contributes to women’s anxiety disorders, depression, personality, and other psychological disorders [41,42].”
BackgroundAnxiety and depression in polycystic ovary syndrome: a comprehensive investigation
Deeks, A. A., Gibson-Helm, M. E., & Teede, H. J. (2010) · Fertility and Sterility · doi:10.1016/j.fertnstert.2009.09.018 · Cited by 246 · cited 1×
“PCOS is also known to be linked to anxiety and depression [7].”
BackgroundMetabolic risk in PCOS: Phenotype and adiposity impact
Moran, L. J., Norman, R. J., & Teede, H. J. (2015) · Trends in Endocrinology & Metabolism · doi:10.1016/j.tem.2014.12.003 · Cited by 245 · cited 2×
“This might be because women with higher BMI are more likely to have other complications [14,15], causing more difficulty in predicting the status of PCOS.”
BackgroundDiagnostic Criteria for Polycystic Ovary Syndrome: Pitfalls and Controversies
Lujan, M. E., Chizen, D. R., & Pierson, R. A. (2008) · Journal of Obstetrics and Gynaecology Canada · doi:10.1016/s1701-2163(16)32915-2 · Cited by 217 · cited 1×
“However, the Rotterdam criteria have been controversial in many studies [9-11].”
BackgroundHirsutism and acne in polycystic ovary syndrome
Archer, J. S., & Chang, R. J. (2004) · Best Practice & Research Clinical Obstetrics & Gynaecology · doi:10.1016/j.bpobgyn.2004.05.007 · Cited by 202
Lean polycystic ovary syndrome (PCOS): an evidence-based practical approach
Toosy, S., Sodi, R., & Pappachan, J. M. (2018) · Journal of Diabetes & Metabolic Disorders · doi:10.1007/s40200-018-0371-5 · Cited by 154 · cited 3×
“Toosy et al [8] noted that a smaller proportion of women with lean PCOS had a normal or low BMI (≤25 kg/m2) and may or may not have symptoms, which makes diagnosis more challenging.”
BackgroundInfluentialRacial and ethnic differences in the polycystic ovary syndrome metabolic phenotype
Engmann, L., Jin, S., Sun, F., Legro, R. S., Polotsky, A. J., Hansen, K. R., Coutifaris, C., Diamond, M. P., Eisenberg, E., Zhang, H., et al. (2017) · American Journal of Obstetrics and Gynecology · doi:10.1016/j.ajog.2017.01.003 · Cited by 144
Changes in the PCOS phenotype with age
Hsu, M.-I. (2013) · Steroids · doi:10.1016/j.steroids.2013.04.005 · Cited by 122 · cited 1×
“Finally, age and diet habits were found to be important factors for PCOS [32,33].”
BackgroundTeede, H. J., Misso, M. L., Costello, M. F., Dokras, A., Laven, J., Moran, L., Piltonen, T., Norman, R. J., & International PCOS Network (2019) · Human Reproduction · doi:10.1093/humrep/dey363 · Cited by 120
Women's experiences of polycystic ovary syndrome diagnosis
Gibson-Helm, M. E., Lucas, I. M., Boyle, J. A., & Teede, H. J. (2014) · Family Practice · doi:10.1093/fampra/cmu028 · Cited by 112 · cited 2×
“Gibson-Helm et al [37,38] reported that many women experience a prolonged and frustrating diagnosis of PCOS because it requires multiple health care providers to JMIR Form Res 2022 | vol. 6 | iss.”
BackgroundDiet and nutrition in polycystic ovary syndrome (PCOS): Pointers for nutritional management
Farshchi, H., Rane, A., Love, A., & Kennedy, R. L. (2007) · Journal of Obstetrics and Gynaecology · doi:10.1080/01443610701667338 · Cited by 87 · cited 1×
“Finally, age and diet habits were found to be important factors for PCOS [32,33].”
BackgroundInfluence of ethnicity on different aspects of polycystic ovary syndrome: a systematic review
Sendur, S. N., & Yildiz, B. O. (2021) · Reproductive BioMedicine Online · doi:10.1016/j.rbmo.2020.12.006 · Cited by 79 · cited 1×
“In previous studies, women experienced variations in PCOS symptoms depending on culture and ethnicity [38-40].”
BackgroundCatBoost Ensemble Approach for Diabetes Risk Prediction at Early Stages
Kumar, P. S., Anisha Kumari, K., Mohapatra, S., Naik, B., Nayak, J., & Mishra, M. (2021) · Odisha International Conference on Electrical Power Engineering, Communication and Computing Technology (ODICON) · doi:10.1109/odicon50556.2021.9428943 · Cited by 77 · cited 1×
“The CatBoost method is one of the newer gradient boosting decision tree models, and it was recently used in diabetes prediction in the study by Kumar et al [20].”
MethodKar, S. (2013) · Journal of Human Reproductive Sciences · doi:10.4103/0974-1208.121422 · Cited by 74 · cited 2×
“Kar et al [24] characterized and classified phenotypes of PCOS in a large cohort of women into subgroups and compared the data of various metabolic complications of these phenotypes.”
BackgroundResultDebates regarding lean patients with polycystic ovary syndrome: a narrative review
Dawood, A., & Goyal, M. (2017) · Journal of Human Reproductive Sciences · doi:10.4103/jhrs.jhrs_77_17 · Cited by 72
Diagnosis of Polycystic Ovary Syndrome: AMH in combination with clinical symptoms
Sahmay, S., Aydin, Y., Oncul, M., & Senturk, L. M. (2014) · Journal of Assisted Reproduction and Genetics · doi:10.1007/s10815-013-0149-0 · Cited by 71 · cited 1×
“Symptoms and AMH have also been reported as important markers for the diagnosis of PCOS [31].”
BackgroundAboueid, S., Liu, R. H., Desta, B. N., Chaurasia, A., & Ebrahim, S. (2019) · JMIR Medical Informatics · doi:10.2196/13445 · Cited by 71 · cited 1×
“This work is well aligned with emerging artificial intelligence and digital health care [21].”
BackgroundPolycystic Ovary Syndrome and Weight Management
Moran, L. J., Lombard, C. B., Lim, S., Noakes, M., & Teede, H. J. (2010) · Women's Health · doi:10.2217/whe.09.89 · Cited by 69 · cited 1×
“In other studies, it has been noted that PCOS is closely associated with obesity and is more prevalent among overweight or obese women than in the general population of women of reproductive age [13-15].”
BackgroundMaugeri, A., Barchitta, M., Basile, G., & Agodi, A. (2021) · Scientific Reports · doi:10.1038/s41598-021-86703-3 · Cited by 64 · cited 1×
“In particular, feature selection, clustering algorithms, and classification have often been adopted for subgroup studies [18,19].”
MethodÁlvarez, J. D., Matias-Guiu, J. A., Cabrera-Martín, M. N., Risco-Martín, J. L., & Ayala, J. L. (2019) · BMC Bioinformatics · doi:10.1186/s12859-019-3027-7 · Cited by 59 · cited 1×
“In particular, feature selection, clustering algorithms, and classification have often been adopted for subgroup studies [18,19].”
MethodPsychosocial Factors Associated with Polycystic Ovary Syndrome: a Case Control Study
Sayyah-Melli, M., Alizadeh, M., Pourafkary, N., Ouladsahebmadarek, E., Jafari-Shobeiri, M., Abbassi, J., Kazemi-Shishvan, M. A., & Sedaghat, K. (2015) · Journal of Caring Sciences · doi:10.15171/jcs.2015.023 · Cited by 43 · cited 1×
“In addition, there is abundant evidence that PCOS substantially contributes to women’s anxiety disorders, depression, personality, and other psychological disorders [41,42].”
BackgroundCharacteristics of heart rate variability in women with polycystic ovary syndrome
Ji, H. R., Woo, H. L., Park, Y.-J., Hwang, D. S., Lee, J., Lee, C. H., Jang, J. B., & Park, K. S. (2018) · Medicine · doi:10.1097/md.0000000000012510 · Cited by 20 · cited 1×
“Regarding pulse rate and respiratory rate being important in subgroup 1, heart rate variability in normal weight women with PCOS has been subject to debate in the literature [35,36].”
BackgroundHeart rate variability in normal-weight patients with polycystic ovary syndrome
Kılıt, C., & Kilit, T. P. (2017) · The Anatolian Journal of Cardiology · doi:10.14744/anatoljcardiol.2017.7473 · Cited by 16 · cited 1×
“Regarding pulse rate and respiratory rate being important in subgroup 1, heart rate variability in normal weight women with PCOS has been subject to debate in the literature [35,36].”
Background[Body composition and polycystic ovary syndrome].
Zabulienė, L., & Tutkuvienė, J. (2010) · PubMed · Cited by 15
Polycystic Ovary Syndrome (PCOS)
Hechtman, L. (2020) · Elsevier eBooks · doi:10.1016/b978-0-323-43044-9.00209-0 · Cited by 7
Polycystic Ovary Syndrome (PCOS)
Donesky, B. W. (2004) · Elsevier eBooks · doi:10.1016/b0-12-475570-4/01037-4 · Cited by 6
Kūno sandara ir policistinių kiaušidžių sindromas
Zabulienė, L., & Tutkuvienė, J. (2010) · Medicina · doi:10.3390/medicina46020021 · Cited by 5
Research works printed but not linked: 1
- 10.5555/3295222.3295230 · doi:10.5555/3295222.3295230
Cited by
A Distinctive Explainable Machine Learning Framework for Detection of Polycystic Ovary Syndrome
Khanna, V. V., Chadaga, K., Sampathila, N., Prabhu, S., Bhandage, V., & Hegde, G. K. (2023) · Applied System Innovation · doi:10.3390/asi6020032 · Cited by 129
“Zigarelli et al. [26] used the same dataset, adopting the CatBoost classifier with K-fold validation evaluating the classifier performance, achieving 82.5% for the invasive methods and 90.1% accuracy for the non-invasive clinical parameters.”
InfluentialSPOSDS: A smart Polycystic Ovary Syndrome diagnostic system using machine learning
Tiwari, S., Kane, L., Koundal, D., Jain, A., Alhudhaif, A., Polat, K., Zaguia, A., Alenezi, F. S., & Althubiti, S. A. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.117592 · Cited by 148
Stafie, C. S., Sufaru, I.-G., Ghiciuc, C. M., Stafie, I.-I., Sufaru, E.-C., Solomon, S. M., & Hancianu, M. (2023) · Diagnostics · doi:10.3390/diagnostics13121995 · Cited by 93
“[140] developed a self-diagnostic prediction model for PCOS, based on different variables, such as hirsutism, acne, an irregular menstrual cycle, the length of the menstrual cycle, and weight gain.”
BackgroundEmpowering early detection: A web-based machine learning approach for PCOS prediction
Rahman, M. M., Islam, A., Islam, F., Zaman, M., Islam, M. R., Sakib, M. S. A., & Babu, H. M. H. (2024) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2024.101500 · Cited by 76
Optimized Machine Learning for the Early Detection of Polycystic Ovary Syndrome in Women
Panjwani, B., Yadav, J., Mohan, V., Agarwal, N., & Agarwal, S. (2025) · Sensors · doi:10.3390/s25041166 · Cited by 77
“…is that it provides PCOS prediction using a dataset of symptoms and test results from invasive tests and ultrasound, an issue that Zigarelli et al. [55] tried to address, presenting a self-diagnostic Sensors 2025 , 25 , 1166 4 of 30 platform with an RF model that considers symptomatic and clinical…”
Metadata
- Authors
- Zigarelli, A., Jia, Z., & Lee, H.
- Affiliations
- University of Massachusetts Amherst (US)
- Corresponding author
- Lee, Hyunsun
- Venue
- JMIR Formative Research
- Publisher
- JMIR Publications Inc.
- Published
- 2022-03-15
- Volume
- 6
- Issue
- 3
- Pages
- e29967
- ISSN
- 2561-326X (ISSN-L)
- PMC
- PMC8965679
- OpenAlex
- W4221092010
- Cited by
- 74
- References
- 42
- Citations by year
- 2026: 22 · 2025: 22 · 2024: 16 · 2023: 11 · 2022: 3
- FWCI
- 9.75
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR Formative Research (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Health Sciences › Medicine › Reproductive Medicine › Ovarian function and disorders
- Physical Sciences › Computer Science › Information Systems › Data Mining and Machine Learning Applications
- Keywords
- polycystic ovary syndrome, prediction model, CatBoost, SHAP feature importance, anti-Müllerian hormone, ovarian follicle count, acanthosis nigricans, hirsutism, menstrual irregularity, digital health platforms, women's health, risk prediction, K-means clustering, principal component analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.