- Research
- Open access
- Published:
Nutritional management recommendation systems in polycystic ovary syndrome: a systematic review
BMC Women's Health volume 24, Article number: 234 (2024)
Abstract
Background
People with polycystic ovary syndrome suffer from many symptoms and are at risk of developing diseases such as hypertension and diabetes in the future. Therefore, the importance of self-care doubles. It is mainly to modify the lifestyle, especially following the principles of healthy eating. The purpose of this study is to review artificial intelligence-based systems for providing management recommendations, especially food recommendations.
Materials and methods
This study started by searching three databases: PubMed, Scopus, and Web of Science, from inception until 6 June 2023. The result was the retrieval of 15,064 articles. First, we removed duplicate studies. After the title and abstract screening, 119 articles remained. Finally, after reviewing the full text of the articles and considering the inclusion and exclusion criteria, 20 studies were selected for the study. To assess the quality of articles, we used criteria proposed by Malhotra, Wen, and Kitchenham. Out of the total number of included studies, seventeen studies were high quality, while three studies were moderate quality.
Results
Most studies were conducted in India in 2021. Out of all the studies, diagnostic recommendation systems were the most frequently researched, accounting for 86% of the total. Precision, sensitivity, specificity, and accuracy were more common than other performance metrics. The most significant challenge or limitation encountered in these studies was the small sample size.
Conclusion
Recommender systems based on artificial intelligence can help in fields such as prediction, diagnosis, and management of polycystic ovary syndrome. Therefore, since there are no nutritional recommendation systems for these patients in Iran, this study can serve as a starting point for such research.
Introduction
Polycystic ovary syndrome affects 8 to 13% of women of reproductive age worldwide, making it the most common endocrine problem. This condition can cause menstrual disorders, lack of ovulation, obesity, acne, hirsutism, hair loss, and baldness. Long-term complications include endometrial cancer, infertility, insulin resistance, type 2 diabetes, high blood pressure, heart disease, depression, and stress. Short-term consequences and complications may also arise [1]. Early diagnosis of the disease is very important and can reduce the duration of the disease and the mortality rate. Studies show that in many diseases, early diagnosis is difficult for health care providers. On the other hand, patients do not have the information related to self-management and do not have the necessary knowledge to obtain this information. Social withdrawal may become more prevalent as physical symptoms like acne, hair loss, and depression manifest [2,3,4]. Following diagnosis, implementing lifestyle modifications to address symptoms like elevated cholesterol and insulin resistance is recognized as an innovative therapeutic approach. Given the prevalent overweight status among women with this syndrome, the significance of adhering to a nutritious diet and engaging in physical activity has been underscored. Consequently, the imperative of devising strategies to promote adherence to healthy dietary habits and facilitate weight loss among affected individuals is deemed essential and inevitable [1].
The use of information technology greatly facilitates the prevention, diagnosis and treatment of chronic diseases and increases their accuracy. Among these technologies, it can be mentioned decision support systems, mobile-based applications, virtual reality, augmented reality, and intelligent decision-making systems. These technologies largely solve the challenge of accessing data and evidence-based information for both patients and medical providers [5,6,7,8,9,10].
In the healthcare domain, recommender systems represent a practical technology enabling self-care through tailored recommendations. Ultrasound imaging can support early disease detection, alleviating healthcare provider workload and expediting diagnosis. This strategy not only conserves resources and reduces expenses but also leverages mobile phone platforms to enhance awareness, disseminate evidence-based knowledge, and foster beneficial shifts in individual behaviors and habits. By harnessing artificial intelligence, a recommender system can enhance dietary practices among patients with PCOS, leveraging the widespread adoption of smartphones to improve access to medical professionals and reliable information [2, 4].
The study conducted by Jan et al. in India in 2022 analyzed six AI models for diagnosing PCOS. They compared these models based on the number of ultrasound images, segmentation, and classification methods. The evaluation highlighted that the Bayesian classifier achieved the highest accuracy of 93.93%. This study underscores the significant potential of AI in diagnosing PCOS and recommends further research to implement this technology effectively [11].
Boyle et al. conducted a study in Australia in 2018 to assess the need for assistance among individuals with PCOS and evaluate mobile health applications in this area. The results show that 98% of participants owned smartphones, 72% had previously used an application for self-care, and 91% expressed willingness to use a PCOS-specific app if available for managing this syndrome. Accurate, evidence-based information was deemed essential in this study, and all the assessed applications met the required quality standards [12].
The research conducted by Portugal et al. in Canada in 2017 focused on the use of machine learning techniques in recommender systems. The study aimed to identify associated issues and assist researchers in implementing these systems more effectively. The findings of the study highlighted various machine learning techniques and their applications, as well as primary and alternative performance criteria [13].
In 2019, Abhari et al. conducted a study in Iran to assess the characteristics of nutritional recommender systems. The study revealed that if these systems are properly designed, implemented, and evaluated, they can serve as effective tools to improve nutrition and promote a healthy lifestyle [14].
The study conducted in India in 2021 by Kaur and colleagues aimed to develop a method for classifying food images to track patients’ meals and provide guidance to nutritionists on recommended tactics and image classification. The research focused on utilizing deep learning approaches, particularly convolutional neural networks (CNNs), to classify Indian food images accurately [15].
After conducting research, we found no systematic review that assesses nutritional recommendations for individuals with PCOS. Given the significance of utilizing companion health and AI in managing chronic conditions like PCOS, this study aims to explore the effects and uses of AI-based systems for PCOS.
Motivation
The advancement of AI in healthcare has made it necessary to use AI-powered recommender systems for predicting, diagnosing, treating, and managing chronic ailments like PCOS. There are several reasons why this is important. Ultrasound images are necessary to diagnose PCOS. However, inaccuracies in counting follicles, high diagnostic test costs in developing countries, time-consuming tests, doctor workload, and diagnostic errors can lead to inaccurate diagnoses. AI techniques can automatically diagnose diseases using ultrasound images, overcoming challenges [16,17,18]. Additionally, since this disease is intricate and has no definitive treatment, the current approach involves a combination of medication and lifestyle changes for disease management. Hence, utilizing recommender systems or self-care systems that prioritize healthy nutrition could be beneficial in mitigating symptoms and lessening the likelihood of associated mental health issues [19, 20]. Besides disease prediction and probability estimation, recommender systems can detect suspicious cases based on AI and take action to prevent disease occurrence or early detection [21].
Contribution
The article aims to explore the use of artificial recommender systems in polycystic ovary syndrome research. We also examine the challenges and limitations of using these systems and algorithms. This article attracts researchers to conduct studies in the field of recommender systems. The sections of this article are as follows:
-
Review of studies from the perspective of publication.
-
Review studies of characteristics.
-
Examining the limitations and challenges of designing recommender systems.
Materials and methods
Study design
The current systematic study was designed and implemented based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.
Data sources
In this study, we were six researchers. The first one determined the search strategy and performed the search in PubMed, Scopus, and Web of Science databases from inception until 6 June 2023. To conduct a search for relevant articles, we utilized a combination of keywords from the article abstracts, including “polycystic ovary syndrome”, “recommender system”, “application program”, “artificial intelligence” and “nutritional program”. We also incorporated Medical Subject Headings (Mesh) into our search strategy. For the complete search strategy, please refer to the supplementary file provided (Table S1-S3).
Selection criteria
One of the researchers determined the inclusion and exclusion criteria based on similar studies, and finally criteria were approved by the supervisor’s opinion. Inclusion and exclusion criteria are as follows:
Inclusion criteria:
-
1)
Studies about recommender systems
-
2)
Articles related to PCOS
-
3)
Articled related to diet management in PCOS
-
4)
Studies that developed an AI system or application
Exclusion criteria:
-
1)
Review articles, meta-analysis, conference abstracts, letters to the editor, book chapter.
-
2)
Articles that are unrelated to the goals of the current research.
-
3)
Articles whose full text is written in non-English language.
-
4)
Articles whose full text is not available for data extraction.
Study selection
In this stage, one of the researchers entered all the articles retrieved from the three databases (PubMed, Web of Science, and Scopus) into the Endnote X9 (Thomson Reuters, Toronto, Ontario, Canada) software. Another researcher removed duplicates. Two of researchers separately, checked the title and abstract of the articles. The items that did not match the inclusion and exclusion criteria were excluded from the study. Finally, by studying and examining the full text of the remaining articles, he selected the articles related to the topic as the final articles. In cases where the two researchers had differing opinions, the supervisor provided the final decision.
Data extraction
A researcher created an Excel form with the guidance of our supervisor. The research team reviewed articles and extracted necessary data elements. The Excel form included data such as the number of articles, publication year and country.
Quality assessment
Two independent reviewers assessed the quality of studies using the Newcastle-Ottawa quality assessment criteria proposed by Malhotra [17], Wen et al. [16], and Kitchenham et al. [18]. The quality assessment criteria consist of eleven questions: Q1) Are the aims of the study clearly defined?; Q2) Are all study questions answered?; Q3) Are the variables used in the study clearly stated?; Q4) Are AI techniques, such as machine learning, clearly defined?; Q5) Is the data set size appropriate?; Q6) Is the data collection method clearly stated?; Q7) Is the study methodology repeatable?; Q8) Are the results and findings clearly presented?; Q9) Are the performance measures used to assess the model(s) clearly stated?; Q10) Are the limitations of the study stated?; Q11) Does the research have value for the academic or industry community? The questions were ranked based on three values: “Yes = 2”, “Partial = 1”, or “No = 0”. Each study could obtain a maximum score of 22 and a minimum score of 0. Criteria used to rank the quality assessment of each study include: i) ≤ 49% = Low quality; ii) 50% and 69% = Moderate quality; iii) above 70% = High quality.
Results
Study selection
Figure 1 shows the process of searching and selecting articles based on the PRISMA flowchart. We found a total of 15,064 articles by searching in PubMed, Web of Science, and Scopus databases. After removing duplicates (n = 6537), We took three steps: (1) screening the article titles, (2) reviewing the article abstracts, and (3) reviewing the full text of the articles and extracting the data by the second group. Based on the predetermined criteria, we eliminated 8,408 studies during the one and two-stage process. In the third stage, from 119 articles unrelated studies (n = 86), articles with unavailable full text (n = 8), review studies (n = 4), and book chapters (n = 1) were excluded. Finally, 20 articles entered the third stage, i.e., a review of the full text of the articles. The research team extracted required data elements such as publication year, country, journal/conference, purpose, study design, sample size, sample age range, results, tools, challenges/limitations, Relevance to the study and system target. We recorded the extracted data in an Excel (Table 1).
Publication analysis
Distribution of studies by year
The studies were conducted from 1997 to 2023. Figure 2 shows the results. 2021 (n = 5) has the most frequency. The second frequency was related to 2018 and 2022. (n = 3).
Distribution of studies by country
The reviewed studies were conducted in eight different countries. The frequency of these studies in each country is depicted in Figs. 2 and 3. India had the highest number of studies (n = 9), followed by Indonesia (n = 3), the United States of America, and China (each with 2), ranking second and third, respectively.
Distribution of articles based on journal/conference name, publisher and impact factor
The articles were appeared in 11 different journals and seven conferences. The journal “Frontiers in Endocrinology” had the highest number (n = 2), while all the other journals had only published 1 article. All conferences, except the “International Conference on Data and Information Science,” presented a paper on the topic. Tables 2 and 3 display the distribution of articles in this field.
Study specifications
Frequency of studies based on AI/application
Based on the survey, most (95%) of the studies focused on models and systems utilizing AI technology [2, 16, 21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37], while only one study resulted in the creation of an AI-based application [18].
Frequency of studies based on the type of system application: When it comes to systems and applications designed for various purposes, they can be classified into three types: prediction, diagnosis, and management. Among all the conducted studies, 60% focused on diagnosis [2, 16, 24,25,26, 29, 30, 32, 33, 35,36,37], 30% on prediction [21,22,23, 27, 31, 34], and none on management. Additionally, 10% of studies focused on both diagnosis and prediction [18, 28]. However, none of the studies examined the role of nutrition in managing PCOS.
Specifications of performance metrics for model evaluation
Table 4 displays the metrics utilized in the articles. MAE [23], RMSE [23, 33], and RRSE [23] calculate three types of errors in implemented models, so the lowest value is considered for an ideal model. MAE in mathematics is the arithmetic equivalent of absolute errors. This criterion only measures the magnitude of the error and does not give a significant indication of the direction of the error. These three criteria have been used in a study to evaluate the model’s performance [23, 33].
The Kappa statistic is a tool that evaluates the effectiveness of a model’s reliability and ensures an accurate representation of changes in collected data. It can range from − 1 to + 1. Despite being a common statistic, there is no consensus on measures of it in health studies. It has been utilized solely in one study [38].
Most studies commonly used Sensitivity [2, 18, 28,29,30, 32,33,34,35,36,37], Specificity [2, 28], and Accuracy [2, 18, 26, 28,29,30, 32,33,34,35,36,37] as criteria. These metrics measures by True Positive (TP), True Negative (TN), False Positive (FP), and False Negative (FN). F1 score or 𝐹 − 𝑀𝑒𝑎𝑠𝑢𝑟e as a harmonic average of accuracy and recall. Where there is a balance between accuracy and recall, it is a better metric [2, 18, 26, 28,29,30, 32,33,34,35,36,37].
Specifications of decision support systems for PCOS prediction
As presented in Table 1, Eight studies [18, 21,22,23, 27, 28, 31, 34] designed a model to predict disease. These studies used Decision trees [22, 31], Topology-preserving forward network [21], multi-layer perceptron [21], NFRS [23], Artificial neural network [23], Apriori algorithm [27], NB classifier method [28], LR [18, 28], KNN [18, 28, 31], CART [28], RF Classifier [18, 28], Gaussian Naive Bayes [18], Fuzzy [31], CNN [34], GCN techniques [34] and SVM [18, 28, 31].
Specifications of decision support systems for PCOS diagnosis
Automated detection models based on database: Eight studies [2, 18, 28,29,30, 32, 36, 37] have implemented automated detection models based on a database to diagnose or screen patients automatically. These studies aimed to diagnose diseases automatically by creating an AI-based model using readily available data or data from those who seek treatment at health centers. The studies all followed a similar methodology. They first collected data from healthy and sick individuals. After that, they performed pre-processing to identify parameters and characteristics. They designed the model using selected techniques and evaluated its performance using model evaluation metrics. The studies utilized various methods such as LR [2, 28,29,30], Bayesian classifier [2], DT [29, 37], SVM [18, 28,29,30, 37], CNN [32], KNN [18, 28], quadratic discriminant classifier [18, 29], RF [18, 28,29,30], CART [28, 30], Gaussian naive Bayes [18], and K-means clustering [36] to develop the automated diagnosis model.
Classification models based on images: Four studies [24,25,26, 35] have developed a diagnostic model for this disease using ultrasound image classification. Ultrasound images were used to train and test the model., and different methods such as CNN [24,25,26], and SVM [26] were used in these studies.
Follicle segmentation models: Two studies [16, 33] created a model to segment follicles in ultrasound images for automatic disease diagnosis. The model specifically diagnoses through follicle segmentation, reducing the time needed for follicle counting. The process began with the publication of images, followed by image processing to segment the follicles. After extracting features, classifiers were used to design the model. Techniques used during the pre-processing stage included histogram equalization, contrast enhancement, and the Wiener filter for noise reduction of the images. The segmentation stage utilized the Fuzzy logicis, Hybrid Intelligent, Water Drop (IWD), KNN, SVM [33] and K-means clustering [16].
Limitations and challenges mentioned in the studies
As presented in Tables 1 and 10 studies pointed out the limitations and refinements. Accordingly, the small volume of sample size and features in 4 studies, the need to conduct more studies in 3 studies, the increase in the time of automatic diagnosis by the system by using more data in one study [26], the unwillingness of patients to disclose reports and clinical data in one [19] and the need to improve accuracy using other classifiers is mentioned in one study [2].
Quality assessment of included studies
The quality assessment of the included studies is detailed in the Supplementary file (Table S4), with 17 studies rated as high quality and three as moderate qualities.
Discussion
The purpose of this study was to conduct a thorough review of recommendation systems for women with PCOS. Specifically, we focused on models or applications that utilized artificial intelligence. We collected information from various sources such as publication year, country, journal or conference, sample size, age of participants, limitations and challenges, and results. During systematic review, we found five studies that shared a similar approach [2, 11, 12, 14, 17, 19]. . We found reasons for using mobile or AI-based recommender systems in PCOS disease management. We will now delve into the study’s findings and other studies.
Abhari et al. investigated nutritional recommendation systems without considering a specific disease. In this study, we reviewed the proposed recommender systems for polycystic ovary syndrome with its various applications [14].
Based on the study results, obesity in people with PCOS, with the escalation of symptoms, increases the cost of treatment and reduces it, especially in infertility. The expenditure of lifestyle modification with the help of health and weight loss is lower than drug therapy. Modifying lifestyle and nutrition using mobile phones is considered a low-cost intervention with a lower percentage of invasion [20]. The results of a 2018 study by Jacqueline A. Boyle and colleagues in Australia showed that a quality disease management application met the needs of patients; however, none of the applications reviewed had quality [12].
As mentioned, early diagnosis of the disease in the early stages is associated with risk reduction of disease consequences. Therefore, we may need recommender systems to reduce risk reduction and time of the diagnosis and increase accuracy [32]. One study by Naila Jan and colleagues in 2023 investigated AI techniques for PCOS diagnosis. The results of this study show that early diagnosis of this disease is difficult despite different symptoms in people, so automatic detection systems can be used as an accurate solution in this field [17]. As with our results, the existence of a limited amount of data is considered as one of the obstacles to the implementation of this type of study [17, 27].
Two studies conducted by Naila Jan and colleagues in India in 2022 reported a PCOS rate of 3.7 to 22.5%, which was higher in urban than rural women [11, 17]. Based on the review, 40.9% of studies have been conducted in India, which justifies the high rate of PCOS. Also, unhealthy lifestyles, including unhealthy eating, can be a reason for most urban women to do this.
Among the models designed to predict PCOS, the best accuracy belongs to a fuzzy logic-based model with an accuracy of 98.2% [31]. In the field of PCOS diagnosis, a hybrid model based on ANN, CNN and InceptionV3 has the best performance among the designed models with an accuracy of 98.12 [35]. RF and SVM were two widely used algorithms with acceptable performance, but the performance of the CNN-based model with 97% accuracy is better than these two algorithms [32]. Based on the study by Naila Jan and colleagues, the best performance belonged to a CNN-based system with an average of 76.36% and a micro-average f1-score of 100%. KNN, ANN, and Fuzzy logic with an accuracy of 97%, 97.5%, and 97.30 were the best classification techniques among the reviewed articles [11, 17].
Another study in 2018 by Jue Xie and colleagues in Australia aimed to investigate the AskPCOS application and the steps involved in creating it. The results showed that the mentioned program is one of the best evidence-based user programs to manage PCOS disease. Due to the support of 5 common languages of the world, it can eliminate the inequality of lack of access to information in developing countries. According to the evaluation, 80% of people were satisfied. The application’s usefulness was 70%. Jacqueline A. Boyle’s study showed that evidence-based application programs positively affect the patient’s awareness [19].
Study limitations
The study has several limitations. (1) only two studies addressed the aspect of concomitant health management in PCOS. (2) there is a lack of access to several paid articles that can provide better results by reviewing them. 3, the case was a similar study in other developing countries such as Iran, which may cause problems in the results of this study in this country. Therefore, we recommend carrying out more specialized studies about recommender systems. We should use AI to modify lifestyles to nutritional patterns, such as providing a dedicated budget, creating motivation, and creating a culture among researchers.
Conclusion
We conducted a systematic review to explore the use of AI and companion health systems in managing polycystic ovary syndrome, with a focus on nutrition. Although AI has primarily been used for disease diagnosis, the positive impact of AI and companion health systems in providing nutrition-based treatment solutions is significant. Therefore, we recommend that countries, particularly those with a high number of affected individuals, prioritize policies that encourage further studies to evaluate the effectiveness of recommendation systems on the nutrition of people with polycystic ovary syndrome. We can work on the quality of life for individuals affected by this condition.
Data availability
The datasets generated and/or analysed during the current study are available in the PubMed, Scopus, and Web of Science databases.
Abbreviations
- AI:
-
Artificial Intelligence
- PCOS:
-
Polycystic ovary syndrome
- SOM:
-
Self-Organizing Map
- TPFN:
-
Topology-Preserving Forward Network
- MLP:
-
Multi-Layer Perceptron
- NFRS:
-
Neural Fuzzy Rough Set Evaluation
- ANN:
-
Artificial Neural Network
- KNN:
-
K-Nearest Neighbor
- CART:
-
Classification and Regression Trees
- SVM:
-
Support Vector Machine
- LR:
-
Logistic Regression
- RF:
-
Random Forest
- SFFS:
-
Sequential Forward Floating Selection
- DT:
-
Decision Tree
- RCT:
-
Clinical Randomized Trial
- NB:
-
Naïve Bayes
- GCN:
-
Graph convolutional network
- CNN:
-
Convolutional Neural Networks
- LDR:
-
Linear Decision Analysis
References
Akre S, Sharma K, Chakole S, Wanjari MB. Recent advances in the management of polycystic ovary syndrome: a review article. Cureus. 2022;14(8):e27689.
Mehrotra P, Chatterjee J, Chakraborty C, Ghoshdastidar B, Ghoshdastidar S, editors. Automated screening of Polycystic Ovary Syndrome using machine learning techniques. 2011 Annual IEEE India Conference; 2011 16–18 Dec. 2011.
Nilashi M, Ahmadi H, Shahmoradi L, Mardani A, Ibrahim O, Yadegaridehkordi EJJSC et al. Knowledge Discovery and diseases Prediction: a comparative study of machine. Learn Techniques. 2017;4(5).
Alotaibi M, Alsinan A, editors. A mobile Polycystic ovarian syndrome management and awareness system for Gulf countries: System architecture. 2016 SAI Computing Conference (SAI); 2016 13–15 July 2016.
Shahmoradi L, Safdari R, Ahmadi H, Zahmatkeshan M. Clinical decision support systems-based interventions to improve medication outcomes: a systematic literature review on features and effects. Med J Islamic Repub Iran. 2021;35:27.
Shahmoradi L, Rezayi SJJoN. Rehabilitation. Cognitive rehabilitation in people with autism spectrum disorder: a systematic review of emerging virtual reality-based approaches. 2022;19(1):91.
Maghooli K, Langarizadeh M, Shahmoradi L, Habibi-Koolaee M, Jebraeily M, Bouraghi HJAIM. Differential diagnosis of erythmato-squamous diseases using classification and regression tree. 2016;24(5):338.
Gopal G, Suter-Crazzolara C, Toldo L, Eberhardt WJCC, Medicine L. Digital transformation in healthcare–architectures of present and future information technologies. 2019;57(3):328–35.
Nilashi M, Ahmadi N, Samad S, Shahmoradi L, Ahmadi H, Ibrahim O, et al. Disease Diagnosis Using Mach Learn Techniques: Rev Classif. 2020;7(1):19–30.
Shahmoradi L, Almasi S, Ahmadi H, Bashiri A, Azadi T, Mirbagherie A, et al. Virtual reality games for rehabilitation of upper extremities in stroke patients. J Bodyw Mov Ther. 2021;26:113–22.
Jan N, Makhdoomi A, Handa P, Goel N, editors. Machine learning approaches in medical image analysis of PCOS. 2022 International Conference on Machine Learning, Computer Systems and Security (MLCSS); 2022: IEEE.
Boyle JA, Xu R, Gilbert E, Kuczynska-Burggraf M, Tan B, Teede H, et al. Ask PCOS: identifying need to inform evidence-based App Development for Polycystic Ovary Syndrome. Semin Reprod Med. 2018;36(1):59–65.
Portugal I, Alencar P, Cowan D. The use of machine learning algorithms in recommender systems: a systematic review. Expert Syst Appl. 2018;97:205–27.
Abhari S, Safdari R, Azadbakht L, Lankarani KB, Niakan Kalhori SR, Honarvar B, et al. A systematic review of Nutrition Recommendation systems: with Focus on Technical Aspects. J Biomedical Phys Eng. 2019;9(6):591–602.
Kaur R, Kumar R, Gupta M. Food Image-based diet recommendation framework to overcome PCOS problem in women using deep convolutional neural network. Comput Electr Eng. 2022;103:108298.
Mandal A, Saha D, Sarkar M, editors. Follicle Segmentation Using K-Means Clustering from Ultrasound Image of Ovary. Proceedings of International Conference on Frontiers in Computing and Systems; 2021 2021//; Singapore: Springer Singapore.
Makhdoomi A, Jan N, Palak P, Goel N, editors. Machine learning techniques for medical images in PCOS. 2022 4th International Conference on Artificial Intelligence and Speech Technology (AIST); 2022 9–10 Dec. 2022.
Vedpathak S, Thakre V, PCOcare:. PCOS detection and prediction using machine learning algorithms. Biosci Biotechnol Res Commun. 2020;13:240–4.
Xie J, Burstein F, Garad R, Teede HJ, Boyle JA. Personalized Mobile Tool AskPCOS delivering evidence-based Quality Information about polycystic ovary syndrome. Semin Reprod Med. 2018;36(1):66–72.
Jiskoot G, Benneheij SH, Beerthuizen A, de Niet JE, de Klerk C, Timman R, et al. A three-component cognitive behavioural lifestyle program for preconceptional weight-loss in women with polycystic ovary syndrome (PCOS): a protocol for a randomized controlled trial. Reproductive Health. 2017;14(1):34.
Lehtinen J-C, Forsström J, Koskinen P, Penttilä T-A, Järvi T, Anttila L. Visualization of clinical data with neural networks, Case study: polycystic ovary syndrome. Int J Med Informatics. 1997;44(2):145–55.
Zhang H, Legro RS, Zhang J, Zhang L, Chen X, Huang H, et al. Decision trees for identifying predictors of treatment effectiveness in clinical trials and its application to ovulation in a study of women with polycystic ovary syndrome. Hum Reprod (Oxford England). 2010;25(10):2612–21.
Rethinavalli S, Manimekalai M. A novel hybrid framework for risk severity of polycystic ovarian syndrome. 2016;9:19–27.
Cahyono B, Adiwijaya, Mubarok MS, Wisesty UN, editors. An implementation of convolutional neural network on PCO classification based on ultrasound image. 2017 5th International Conference on Information and Communication Technology (ICoIC7); 2017 17–19 May 2017.
Dewi RM, Adiwijaya, Wisesty UN, Jondri. Classification of polycystic ovary based on ultrasound images using competitive neural network. J Phys: Conf Ser. 2018;971(1):012005.
Thufailah IF, Adiwijaya, Wisesty UN, Jondri. An implementation of Elman neural network for polycystic ovary classification based on ultrasound images. J Phys: Conf Ser. 2018;971(1):012016.
Boddu V, Bollineni A, Bhargav K, Sarangi S, Chilla M. Application of the Apriori Algorithm for Prediction of polycystic ovarian syndrome (PCOS). 2018. p. 934–44.
Denny A, Raj A, Ashok A, Ram CM, George R, editors. i-HOPE: Detection And Prediction System For Polycystic Ovary Syndrome (PCOS) Using Machine Learning Techniques. TENCON 2019–2019 IEEE Region 10 Conference (TENCON); 2019 17–20 Oct. 2019.
Adla YAA, Raydan DG, Charaf MZJ, Saad RA, Nasreddine J, Diab MO, editors. Automated Detection of Polycystic Ovary Syndrome Using Machine Learning Techniques. 2021 Sixth International Conference on Advances in Biomedical Engineering (ICABME); 2021 7–9 Oct. 2021.
Hassan M, Mirza T. Comparative Analysis of Machine Learning Algorithms in diagnosis of polycystic ovarian syndrome. Int J Comput Appl. 2020;Volume 175.
Kodipalli A, Devi S. Prediction of PCOS and Mental Health using fuzzy inference and SVM. Front Public Health. 2021;9:789569.
Lv W, Song Y, Fu R, Lin X, Su Y, Jin X, et al. Deep Learning Algorithm for Automated Detection of Polycystic Ovary Syndrome using scleral images. Front Endocrinol. 2021;12:789878.
N.S N, Novel Hybrid RDA. Algorithm with Extreme Learning adaptive neuro-fuzzy inference system based follicle identification for polycystic ovary syndrome in Ultrasound images. Webology. 2021;18:594–616.
Zhang N, Wang H, Xu C, Zhang L, Zang T. DeepGP: an Integrated Deep Learning Method for Endocrine Disease Gene Prediction using. Omics Data. 2021;9.
Hosain AS, Mehedi MHK, Kabir IE, editors. Pconet: A convolutional neural network architecture to detect polycystic ovary syndrome (pcos) from ovarian ultrasound images. 2022 International Conference on Engineering and Emerging Technologies (ICEET); 2022: IEEE.
Zigarelli A, Jia Z, Lee H. Machine-aided self-diagnostic prediction models for polycystic ovary syndrome: Observational Study. JMIR Formative Res. 2022;6(3):e29967.
Nsugbe E. An artificial intelligence-based decision support system for early diagnosis of polycystic ovaries syndrome. Healthc Anal. 2023;3:100164.
McHugh ML. Interrater reliability: the kappa statistic. Biochemia Med. 2012;22(3):276–82.
Acknowledgements
We would like to express our gratitude to the Tehran University of Medical Sciences for their invaluable support.
Funding
No funding sources.
Author information
Authors and Affiliations
Contributions
ABY and FS collected the data and wrote the first manuscript. LSH, LA, ShR.NK and JF supervised all stages of the study. All authors reviewed and approved the final manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
This study was carried out in the first phase of the thesis entitled “Design and evaluation of the application program of nutrition advisor for women with polycystic ovary syndrome (PCOS)” and was approved by the Research Ethics Committee of Tehran University of Medical Sciences, Tehran, Iran (ethical code: IR.TUMS.SPH.REC.1402.198).
Consent for publication
Not applicable.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
About this article
Cite this article
Shahmoradi, L., Azadbakht, L., Farzi, J. et al. Nutritional management recommendation systems in polycystic ovary syndrome: a systematic review. BMC Women's Health 24, 234 (2024). https://doi.org/10.1186/s12905-024-03074-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s12905-024-03074-3