Next Article in Journal
Strategic Control Enhancements for Frequency Deviation Management in European Power Systems
Previous Article in Journal
Extending the KT Cellular Automata Model for Signalized Urban Traffic
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 16
Issue 13
10.3390/app16136469
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Modular Digital Health Architecture for Longitudinal Menstrual Cycle Monitoring: System Design and Formative Usability Evaluation

by
Tomasz Bolesław Cedro
1,2,3,*,
Grzegorz Południewski
1,3 and
Wojciech Michał Glinkowski
3,4,*
1
IQCREDO SP. Z O.O., 00-105 Warsaw, Poland
2
CeDeROM, 01-864 Warsaw, Poland
3
Polish Telemedicine and e-Health Society, 03-728 Warsaw, Poland
4
Department of Medical Informatics and Telemedicine, Medical University of Warsaw, 02-091 Warsaw, Poland
*
Authors to whom correspondence should be addressed.
Appl. Sci. 2026, 16(13), 6469; https://doi.org/10.3390/app16136469
Submission received: 29 May 2026 / Revised: 13 June 2026 / Accepted: 26 June 2026 / Published: 29 June 2026
(This article belongs to the Special Issue Digital Healthcare IoT and Sensing Platforms)
Browse Figures
Versions Notes

Abstract

Background: Longitudinal menstrual cycle monitoring requires digital health systems capable of handling individual variability, irregular sampling, and incomplete real-world observations. Most consumer-focused menstrual-tracking applications depend on simplified calendar-based logic, offering limited support for transparent longitudinal data handling, interoperability, and the management of irregular real-world observations. Objective: This study presents the design, implementation, and formative evaluation of a non-clinical digital health infrastructure for longitudinal menstrual cycle monitoring, with an emphasis on modular system architecture, longitudinal data processing, and user-perceived usability. Methods: A modular digital health system was developed in accordance with separation-of-concerns and privacy-by-design principles, combining a backend analytical infrastructure with a mobile application interface. The architecture was designed to support longitudinal data acquisition, variability-aware processing, and extensibility while remaining independent of proprietary analytical services. System evaluation included technical and functional verification, formative usability assessment, and quality evaluation using the user version of the Mobile App Rating Scale (uMARS). Results: In the uMARS evaluation (N = 63), the mean total score across core domains was 3.11 ± 0.76. Information quality (3.44 ± 0.85) and functionality (3.27 ± 0.88) received the highest ratings, whereas engagement (2.83 ± 0.84) received the lowest, consistent with the system’s prototype character. Internal consistency was high (Cronbach’s α = 0.91), and sensitivity analysis restricted to female participants yielded results comparable to those of the full sample. Conclusions: The proposed system demonstrates the technical and functional feasibility of a modular digital health architecture for longitudinal menstrual cycle monitoring under heterogeneous real-world data conditions. The findings support the use of variability-aware and extensible monitoring infrastructures as a foundation for future applied research and iterative development of women’s digital health systems without making diagnostic or predictive clinical claims.

1. Introduction

Longitudinal biological processes have long been a subject of interest in biomedical sciences, with an emphasis on system dynamics, feedback mechanisms, and temporal variability rather than on isolated observations [1,2]. Biological rhythms, including the menstrual cycle, exhibit substantial intra- and interindividual variability over time [3,4,5,6,7]. From a systems perspective, such processes are better represented as longitudinally variable processes rather than as fixed or deterministic sequences.
The menstrual cycle is a particularly challenging example of a longitudinal physiological process [8,9,10]. Cycle length, phase duration, and associated physiological signals vary substantially over time and among individuals, whereas observations are often irregular, incomplete, and subject to measurement noise [11,12,13,14]. Despite this complexity, many widely used digital tools for menstrual cycle tracking rely on simplified calendar-based logic or static heuristics that inadequately capture the longitudinal variability and irregularity of the underlying processes [15,16,17,18,19].
Advances in digital health technologies have enabled the large-scale collection of longitudinal, user-generated health data outside controlled clinical environments [20,21,22]. However, from a biomedical and systems-engineering standpoint, such data pose additional challenges related to signal reliability, irregular sampling, and missing observations [9,23,24]. Effective system-level modeling necessitates architectures and analytical pipelines that can handle uncertainty, continuously update longitudinal representations as new observations arise, and preserve the interpretability of the model outputs [25,26].
Within the field of biomedical engineering, there is growing recognition that longitudinal health data should be treated as an evolving process rather than as a collection of independent measurements [27,28,29]. Modeling approaches that explicitly account for variability and temporal structure are central to this perspective [30,31]. However, these principles are rarely implemented transparently in consumer-oriented systems that integrate data acquisition, processing, and presentation into a single framework [32,33,34].
To contextualize the proposed framework within the current landscape of menstrual health applications, a conceptual comparison between typical consumer-oriented menstrual tracking apps and the proposed architecture is presented in Table 1.
As shown in Table 1, the proposed framework differs from typical consumer-oriented applications by prioritizing longitudinal data representation, modular analytical processing, and architectural transparency over single-purpose cycle prediction functionality.
Unlike most commercially available menstrual tracking applications, which primarily focus on cycle prediction and user-facing functionality, the proposed system was conceived as a modular research and development platform for longitudinal reproductive-health monitoring. The architecture enables multiple analytical modules to operate on the same longitudinal dataset, facilitating comparative evaluation of alternative modeling approaches while preserving the separation among the data acquisition, analytical processing, and presentation layers. Future extensions may include integration of multimodal physiological measurements, wearable and dedicated sensing devices, fertility-awareness applications, infertility support tools, and other reproductive-health use cases built upon the same underlying infrastructure.
This study presents the design, implementation, and formative evaluation of a modular digital health system intended to support longitudinal menstrual cycle monitoring under heterogeneous real-world data conditions. The evaluation focused on system architecture, analytical processing principles, technical and functional feasibility, and user-perceived usability, assessed using the user version of the Mobile Application Rating Scale (uMARS) [35], without making diagnostic, predictive, or clinical-effectiveness claims.

2. Materials and Methods

2.1. System Requirements and Design Goals

The system was designed as a non-clinical, research-oriented eHealth platform intended to support longitudinal observations and structured representation of women’s menstrual cycle-related data. The primary design objective was to enable continuous, user-driven data collection and analysis outside controlled clinical environments while maintaining transparency in data processing and methodological assumptions.
The key system requirements include support for long-term observation across multiple cycles, handling irregular and incomplete data, and an explicit separation between the data acquisition, analytical processing, and presentation layers of the system. The system was intentionally conceived as an assistive and exploratory information system rather than as a diagnostic or decision-support tool. This platform does not generate automated clinical interpretations, diagnostic outputs, or treatment recommendations.
From an engineering perspective, additional design goals include modularity, extensibility, and independence from proprietary technology. These requirements were adopted to facilitate iterative development, independent verification, and future methodological extensions without requiring any restructuring of the core system.

2.2. System Architecture

The system was implemented as a modular multilayer digital health architecture composed of three primary components: (i) an Internet-based backend platform, (ii) a mobile application serving as the primary user interface, and (iii) optional external measurement devices acting as auxiliary data sources for the mobile application. This layered design was adopted to support longitudinal data handling, modular extensibility, interoperability, and separation of concerns, while preserving the platform’s exploratory, non-clinical scope. Each component fulfills a distinct role within the system and communicates through well-defined interfaces. The overall modular architecture of the proposed digital health infrastructure is presented in Figure 1.
The backend Internet platform is responsible for centralized data storage, authentication, and execution of analytical routines. The backend exposes a set of application programming interfaces (APIs) enabling secure communication with client applications and internal analytical services. Analytical processing is performed on the server side to ensure consistent model execution and decouple algorithmic updates from client-side deployment.
The mobile application serves as the primary user interface, enabling manual data entry, visualization of longitudinal patterns, and user interaction with stored information. The client application does not perform analytical inference locally; instead, it retrieves processed summaries generated by the backend infrastructure, enabling a lightweight and maintainable client design.
Optional external measurement devices are treated as interchangeable auxiliary input sources rather than mandatory system elements. Their integration is mediated through standardized data representations, supporting device-level interoperability without modifying the core system logic.
The architectural separation of concerns minimizes component coupling, allowing individual subsystems, analytical modules, or data acquisition components to be modified, replaced, or extended without affecting overall system functionality. This design supports incremental system evolution and accommodates heterogeneous longitudinal data sources commonly encountered in consumer-oriented digital health applications.

2.3. Core Design Principles

The system was designed according to a set of engineering and digital health infrastructure principles intended to support longitudinal monitoring across heterogeneous real-world conditions while preserving modularity, extensibility, and transparent data handling. The principal design assumptions and their practical implementation within the proposed architecture are summarized in Table 2.

2.4. Data Acquisition and Management

The system supports the acquisition of longitudinal data from user-reported observations and, optionally, consumer-accessible physiological proxy data. These inputs may include cycle timing information, basal body temperature trends, and results from home-based biochemical tests of body fluids, such as urine. Such measurements have been reported to correlate with underlying hormonal processes; however, they are considered indirect indicators rather than precise laboratory-grade measurements.
All incoming observations were stored as time-stamped data records, accompanied by structured metadata describing the source, acquisition context, and measurement type. This unified representation enables the management of heterogeneous inputs within a common data model and supports subsequent analyses.
Data storage used a structured relational schema that was optimized for repeated observations over time. The system was designed to tolerate missing, delayed, or irregular data entries, which are common in self-reported datasets. No third-party analytics services or external cloud-based inference platforms were employed, and all data processing was performed within the system infrastructure.
User-reported symptoms and subjective observations were used as contextual data. These inputs are not interpreted diagnostically; rather, they enrich the temporal dataset and provide additional context for exploratory analyses and reflective self-monitoring.

2.5. Algorithms and Data Processing Pipeline

The analytical core of the system was designed to support longitudinal handling of heterogeneous menstrual cycle-related observations acquired under real-world conditions rather than isolated single-point measurements. The analytical framework emphasizes temporal continuity, individual variability, and incremental updating of user-specific longitudinal representations as new observations become available.
Incoming data are represented as time-stamped observations accompanied by structured metadata describing the acquisition context, signal provenance, and measurement type. Before analytical processing, the data undergo structural validation to ensure temporal consistency and completeness of the required attributes. This validation is intentionally limited to data integrity and does not impose physiological thresholds or clinical constraints, thereby preserving the system’s exploratory, non-clinical nature.
The analytical pipeline maintains individualized longitudinal representations of recurring temporal patterns and variability ranges observed across successive cycles. Longitudinal summaries are updated incrementally as new observations are acquired rather than recalculated exclusively through retrospective batch processing. Variability, irregular sampling, and missing observations are treated as expected characteristics of real-world longitudinal monitoring data rather than as artifacts to be excluded.
From a modeling perspective, the processing framework supports heterogeneous temporal behavior and unequal sampling intervals commonly encountered in self-reported and consumer-generated health data. Instead of enforcing rigid cycle segmentation rules, the system preserves variability-aware temporal representations that explicitly acknowledge uncertainty associated with irregular and incomplete observations.
Analytical components are implemented as modular services operating within the backend infrastructure. This modularity enables alternative modeling strategies, parameterizations, and future methodological extensions to be evaluated, replaced, or refined without altering the data acquisition layer or presentation interface. Artificial intelligence and machine-learning methods are treated as optional analytical extensions and do not generate autonomous clinical classifications, diagnoses, or treatment recommendations.
To illustrate the modular nature of the analytical framework, representative analytical modules currently implemented or evaluated within the platform are summarized in Table 3. These modules demonstrate alternative approaches to longitudinal cycle representation and are presented as exploratory analytical components rather than validated clinical prediction models.
The platform architecture allows multiple analytical modules to operate on the same longitudinal dataset. This design enables comparative evaluation of alternative modeling approaches and supports future methodological development without modification of the underlying data acquisition or storage infrastructure.
When new observations are recorded, individualized longitudinal summaries are incrementally updated to include cycle-length distributions, variability ranges, rolling averages, and trend descriptors. Missing observations do not interrupt processing; instead, summaries are recalculated using available observations while preserving temporal continuity.
The processing pipeline produces descriptive and model-derived summaries intended to characterize temporal patterns, stability, variability, and deviations relative to individualized reference behavior. These outputs are designed to support exploratory longitudinal interpretation rather than deterministic inference. Figure 2 summarizes the longitudinal data acquisition and analytical workflow implemented in the proposed system.

2.6. Privacy, Security, and Ethical Design

Privacy protection is incorporated as a foundational design principle. User identification was minimized, and personal data were processed solely for system operation and longitudinal representation. Data transmission between system components is secured through encrypted communication channels, and access to stored data is restricted through authentication and authorization mechanisms.
The system was implemented using a software stack that incorporates Free and Open-Source Software components operating under permissive licenses, including BSD, MIT, Apache, and GNU licenses. The architectural philosophy intentionally prioritizes interoperability, transparency, long-term maintainability, and avoidance of vendor lock-in. Although the current platform is not released as an open-source product, its implementation was intentionally designed around open technologies and modular interfaces to facilitate future extensibility and independent verification.
The system was designed as a low-risk digital health application intended for observational and assistive use. Ethical considerations related to user consent and data handling were addressed by clearly communicating the system’s scope and limitations. Clinical deployment, regulatory certification, and device-level validation were explicitly beyond the scope of this study.
The usability assessment was intended as a formative exploratory evaluation of user experience and perceived usability rather than a clinical validation or effectiveness study.
The research protocol was reviewed and approved on 21 February 2022, and was conducted in accordance with the principles of scientific ethics and the Declaration of Helsinki by the Bioethical Commission (approval no. AKBE/60/2022).

2.7. Participants and Recruitment

Eligible participants were adult Polish-speaking smartphone users. Participants were recruited through an open, convenience-based online recruitment strategy. Members of the Polish Telemedicine and e-Health Society disseminated information about the study. They invited potential participants to take part during professional presentations, educational meetings, and conferences, as well as through online communication channels. Eligible participants were adults, Polish-speaking smartphone users who voluntarily agreed to test the application and complete the uMARS questionnaire. Before participation, all individuals received detailed information about the study’s purpose and procedures, and provided informed consent electronically. Participants were informed that participation was voluntary and that they could withdraw from the study at any time without consequences.
A total of 66 questionnaires were collected. Of these, 63 contained complete data for the core uMARS domains and were included in the main analysis. The recruited sample included 42 women (63.6%) and 24 men (36.4%). The median age was 19 years, with a range of 18 to 70 years (mean age: 21.7 ± 9.6 years). Most respondents were students or pupils, reflecting the convenience-based, early-stage, formative nature of the evaluation. 95.5% of participants were Android users. No personally identifiable data was collected. All data were collected and anonymized. Data storage and processing complied with applicable data protection regulations, and access to the data was restricted to members of the research team. The participants did not receive any financial or other compensation for their participation.

2.8. Technical and Functional System Evaluation

Basic technical and functional evaluations were conducted to verify system operability and internal consistency before the user-facing usability assessment. The evaluation focused on the functional completeness of the core workflows, including data entry, secure data transmission, backend storage, analytical processing, and retrieval of summarized outputs using a mobile application.
The system components were tested to confirm the proper handling of time-stamped longitudinal data, tolerance for missing or irregular entries, and stability of backend services under routine usage conditions. Functional verification confirmed that the analytical routines were executed as intended and that updates to the model-derived summaries were consistently reflected in the app user interface. Representative technical verification scenarios, predefined acceptance criteria, and observed outcomes are summarized in Supplementary Table S2.
This evaluation was formative and aimed to confirm technical readiness for an exploratory user evaluation rather than assess performance, scalability, or regulatory compliance.
The proposed system was designed as an exploratory longitudinal monitoring infrastructure rather than a clinical diagnostic or predictive decision-making system.

3. Results

3.1. Implemented System and Functional Capabilities

The proposed system architecture was implemented as a modular eHealth platform that integrated a backend data management and analytics layer with a mobile application interface. The system supports the longitudinal acquisition of user-reported data, secure data transmission, centralized storage, and server-side analytical processing.
Heterogeneous inputs are stored in a unified, time-stamped data representation, enabling incremental updates to model-derived summaries as new observations are made. Separating the data acquisition, analytical processing, and presentation layers enables independent modification of system components and supports iterative system development.
Analytical routines operate on aggregated longitudinal data and generate descriptive and model-based summaries that characterize temporal patterns and intra-individual variability. These summaries were presented to users via a mobile application in a trend-oriented and uncertainty-aware manner. The system does not generate deterministic classifications, predictions, or clinical interpretations.

3.2. Technical and Functional Evaluation

A technical and functional evaluation was conducted to verify the proper operation of the core system workflow before the user-facing usability assessment. The evaluation confirmed the functionality of data entry, secure data transmission, backend storage, the execution of analytical routines, and the retrieval of processed summaries via the mobile application. Technical verification confirmed correct operation of all core workflows.
The system demonstrated the correct handling of time-stamped longitudinal records, including tolerance for missing and irregular entries. Updates to the analytical components were consistently reflected in the user interface without disrupting the data acquisition or storage. This evaluation was limited to verifying functional readiness and internal consistency and did not include performance benchmarking or scalability testing.

3.3. User-Perceived Usability and Quality (uMARS)

The Polish end-user version of the Mobile Application Rating Scale (uMARS) was used as an internationally acclaimed tool for mHealth application quality assessment. A total of 66 uMARS questionnaires were collected; 63 contained complete data for the core uMARS domains and were included in the analysis. Incomplete questionnaires were excluded from the study. The mean total uMARS score, calculated as the average of the engagement, functionality, aesthetics, and information quality domains (Table 4), was 3.11 ± 0.76 (on a 1–5 scale).
Among the individual domains, information quality received the highest rating (3.44 ± 0.85), followed by functionality (3.27 ± 0.88). Aesthetics achieved a moderate score (3.08 ± 0.97), and engagement received the lowest score (2.83 ± 0.84). The subjective quality scores were lower (2.08 ± 0.92) and reflected overall user appraisal rather than core usability.
The internal consistency of the uMARS core domain was high (Cronbach’s α = 0.91).
A sensitivity analysis restricted to female participants yielded domain scores and total uMARS values comparable to those of the full sample, with no material differences in the domain rankings or overall results (Supplementary Table S1). The analysis was descriptive and intended to assess robustness rather than test group differences. Supplementary Table S1 compares the uMARS domain scores between the full sample and the female participants. Values are presented as mean ± standard deviation (SD).

4. Discussion

4.1. Summary of System Design and Evaluation Findings

This study presents the design and implementation of a non-clinical eHealth system for longitudinal monitoring of menstrual cycle-related data, combining a modular system architecture with exploratory algorithmic processing and formative usability evaluation. The results highlight both the system’s technical feasibility and user-perceived strengths and limitations at its current stage of development.

4.2. Engineering Trade-Offs and Biomedical Perspective

This study approaches menstrual cycle monitoring from biomedical sciences and systems engineering perspectives, emphasizing longitudinal variability and temporal changes rather than deterministic classification or prediction. This perspective is particularly relevant for longitudinal biological processes characterized by irregular sampling, incomplete observations, and substantial inter- and intra-individual heterogeneity.
The central trade-off in the proposed system is between model complexity and interpretability. Instead of implementing highly parameterized black-box predictive models, the system adopts a longitudinal variability-aware representation that prioritizes modularity and robustness in the presence of noisy or user-generated data. This choice reflects a deliberate design decision aligned with bioengineering principles, in which system transparency and stability are often more critical than short-term predictive accuracy, making it distinct from current market trends and other consumer-oriented solutions.
From a bioengineering standpoint, the platform can be viewed as a longitudinal monitoring infrastructure in which longitudinal summaries are updated incrementally as new observations are added. Simultaneously, rigid input segmentation was avoided, thereby ensuring flexibility in representing diverse patterns. Although the system is not yet positioned as a diagnostic or decision-support tool, it provides a structured foundation for future extensions, validation studies, and comparative modeling, including studies conducted under appropriate regulatory and ethical frameworks to meet stringent digital health requirements.
From an engineering perspective, the current architecture meets the key requirements for handling longitudinal health data, including modularity, extensibility, and tolerance to irregular and incomplete data. The explicit separation between the data storage, analytics, and presentation layers enables independent evolution of the system and aligns with best practices for research-oriented digital health platforms.
The inclusion of a usability evaluation using the uMARS instrument serves as supporting evidence of the quality of the current system implementation rather than as a primary outcome. In biomedical sciences, such evaluations complement technical verification by providing insights into how users interact with longitudinal data representations. Importantly, usability feedback enables system refinement without constraining the analytical core or imposing clinically oriented performance metrics [35,36,37,38].
Ethical considerations have influenced several key engineering decisions in the system’s design. The non-clinical positioning of the platform, explicit avoidance of diagnostic functionality, and separation between data acquisition and analytical processing were adopted not only for methodological clarity but also to reduce the ethical and regulatory risks associated with user-generated health data [39,40,41,42,43].

4.3. Implications for System-Oriented Digital Health Research

An important characteristic of the proposed framework is that it serves as a modular experimentation environment rather than a single-purpose application. The architecture was intentionally designed to accommodate multiple analytical modules, evolving modeling approaches, and future integration of additional physiological data streams [44,45,46]. This flexibility may support comparative evaluation of alternative prediction strategies and facilitate the development of broader reproductive-health applications, including fertility-awareness support, infertility-related monitoring, family-planning tools, and other longitudinal women’s health services. Beyond menstrual cycle monitoring itself, the framework may therefore serve as a reusable methodological infrastructure for future digital health research and development.
The usability evaluation using the uMARS instrument provides complementary information regarding user-perceived quality and system acceptance [47,48,49].
Higher scores for information quality and functionality suggest that users perceived the system as understandable, reliable, and operationally coherent. [43,50,51].
Lower Engagement and Subjective Quality scores are consistent with an early-stage prototype and indicate areas that require further refinement before broader implementation.
Several factors likely contributed to the lower engagement ratings. During the current stage of development, priority was intentionally given to architectural robustness, longitudinal data management, privacy protection, and analytical functionality rather than to advanced user experience design. Future iterations should focus on improving engagement through enhanced visualizations, personalized reminders, configurable user interfaces, educational content, and longitudinal feedback mechanisms. Additional approaches, including gamification-inspired features, reward-based participation strategies, and physician-guided monitoring pathways, may further support sustained user adherence and long-term engagement.
Importantly, the usability findings should be interpreted as formative feedback rather than evidence of effectiveness, clinical utility, or readiness for deployment within healthcare systems. The consistency of the uMARS results observed in both the full sample and the women-only sensitivity analysis supports the robustness of the usability assessment within the intended user population [52,53].
Taken together, the technical implementation and usability findings illustrate the challenges and opportunities associated with developing transparent, modular, and non-clinical digital health systems. The proposed framework prioritizes longitudinal representation of biological variability over deterministic outputs. It may therefore provide a useful foundation for future research-oriented digital health infrastructures operating under heterogeneous real-world conditions.

4.4. Study and System Limitations

The present study had several limitations that should be acknowledged when interpreting the results. First, the system evaluation was intentionally limited to formative and non-clinical scopes. No claims were made regarding the diagnostic accuracy, predictive performance, or clinical effectiveness. The analytical components were designed to support the exploratory representation of longitudinal patterns rather than validated clinical inferences.
Second, the system’s data sources included user-reported observations and consumer-accessible physiological proxy signals. These data are inherently subject to noise, missing data, and reporting bias. Although the system architecture and analytical pipeline explicitly accommodate irregular sampling and incomplete observations, these factors may still influence the stability and interpretability of the model-derived summaries.
Third, the usability evaluation using the uMARS instrument was conducted with a relatively homogeneous sample and reflected user-perceived quality early in system development. The evaluation focused on usability and perceived information quality rather than on long-term general-population engagement, behavioral impact, or real-world technology adoption. Therefore, the reported usability scores should be interpreted as formative rather than definitive indicators of the system’s maturity.
Fourth, the recruitment strategy may have introduced selection bias. Participants were recruited through online invitations and announcements disseminated by members of a professional telemedicine society during presentations, educational meetings, and conferences. The resulting sample was dominated by young, digitally literate users, particularly students, and therefore should not be considered representative of the general population of potential users of menstrual health apps. Consequently, the usability findings should be interpreted as formative feedback from an early, friendly user-testing cohort rather than as evidence of general-population acceptability.
Finally, the technical and functional evaluations were limited to verifying core workflows and system operability. Performance benchmarking, scalability testing, cybersecurity hardening, and regulatory compliance assessment were outside the scope of this study and remain to be addressed in subsequent development stages.

4.5. Directions for Future Development

Future studies should focus on a systematic empirical evaluation of the proposed system under controlled and semi-controlled conditions. This includes the collection of larger, more diverse longitudinal datasets, which will enable future assessments of model robustness across different usage patterns and degrees of data completeness.
Further development of the analytical pipeline may involve a comparative evaluation of alternative statistical and probabilistic modeling approaches and the incorporation of additional contextual covariates. Such extensions should be accompanied by transparent reporting of the modeling assumptions and uncertainties to preserve interpretability.
From a system perspective, future efforts should focus on optimizing user engagement and interaction design, informed by iterative usability testing and user feedback. The engagement-related domains identified in the uMARS evaluation represent clear targets for refining the design.
Finally, any transition toward clinical or decision-support use would require dedicated validation studies, regulatory reviews, and ethical assessments within appropriate governance frameworks. These steps are explicitly beyond the scope of the present study but are recognized as necessary prerequisites for broader deployment in the healthcare sector.
Future extensions may also explore fertility-awareness support, infertility-related monitoring, multimodal physiological sensing, and integration with wearable or dedicated reproductive-health devices.

5. Conclusions

This study describes the design, implementation, and formative evaluation of a modular eHealth system for the longitudinal monitoring of women’s menstrual cycle-related data. This study focused on system architecture, data-handling strategies, and algorithmic processing rather than on clinical validation or diagnostic performance.
By integrating longitudinal data representation with variability-aware analytical approaches, the system provides a structured framework for exploring temporal patterns. The modular and extensible design supports methodological transparency, independent verification, and future system evolution, aligning with research-oriented use cases in digital health.
The incorporation of a uMARS-based usability evaluation adds a user-centered perspective to the system description, identifying both strengths and areas for further refinement. These findings support iterative system development while maintaining clear boundaries between the provision of assistive information and clinical decision-making.
Overall, the proposed system provides a methodological and architectural foundation for future applied studies and empirical evaluations to advance longitudinal digital health monitoring in women’s health. Beyond menstrual cycle monitoring itself, the proposed architecture may serve as a reusable methodological foundation for future applications in reproductive health, fertility awareness, and longitudinal women’s digital health.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/app16136469/s1, Supplementary Table S1. Comparison of uMARS domain scores between the full study sample and the female participant subgroup. Supplementary Table S2. Representative technical verification scenarios.

Author Contributions

Conceptualization, T.B.C., G.P. and W.M.G.; Methodology, T.B.C., G.P. and W.M.G.; Software, T.B.C.; Validation, T.B.C., G.P. and W.M.G.; Formal analysis, T.B.C., G.P. and W.M.G.; Investigation, T.B.C., G.P. and W.M.G.; Resources, T.B.C. and G.P.; Data curation, T.B.C., G.P. and W.M.G.; Writing—original draft, T.B.C., G.P. and W.M.G.; Writing—review & editing, T.B.C. and W.M.G.; Visualization, T.B.C. and W.M.G.; Supervision, T.B.C., G.P. and W.M.G.; Project administration, T.B.C. and G.P.; Funding acquisition, T.B.C. and G.P. All authors have read and agreed to the published version of the manuscript.

Funding

Research and Development project co-founded with an EU grant RPMA.01.02.00-14-a180/18.

Institutional Review Board Statement

This study was approved by the Bioethical Commission (approval no. AKBE/60/2022). The research protocol was reviewed and approved on 21 February 2022, and was conducted in accordance with the principles of scientific ethics and the Declaration of Helsinki.

Informed Consent Statement

Informed consent was obtained electronically from all participants prior to their participation in the study. Participants were required to read the study information and actively select an “I agree” checkbox before accessing the questionnaire.

Data Availability Statement

The original contributions presented in this study are included in the article and Supplementary Material. Further inquiries can be directed to the corresponding author.

Acknowledgments

During the preparation of this manuscript, the authors used ChatGPT (OpenAI, GPT-5 series; https://chatgpt.com; accessed 17 June 2026) for language refinement, stylistic editing, organizational support, and assistance in the graphical implementation of Figure 1 and Figure 2 based on an author-defined conceptual framework. Generative artificial intelligence tools were used solely to assist in the graphical rendering of Figure 1 and Figure 2 according to concepts and structural designs fully specified by the authors. All scientific content and final editorial decisions remained under the authors’ control. Grammarly (Grammarly Inc.; https://www.grammarly.com; accessed 17 June 2026) was used for grammar and language checking, and Paperpal (Cactus Communications; https://paperpal.com; accessed 17 June 2026) was used for the final manuscript journal submission check. All scientific interpretation, literature analysis, classification development, and final manuscript decisions were performed exclusively by the authors.

Conflicts of Interest

Author Tomasz Bolesław Cedro is a founder of CeDeROM and IQCREDO company. Author Grzegorz Południewski is a founder of IQCREDO. The remaining authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Lehrer, P.; Eddie, D. Dynamic processes in regulation and some implications for biofeedback and biobehavioral interventions. Appl. Psychophysiol. Biofeedback 2013, 38, 143–155. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Patharkar, A.; Cai, F.; Al-Hindawi, F.; Wu, T. Predictive modeling of biomedical temporal data in healthcare applications: Review and future directions. Front. Physiol. 2024, 15, 1386760. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Kembro, J.M.; Flesia, A.G.; Acosta-Rodríguez, V.A.; Takahashi, J.S.; Nieto, P.S. Dietary restriction modulates ultradian rhythms and autocorrelation properties in mice behavior. Commun. Biol. 2024, 7, 303. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Pittayakanchit, W.; Lu, Z.; Chew, J.; Rust, M.J.; Murugan, A. Biophysical clocks face a trade-off between internal and external noise resistance. eLife 2018, 7, e37624. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  5. Rahman, S.A.; Grant, L.K.; Gooley, J.J.; Rajaratnam, S.M.W.; Czeisler, C.A.; Lockley, S.W. Endogenous Circadian Regulation of Female Reproductive Hormones. J. Clin. Endocrinol. Metab. 2019, 104, 6049–6059. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  6. Derry, G.; Derry, P. Characterization of chaotic dynamics in the human menstrual cycle. Nonlinear Biomed. Phys. 2010, 4, 5. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  7. Ecochard, R.; Stanford, J.B.; Fehring, R.J.; Schneider, M.; Najmabadi, S.; Gronfier, C. Evidence that the woman’s ovarian cycle is driven by an internal circamonthly timing system. Sci. Adv. 2024, 10, eadg9646. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  8. Lin, G.; Li, J.Y.; Kalani, K.; Truong, K.N.; Mariakakis, A. A longitudinal dataset of physiological, hormonal, metabolic, and self-reported menstrual health data. Sci. Data 2026, 13, 411. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  9. Li, K.; Urteaga, I.; Wiggins, C.H.; Druet, A.; Shea, A.; Vitzthum, V.J.; Elhadad, N. Characterizing physiological and symptomatic variation in menstrual cycles using self-tracked mobile-health data. npj Digit. Med. 2020, 3, 79. [Google Scholar] [PubMed]
  10. Fehring, R.J.; Schneider, M.; Raviele, K. Variability in the phases of the menstrual cycle. J. Obstet. Gynecol. Neonatal Nurs. 2006, 35, 376–384. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  11. Najmabadi, S.; Schliep, K.C.; Simonsen, S.E.; Porucznik, C.A.; Egger, M.J.; Stanford, J.B. Menstrual bleeding, cycle length, and follicular and luteal phase lengths in women without known subfertility: A pooled analysis of three cohorts. Paediatr. Perinat. Epidemiol. 2020, 34, 318–327. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. Henry, S.; Shirin, S.; Goshtasebi, A.; Prior, J.C. Prospective 1-year assessment of within-woman variability of follicular and luteal phase lengths in healthy women prescreened to have normal menstrual cycle and luteal phase lengths. Hum. Reprod. 2024, 39, 2565–2574. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. Grieger, J.A.; Norman, R.J. Menstrual Cycle Length and Patterns in a Global Cohort of Women Using a Mobile Phone App: Retrospective Cohort Study. J. Med. Internet Res. 2020, 22, e17109. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  14. Mackenzie, A.C.L.; Chung, S.; Hoppes, E.; Mickler, A.K.; Cartwright, A.F. Measurement of changes to the menstrual cycle: A transdisciplinary systematic review evaluating measure quality and utility for clinical trials. PLoS ONE 2024, 19, e0306491. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. Houghton, L.C.; Elhadad, N. Practice Note: ‘If Only All Women Menstruated Exactly Two Weeks Ago’: Interdisciplinary Challenges and Experiences of Capturing Hormonal Variation Across the Menstrual Cycle. In The Palgrave Handbook of Critical Menstruation Studies; Bobel, C., Winkler, I.T., Fahs, B., Hasson, K.A., Kissling, E.A., Roberts, T.A., Eds.; Palgrave Macmillan: Singapore, 2020; pp. 725–732. [Google Scholar]
  16. Johnson, S.; Marriott, L.; Zinaman, M. Can apps and calendar methods predict ovulation with accuracy? Curr. Med. Res. Opin. 2018, 34, 1587–1594. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  17. Freis, A.; Freundl-Schütt, T.; Wallwiener, L.M.; Baur, S.; Strowitzki, T.; Freundl, G.; Frank-Herrmann, P. Plausibility of Menstrual Cycle Apps Claiming to Support Conception. Front. Public Health 2018, 6, 98. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  18. Schantz, J.S.; Fernandez, C.S.P.; Anne Marie, Z.J. Menstrual Cycle Tracking Applications and the Potential for Epidemiological Research: A Comprehensive Review of the Literature. Curr. Epidemiol. Rep. 2021, 8, 9–19. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Broad, A.; Biswakarma, R.; Harper, J.C. A survey of women’s experiences of using period tracker applications: Attitudes, ovulation prediction and how the accuracy of the app in predicting period start dates affects their feelings and behaviors. Womens Health 2022, 18, 17455057221095246. [Google Scholar]
  20. Lavallee, D.C.; Lee, J.R.; Austin, E.; Bloch, R.; Lawrence, S.O.; McCall, D.; Munson, S.A.; Nery-Hurwit, M.; Amtmann, D. mHealth and patient generated health data: Stakeholder perspectives on opportunities and barriers for transforming healthcare. Mhealth 2020, 6, 8. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  21. Guardado, S.; Karampela, M.; Isomursu, M.; Grundstrom, C. Use of Patient-Generated Health Data from Consumer-Grade Devices by Health Care Professionals in the Clinic: Systematic Review. J. Med. Internet Res. 2024, 26, e49320. [Google Scholar] [PubMed]
  22. Khaled, M.; Sareban, M.; Kreuzthaler, M.; Schulz, S.; Hussein, R. Overview of Existing Tools for Extracting Health and Fitness Data from mHealth Apps. Stud. Health Technol. Inform. 2022, 295, 49–50. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  23. West, P.; Van Kleek, M.; Giordano, R.; Weal, M.; Shadbolt, N. Information Quality Challenges of Patient-Generated Data in Clinical Practice. Front. Public Health 2017, 5, 284. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Abdolkhani, R.; Gray, K.; Borda, A.; DeSouza, R. Patient-generated health data management and quality challenges in remote patient monitoring. JAMIA Open. 2019, 2, 471–478. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  25. Namli, T.; Anıl Sınacı, A.; Gönül, S.; Herguido, C.R.; Garcia-Canadilla, P.; Muñoz, A.M.; Esteve, A.V.; Erturkmen, G.B.L. A scalable and transparent data pipeline for AI-enabled health data ecosystems. Front. Med. 2024, 11, 1393123. [Google Scholar] [CrossRef] [Scilit]
  26. Kimpton, L.M.; Paun, L.M.; Colebank, M.J.; Volodina, V. Challenges and opportunities in uncertainty quantification for healthcare and biological systems. Philos. Trans. A Math. Phys. Eng. Sci. 2025, 383, 20240232. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  27. Rose, S.M.S.-F.; Contrepois, K.; Moneghetti, K.J.; Zhou, W.; Mishra, T.; Mataraso, S.; Dagan-Rosenfeld, O.; Ganz, A.B.; Dunn, J.; Hornburg, D.; et al. A longitudinal big data approach for precision health. Nat. Med. 2019, 25, 792–804. [Google Scholar] [CrossRef] [Scilit]
  28. Mueller-Peltzer, M.; Feuerriegel, S.; Molgaard Nielsen, A.; Kongsted, A.; Vach, W.; Neumann, D. Longitudinal healthcare analytics for disease management: Empirical demonstration for low back pain. Decis. Support Syst. 2020, 132, 113271. [Google Scholar] [CrossRef] [Scilit]
  29. Taheriyoun, A.R.; Ross, A.; Safikhani, A.; Soudbakhsh, D.; Rahnavard, A. Longitudinal omics data analysis: Approaches and applications. Comput. Struct. Biotechnol. J. 2026, 31, 301–315. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  30. Cauchi, M.; Mills, A.R.; Lawrie, A.; Kiely, D.G.; Kadirkamanathan, V. Individualized survival predictions using state space model with longitudinal and survival data. J. R. Soc. Interface 2024, 21, 20230682. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  31. Hosseinzadeh, M.M.; Cannataro, M.; Guzzi, P.H.; Dondi, R. Temporal networks in biology and medicine: A survey on models, algorithms, and tools. Netw. Model. Anal. Health Inform. Bioinform. 2023, 12, 10. [Google Scholar] [PubMed]
  32. Kaplan, B. Seeing through health information technology: The need for transparency in software, algorithms, data privacy, and regulation. J. Law Biosci. 2020, 7, lsaa062. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  33. Quinn, T.P.; Jacobs, S.; Senadeera, M.; Le, V.; Coghlan, S. The three ghosts of medical AI: Can the black-box present deliver? Artif. Intell. Med. 2022, 124, 102158. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  34. Mohan, S.; Jenkins, J. Flowing data: Women’s views and experiences on privacy and data security when using menstrual cycle tracking apps. Oxf. Open Digit. Health 2025, 3, oqaf011. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Glinkowski, W.M.; Cedro, T.; Nowicka, J.; Doniec, R.; Michalak, B.; Glinkowska, B.; Stoyanov, S. Translation and Validation of the User Version of the Mobile Application Rating Scale Into the Polish Language: Cross-Sectional Methodological Study. JMIR Form. Res. 2026, 10, e65067. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Muro-Culebras, A.; Escriche-Escuder, A.; Martin-Martin, J.; Roldán-Jiménez, C.; De-Torres, I.; Ruiz-Muñoz, M.; Gonzalez-Sanchez, M.; Mayoral-Cleries, F.; Biró, A.; Tang, W.; et al. Tools for Evaluating the Content, Efficacy, and Usability of Mobile Health Apps According to the Consensus-Based Standards for the Selection of Health Measurement Instruments: Systematic Review. JMIR mHealth uHealth 2021, 9, e15433. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  37. Stoyanov, S.R.; Hides, L.; Kavanagh, D.J.; Wilson, H. Development and Validation of the User Version of the Mobile Application Rating Scale (uMARS). JMIR mHealth uHealth 2016, 4, e72. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  38. Cho, H.; Yen, P.Y.; Dowding, D.; Merrill, J.A.; Schnall, R. A multi-level usability evaluation of mobile health applications: A case study. J. Biomed. Inform. 2018, 86, 79–89. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  39. Rossmaier, L.W.S. Commercial mHealth Apps and Unjust Value Trade-Offs: A Public Health Perspective. Public Health Ethics 2022, 15, 277–288. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  40. Mennella, C.; Maniscalco, U.; De Pietro, G.; Esposito, M. Ethical and regulatory challenges of AI technologies in healthcare: A narrative review. Heliyon 2024, 10, e26297. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  41. Williams, M.; Karim, W.; Gelman, J.; Raza, M. Ethical data acquisition for LLMs and AI algorithms in healthcare. npj Digit. Med. 2024, 7, 377. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  42. Hammack-Aviran, C.M.; Brelsford, K.M.; Beskow, L.M. Ethical Considerations in the Conduct of Unregulated mHealth Research: Expert Perspectives. J. Law Med. Ethics. 2020, 48, 9–36. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  43. Maeckelberghe, E.; Zdunek, K.; Marceglia, S.; Farsides, B.; Rigby, M. The ethical challenges of personalized digital health. Front. Med. 2023, 10, 1123863. [Google Scholar] [CrossRef] [Scilit]
  44. McNamee, D.; Wolpert, D.M. Internal Models in Biological Control. Annu. Rev. Control. Robot. Auton. Syst. 2019, 2, 339–364. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  45. Paganelli, A.I.; Mondéjar, A.G.; da Silva, A.C.; Silva-Calpa, G.; Teixeira, M.F.; Carvalho, F.; Raposo, A.; Endler, M. Real-time data analysis in health monitoring systems: A comprehensive systematic literature review. J. Biomed. Inform. 2022, 127, 104009. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  46. Abdolkhani, R.; Gray, K.; Borda, A.; DeSouza, R. Recommendations for the Quality Management of Patient-Generated Health Data in Remote Patient Monitoring: Mixed Methods Study. JMIR mHealth uHealth 2023, 11, e35917. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  47. Hutton, L.; Price, B.A.; Kelly, R.; McCormick, C.; Bandara, A.K.; Hatzakis, T.; Meadows, M.; Nuseibeh, B. Assessing the Privacy of mHealth Apps for Self-Tracking: Heuristic Evaluation Approach. JMIR mHealth uHealth 2018, 6, e185. [Google Scholar] [PubMed]
  48. Grosman-Rimon, L.; Wegier, P. With advancement in health technology comes great responsibility—Ethical and safety considerations for using digital health technology: A narrative review. Medicine 2024, 103, e39136. [Google Scholar] [PubMed]
  49. Benjumea, J.; Ropero, J.; Rivera-Romero, O.; Dorronzoro-Zubiete, E.; Carrasco, A. Privacy Assessment in Mobile Health Apps: Scoping Review. JMIR mHealth uHealth 2020, 8, e18868. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  50. Gomis-Pastor, M.; Berdún, J.; Borrás-Santos, A.; López, A.D.D.; Rama, B.F.-M.; García-Esquirol, Ó.; Gratacòs, M.; Rodríguez, G.D.O.; Cruz, R.P.; Real, J.; et al. Clinical Validation of Digital Healthcare Solutions: State of the Art, Challenges and Opportunities. Healthcare 2024, 12, 1057. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  51. Alifia, R.R.; Sadeghi, M.; Eluru, M.; Jafari, M.; Grando, M.A. Bridging ethical gaps in digital health research: A framework for informed consent aligned with NIH guidance. BMC Med. Ethics 2025, 26, 132. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  52. Roberts, A.E.; Davenport, T.A.; Wong, T.; Moon, H.-W.; Hickie, I.B.; LaMonica, H.M. Evaluating the quality and safety of health-related apps and e-tools: Adapting the Mobile App Rating Scale and developing a quality assurance protocol. Internet Interv. 2021, 24, 100379. [Google Scholar] [PubMed]
  53. Sharma, S.; Kumar, B.A. A systematic review of user-based usability testing practices in self-care mHealth apps. Digit. Health 2025, 11, 20552076251374184. [Google Scholar]
Applsci 16 06469 g001
Figure 1. Modular multilayer architecture of the proposed digital health system for longitudinal menstrual cycle monitoring. The system separates data acquisition, backend analytical processing, longitudinal storage, and user-facing visualization layers to support modularity, extensibility, and interoperability under heterogeneous real-world data conditions.
Figure 1. Modular multilayer architecture of the proposed digital health system for longitudinal menstrual cycle monitoring. The system separates data acquisition, backend analytical processing, longitudinal storage, and user-facing visualization layers to support modularity, extensibility, and interoperability under heterogeneous real-world data conditions.
Applsci 16 06469 g001
Applsci 16 06469 g002
Figure 2. Longitudinal data acquisition and processing workflow. The pipeline supports heterogeneous user-generated observations, incremental updating of longitudinal records, variability-aware processing, and exploratory trend-oriented summaries without generating diagnostic or predictive clinical outputs.
Figure 2. Longitudinal data acquisition and processing workflow. The pipeline supports heterogeneous user-generated observations, incremental updating of longitudinal records, variability-aware processing, and exploratory trend-oriented summaries without generating diagnostic or predictive clinical outputs.
Applsci 16 06469 g002
Table 1. Conceptual comparison of common design characteristics reported for consumer menstrual tracking applications and the proposed framework.
Table 1. Conceptual comparison of common design characteristics reported for consumer menstrual tracking applications and the proposed framework.
CharacteristicTypical Consumer AppsProposed Framework
Primary objectiveCycle prediction and user engagementLongitudinal monitoring infrastructure
Architecture transparencyUsually not disclosedExplicitly described
Analytical processingOften proprietaryTransparent and modular
Handling of missing dataVariableExplicit tolerance
Interoperability orientationLimited or vendor-specificExplicit design objective
Longitudinal variability modellingUsually simplifiedCore design principle
Research-oriented architectureRarePrimary objective
Technology strategyMixed, frequently proprietaryOpen technologies and modular interfaces
Clinical decision supportSometimes impliedExplicitly excluded
Table 2. Core engineering principles and their implementation within the proposed digital health system.
Table 2. Core engineering principles and their implementation within the proposed digital health system.
Design PrincipleEngineering RationaleSystem Implementation
Longitudinal time-stamped observationsPreserve temporal continuityAll observations stored as timestamped records
Tolerance for incomplete and irregular dataReflect real-world user behaviorMissing observations retained without exclusion
Modular layered infrastructureReduce component couplingIndependent acquisition, analytics, and presentation layers
Extensible analytical modulesSupport future methodological evolutionBackend analytical services can be replaced independently
Interoperable external data integrationFacilitate interoperabilityStandardized data representation layer
Non-diagnostic exploratory frameworkMaintain methodological transparencyNo automated diagnosis, treatment recommendations, or clinical decision support
Table 3. Representative analytical modules implemented within the platform.
Table 3. Representative analytical modules implemented within the platform.
ModuleInput DataAnalytical ApproachOutput
ai_cycle_calendarHistorical cycle datesCalendar repetition heuristicPredicted cycle timing
ai_cycle_statCycle historyDescriptive statistics and variability estimationIndividualized cycle ranges
ai_cycle_mlMultimodal inputs (temperature, hormone measurements, symptoms)Experimental machine-learning approachesExploratory cycle predictions
Table 4. uMARS domain scores (mean ± SD).
Table 4. uMARS domain scores (mean ± SD).
DomainMean ± SD
Engagement2.83 ± 0.84
Functionality3.27 ± 0.88
Aesthetics3.08 ± 0.97
Information3.44 ± 0.85
uMARS total (A–D)3.11 ± 0.76
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Cedro, T.B.; Południewski, G.; Glinkowski, W.M. A Modular Digital Health Architecture for Longitudinal Menstrual Cycle Monitoring: System Design and Formative Usability Evaluation. Appl. Sci. 2026, 16, 6469. https://doi.org/10.3390/app16136469

AMA Style

Cedro TB, Południewski G, Glinkowski WM. A Modular Digital Health Architecture for Longitudinal Menstrual Cycle Monitoring: System Design and Formative Usability Evaluation. Applied Sciences. 2026; 16(13):6469. https://doi.org/10.3390/app16136469

Chicago/Turabian Style

Cedro, Tomasz Bolesław, Grzegorz Południewski, and Wojciech Michał Glinkowski. 2026. "A Modular Digital Health Architecture for Longitudinal Menstrual Cycle Monitoring: System Design and Formative Usability Evaluation" Applied Sciences 16, no. 13: 6469. https://doi.org/10.3390/app16136469

APA Style

Cedro, T. B., Południewski, G., & Glinkowski, W. M. (2026). A Modular Digital Health Architecture for Longitudinal Menstrual Cycle Monitoring: System Design and Formative Usability Evaluation. Applied Sciences, 16(13), 6469. https://doi.org/10.3390/app16136469

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop