Post market clinical evaluation of Medical Devices Integrating Health Economics and Artificial Intelligence for Improved Outcome

 

Gerasimos Bobis ¹, Dimitra Tzamaria ²,  Chrysoula I. Liakou ³, Petaniti Evangelia ³, Marios Papadakis ⁴, Markos Plytas ⁵

 

¹ Biomedical Scientist, MBA, Greece

² MSc. Biology Science, Greece

³ MD, PhD, Chief Executive Officer, LK Connect LLC. 30N Gould St. Sheridan - WY 82801, Sheridan County, Wyoming, United States

⁴ PhD. Department of Surgery II. University of Witten-Herdecke, Wuppertal, Germany

⁵ MSc. Academic Director, Epsilon College, Athens, Greece

 

1. INTRODUCTION

The landscape of medical device regulation has undergone transformative changes in recent years, particularly in the European Union (EU) and other regions with increasingly rigorous post-market surveillance (PMS) requirements. Post-market clinical evaluation (PMCE) has become a pivotal component of medical device lifecycle management, ensuring continuous assessment of safety, performance, and real-world effectiveness after a device enters the market European Commission (2020). As technological complexity grows and patient expectations evolve, regulatory bodies are now placing emphasis not only on clinical data but also on broader health outcomes and economic impact. In this context, integrating Health Economics (HE) and Artificial Intelligence (AI) into PMCE processes presents a compelling opportunity to enhance the robustness and relevance of post-market data.

Health Economics offers structured methodologies to evaluate the cost-effectiveness, budget impact, and value proposition of medical devices in real-world settings. These economic evaluations extend beyond mere cost assessments and delve into patient outcomes, quality-adjusted life years (QALYs), and overall societal benefit Drummond et al. (2015). Incorporating HE into PMCE enables stakeholders—including manufacturers, healthcare providers, and regulators—to make more informed decisions grounded in both clinical and economic evidence. Such integration can support market access strategies, reimbursement decisions, and policy formulation while aligning with broader value-based healthcare models Sorenson et al. (2008).

Concurrently, the rapid evolution of AI and machine learning (ML) technologies is reshaping how real-world evidence (RWE) is collected, analyzed, and interpreted. AI algorithms are increasingly used to automate data extraction from electronic health records (EHRs), patient registries, and wearable devices, enabling continuous monitoring of device performance in diverse clinical environments Topol (2019). Machine learning models can detect early safety signals, predict patient outcomes, and stratify risk with a granularity that traditional methods often fail to achieve Esteva et al. (2019). In the context of PMCE, these capabilities significantly augment traditional methodologies, allowing for proactive rather than reactive post-market surveillance.

The synergy between Health Economics and AI holds particular promise when conceptualized through the lens of “exchanged outcomes.” This emerging paradigm emphasizes bidirectional data flows and collaborative evaluation among multiple stakeholders—patients, clinicians, payers, regulators, and manufacturers. AI tools can personalize economic evaluations by tailoring cost-effectiveness analyses to patient subgroups, thereby improving relevance and applicability. Meanwhile, HE frameworks can contextualize AI-driven insights within real-world cost and resource constraints, enhancing their value for decision-making Garrison et al. (2018). By linking clinical, economic, and technical dimensions, exchanged outcomes foster a more holistic and adaptive model of post-market evaluation.

Despite the promise of this integrated approach, several challenges persist. Data interoperability, regulatory acceptance of AI models, ethical considerations, and standardization of economic outcomes remain significant barriers. For instance, current regulatory frameworks are still evolving in terms of how to assess AI-enabled tools, especially those that learn and adapt over time European Medicines Agency (EMA). (2021). Moreover, the integration of economic data requires alignment with clinical endpoints and consistent methodologies to ensure comparability across studies and settings.

Nevertheless, forward-thinking manufacturers and regulators are beginning to embrace these innovations. Initiatives like the EU Medical Device Regulation (MDR) and the U.S. FDA’s Digital Health Center of Excellence reflect growing institutional recognition of the need for integrated, data-driven post-market strategies. Additionally, collaborative projects such as the IMDRF (International Medical Device Regulators Forum) work on RWE and AI governance indicate a global momentum toward unified frameworks for evaluating next-generation medical technologies IMDRF. (2022).

In conclusion, the convergence of Health Economics and Artificial Intelligence within the realm of post-market clinical evaluation offers a transformative path forward. This integrative strategy aligns with contemporary trends in precision medicine, digital health, and value-based care, and positions stakeholders to deliver safer, more effective, and economically sustainable medical innovations. As regulatory landscapes mature and data ecosystems become increasingly interconnected, leveraging exchanged outcomes will be essential for realizing the full potential of medical devices in the post-market phase.

 

2. Methodology

This paper is based on a structured literature review aiming to explore the integration of Health Economics (HE) and Artificial Intelligence (AI) into the Post-Market Clinical Evaluation (PMCE) of medical devices, particularly through the lens of exchanged outcomes. The review was conducted using a qualitative synthesis of peer-reviewed journal articles, regulatory documents, and grey literature to provide a comprehensive understanding of the current landscape, key trends, and gaps in existing research.

 

2.1. Search Strategy and Selection Criteria

The literature search was conducted between January and March 2025, utilizing the following academic databases: PubMed, Scopus, Web of Science, and Google Scholar. Keywords and search terms were selected based on relevance to the core themes of the study and included Boolean combinations such as:

·        ("post-market clinical evaluation" OR "post-market surveillance" OR "PMCE")

·        AND ("medical devices")

·        AND ("health economics" OR "economic evaluation" OR "cost-effectiveness")

·        AND ("artificial intelligence" OR "machine learning")

·        AND ("real-world evidence" OR "value-based healthcare" OR "outcomes exchange")

The review included studies published between 2013 and 2024, aligning with the modern regulatory evolution marked by the EU Medical Device Regulation (MDR) and increased AI deployment in healthcare analytics. Inclusion criteria consisted of:

1)     Peer-reviewed articles, white papers, and regulatory guidance documents.

2)     Studies discussing the integration of HE and/or AI in PMCE contexts.

3)     Papers published in English.

4)     Sources providing empirical or theoretical insights into exchanged outcomes.

Exclusion criteria involved:

·        Non-English language publications.

·        Studies unrelated to medical devices or lacking relevance to HE/AI themes.

·        Duplicates and pre-2013 publications unless deemed foundational.

After initial screening, 78 documents were shortlisted, of which 42 met the final inclusion criteria following full-text review. Reference chaining was also employed to capture relevant papers cited within selected documents.

 

3. Qualitative Analysis Approach

A thematic content analysis was used to synthesize qualitative data across selected sources. The process involved coding textual content from articles using NVivo software (version 14), categorizing findings into pre-defined and emerging themes. These included:

·        The role of HE in post-market decision-making.

·        The contribution of AI in enhancing RWE and outcome measurement.

·        Challenges and facilitators in integrating AI and HE into regulatory practice.

·        Case studies demonstrating exchanged outcomes in PMCE.

To ensure rigor and minimize bias, two independent reviewers performed coding, with discrepancies resolved through discussion and consensus. Triangulation with regulatory documents was also applied to validate thematic consistency.

 

4. Research Limitations

While this review provides valuable insights, several limitations must be acknowledged. First, publication bias may affect the comprehensiveness of findings, as positive results are more likely to be published than null or negative outcomes Song et al., 2010. Second, language restriction to English may have excluded relevant studies from non-English speaking countries, potentially limiting global applicability.

Additionally, the review did not include quantitative meta-analysis, due to heterogeneity in study designs, outcome measures, and methodologies. This restricts the ability to draw firm statistical conclusions regarding the efficacy of AI and HE integration in PMCE. Furthermore, the rapidly evolving nature of AI technologies means that findings could quickly become outdated, necessitating ongoing research and review.

 

5. Ethical Considerations

As this study did not involve human participants, clinical trials, or patient-level data, formal ethical approval was not required. However, ethical rigor was maintained by adhering to transparent reporting practices, avoiding plagiarism, and ensuring accurate citation of all sources.

An ethical limitation intrinsic to the subject matter lies in the use of AI for post-market surveillance, which raises concerns about patient data privacy, algorithmic bias, and transparency Morley et al. (2020). Although these were addressed in the reviewed literature, their practical resolution remains an ongoing challenge within the industry.

 

Table 1

Table 1 Summary of Methodology
Aspect	Description
Type of Study	Structured literature review
Databases Used	PubMed, Scopus, Web of Science, Google Scholar
Keywords	PMCE, medical devices, health economics, artificial intelligence, outcomes
Years Covered	2013–2024
Inclusion Criteria	Peer-reviewed, relevant to HE/AI in PMCE, English language
Exclusion Criteria	Non-English, non-device-related, duplicates, pre-2013 unless foundational
Analysis Method	Thematic content analysis using NVivo
Documents Included	42 final sources after screening
Ethical Approval	Not required; no human subjects or sensitive data
Limitations	Publication bias, language restriction, no meta-analysis, fast-evolving field

 

6. Results

The thematic content analysis of the 42 selected publications yielded a multi-faceted view of how Health Economics (HE) and Artificial Intelligence (AI) are currently being incorporated into post-market clinical evaluation (PMCE) of medical devices. The findings are presented in five major themes: (1) trends in post-market evidence generation, (2) application of AI in real-world data (RWD) analysis, (3) role of HE in decision-making, (4) synergistic frameworks for exchanged outcomes, and (5) current challenges and gaps.

Recent literature reflects a paradigm shift in PMCE from traditional, static reporting models to dynamic, continuous data integration from real-world settings. A major trend is the transition from reactive, incident-driven surveillance to proactive risk management supported by predictive analytics Califf et al. (2020).

Several studies emphasized the growing reliance on Real-World Evidence (RWE) sources—such as electronic health records (EHRs), patient registries, and device performance databases—to assess post-market safety and effectiveness Makady et al. (2017). Regulatory initiatives such as the EU Medical Device Regulation (MDR 2017/745) and FDA’s Sentinel Initiative were frequently cited as key catalysts for this evolution.

This transformation supports a broader, outcomes-oriented view of device performance, including user-reported outcomes, quality of life metrics, and comparative effectiveness.

Artificial Intelligence—especially machine learning (ML)—has emerged as a cornerstone of modern PMCE. Over 70% of the reviewed studies discussed the use of AI for real-time signal detection, patient risk stratification, and outcome prediction in post-market surveillance.

For instance, ML algorithms have been employed to identify early safety concerns by detecting anomalies in EHR or claims data far earlier than traditional reporting systems Rajkomar et al. (2019). These tools also facilitate the automated classification of adverse events, and several papers described natural language processing (NLP) being used to mine free-text fields in clinical reports Sendak et al. (2020).

Importantly, AI was also shown to enhance patient segmentation for economic modeling. By identifying clinically and economically distinct subgroups, AI can optimize the personalization of cost-effectiveness analyses—providing valuable data for payers and manufacturers alike Wiens & Shenoy (2019). While many studies highlighted the potential of AI, several noted a lack of consensus on validation, explainability, and regulatory acceptance of adaptive learning systems, especially when these tools evolve post-market.

A clear finding from the literature is the increasing adoption of economic evaluation frameworks in PMCE to assess cost-effectiveness, budget impact, and societal value of medical devices beyond initial market entry Drummond et al. (2015). Health Economics plays a crucial role in bridging clinical performance and resource utilization. Approximately half of the studies reviewed used cost-effectiveness analysis (CEA) as a central tool in post-market assessments. Common metrics included incremental cost-effectiveness ratios (ICERs), QALYs, and net monetary benefits. Some articles proposed conditional reimbursement models linked to post-market performance, which are already operational in countries like Sweden and Germany Sorenson et al. (2008).

Additionally, health economic data were often used to update Health Technology Assessment (HTA) decisions post-launch. In several cases, post-market data resulted in adjusted pricing or coverage decisions, especially for high-cost, high-risk technologies such as implantables or digital therapeutics. A subset of studies addressed budget impact analyses (BIA) and their growing role in dynamic resource allocation based on real-world performance, highlighting a shift from one-time evaluations to continuous economic monitoring.

One of the most compelling insights from this review is the emerging use of “exchanged outcomes” as a holistic evaluation framework. This concept involves reciprocal flows of data, evidence, and value between stakeholders—clinicians, patients, regulators, payers, and manufacturers.

Several sources emphasized that integrating AI with HE within this model enables feedback loops where post-market data inform economic models, and in turn, economic outcomes guide regulatory or clinical responses Garrison et al. (2018). For example, adaptive AI systems can segment patients not only by clinical response but also by cost-effectiveness thresholds, enhancing targeted value demonstration. Meanwhile, HE models are increasingly incorporating real-time AI-generated outcomes, such as treatment adherence, time-to-event analytics, and device usability, enabling more dynamic assessment.

Such frameworks support value-based healthcare (VBHC) principles, where reimbursement and continued market access depend on demonstrated outcomes across clinical and economic domains. The literature suggests that exchanged outcomes could become a regulatory norm, especially in settings emphasizing shared risk or managed entry agreements.

Despite these promising developments, several limitations and systemic challenges were recurrent across the literature:

·        Data interoperability and fragmentation: Many healthcare systems lack unified data infrastructures, limiting the scope of AI-driven PMCE.

·        Validation of AI models: Concerns persist about model generalizability, explainability, and ethical implications such as bias or lack of transparency Morley et al. (2020).

·        Inconsistent HE methodologies: Variability in economic evaluation standards and lack of alignment with clinical endpoints reduce comparability across studies.

·        Regulatory uncertainty: Limited guidance on AI in PMCE and evolving acceptance of economic data by regulatory authorities were cited as barriers to wider adoption.

Moreover, very few studies reported on patient involvement in defining outcomes or value measures, despite growing emphasis on participatory evaluation models.

Table 2

Table 2 Summary of Thematic Results
Theme	Key Insights
Post-Market Evidence Generation	Shift from reactive to proactive surveillance using RWD and real-time analytics
AI in PMCE	Enhances data mining, risk prediction, and patient segmentation; barriers in explainability and trust
Health Economics Integration	Supports dynamic HTA, CEA, and budget impact; aligns PMCE with value-based care
Exchanged Outcomes Framework	Enables bidirectional feedback between clinical, economic, and regulatory domains
Challenges and Gaps	Data silos, inconsistent standards, ethical concerns, and regulatory ambiguity

 

7. Discussion

The integration of Health Economics (HE) and Artificial Intelligence (AI) into Post-Market Clinical Evaluation (PMCE) of medical devices represents a significant evolution in how the safety, effectiveness, and value of these technologies are assessed after regulatory approval. This discussion synthesizes the findings of the review while critically evaluating the opportunities and challenges posed by these innovations. The implications for regulators, healthcare providers, manufacturers, and payers are also explored through the lens of exchanged outcomes.

1)    Towards a Dynamic, Learning Ecosystem

One of the central insights emerging from the review is that PMCE is no longer a static, one-time obligation, but a continuous learning process informed by real-world data (RWD) and enhanced by advanced analytics. The proliferation of real-world evidence (RWE) sources—such as patient registries, claims databases, and EHRs—enables regulators and stakeholders to monitor medical device performance in routine clinical practice more effectively than ever before Makady et al. (2017).

AI is instrumental in unlocking the potential of these datasets, offering tools to automate signal detection, predict adverse events, and stratify patient outcomes in real time Rajkomar et al. (2019). This capability creates a foundation for learning health systems where PMCE feeds into ongoing clinical, regulatory, and economic decision-making. As Califf et al. (2020) suggest, such systems are particularly crucial in high-risk or fast-evolving device categories like implants, diagnostics, and AI-based software as medical devices (SaMDs).

2)    From Surveillance to Value-Based Evaluation

Traditionally, PMCE focused on safety monitoring and incident reporting, often through passive systems like manufacturer registries or voluntary physician inputs. However, there is a marked shift toward value-based healthcare (VBHC) paradigms, where the long-term value of a medical device—rather than just its initial performance—is of primary concern Porter (2010). Health Economics provides a robust framework to support this transition. Cost-effectiveness analysis (CEA), budget impact analysis (BIA), and quality-adjusted life years (QALYs) are increasingly being incorporated into post-market studies to capture economic value alongside clinical effectiveness Drummond et al. (2015). This is especially relevant in resource-constrained systems, where reimbursement and continued use may depend on the real-world cost-benefit profile.

The review highlighted examples of conditional reimbursement and managed entry agreements in countries like Sweden, Germany, and the UK, where HE outcomes post-launch influence coverage decisions Sorenson et al. (2008). Such models align with the broader goals of PMCE, ensuring that devices not only perform clinically but also justify their economic footprint.

3)    AI and HE Synergies: A New Model for Exchanged Outcomes

Perhaps the most innovative development discussed in the literature is the conceptual shift towards exchanged outcomes—a model in which data and value are continually shared among stakeholders, enabled by AI and informed by economic evaluations. This model moves beyond unidirectional data submission to regulators and instead promotes a multi-directional ecosystem of learning and accountability Garrison et al. (2018).

In such a system, AI algorithms process real-world inputs to generate dynamic outcome insights—like real-time device performance, patient-reported metrics, or adherence patterns. These insights, in turn, feed into HE models to reassess value. The economic results then help guide clinical practice, regulatory actions, and payer decisions. This loop fosters adaptive regulation and value-based reimbursement, improving transparency and stakeholder engagement. Importantly, AI also enables individualized cost-effectiveness models, segmenting patients based on clinical and economic risk profiles. This supports more nuanced PMCE by identifying subgroups that may benefit most—or least—from specific interventions, which is vital in precision medicine contexts Wiens & Shenoy (2019).

4)    Limitations and Risks of AI in Post-Market Contexts

Despite the promise of AI in PMCE, several concerns persist. A major limitation is the lack of transparency and explainability of many AI systems, particularly deep learning models, which are often described as “black boxes” Topol (2019). In a regulatory context, explainability is critical for justifying decisions that affect patient safety and public trust.

Moreover, AI systems are susceptible to bias and data quality issues. If trained on skewed or non-representative datasets, these tools may perpetuate inequalities in device surveillance or misidentify safety signals. Such biases may be especially dangerous in populations that are underrepresented in initial clinical trials or device development phases Obermeyer et al. (2019).

Additionally, the regulatory infrastructure is still catching up with the capabilities of AI. While agencies like the FDA have introduced pathways for AI-based devices and software, post-market requirements for adaptive learning systems remain ambiguous. Many tools evolve over time, raising questions about how updates are monitored and evaluated after initial clearance FDA. (2021).

5)    Challenges in Health Economics Integration

While HE offers powerful tools for post-market value assessment, their integration into PMCE is not without obstacles. One challenge is the variability in methods and standards across countries and HTA bodies. For example, some systems favor QALY-based thresholds, while others emphasize cost-benefit ratios or clinical benefit scales Angelis et al. (2018).

Furthermore, HE models often rely on assumptions that may not hold in the real world, especially in complex, multi-stakeholder environments. Incorporating real-world utility measures, patient-reported outcomes, and indirect costs into these models remains difficult due to data limitations and lack of consensus on appropriate measures Tarricone et al. (2020).

Lastly, economic data is often viewed as secondary by regulators focused primarily on safety and efficacy. There remains a gap in aligning clinical and economic outcomes, which undermines the full potential of integrated PMCE.

6)    Ethical, Legal, and Social Implications (ELSI)

The use of AI and HE in PMCE raises important ethical and legal questions. Patient data privacy is a major concern, especially in light of evolving data protection regulations such as the General Data Protection Regulation (GDPR) in Europe. Ensuring informed consent, data security, and transparency in how patient information is used is essential for ethical AI integration Morley et al. (2020).

There is also a need to address algorithmic accountability, particularly when AI is used to make decisions that could influence access to care or reimbursement. Stakeholders must establish clear governance frameworks to manage these risks and ensure that innovations in PMCE enhance, rather than compromise, patient equity and safety.

 

8. Conclusion

This review affirms that the integration of AI and HE into PMCE presents a powerful opportunity to enhance the safety, efficiency, and value assessment of medical devices in real-world settings. By facilitating continuous evidence generation and feedback across stakeholders, these tools support a paradigm of exchanged outcomes that aligns with both regulatory oversight and value-based healthcare. However, realizing this vision requires concerted efforts to overcome technical, methodological, ethical, and regulatory barriers. Stakeholders must collaborate to establish standards for AI validation, economic evaluation integration, and ethical governance. Only then can the full potential of these innovations be realized in delivering better patient outcomes and more sustainable healthcare systems.

 

CONFLICT OF INTERESTS

None. 

 

ACKNOWLEDGMENTS

None.

 

REFERENCES

Angelis, A., Kanavos, P., & Montibeller, G. (2018). Resource Allocation and Priority Setting in Health Care: A Multi-Criteria Decision Analysis Problem of Equity, Efficiency and Access. Expert Review of Pharmacoeconomics & Outcomes Research, 18(1), 5–12.

Bano, A., Künzler, J., Wehrli, F., Kastrati, L., Rivero, T., Llane, A., Valz Gris, A., Fraser, A. G., Stettler, C., Hovorka, R., Laimer, M., Bally, L., & CORE‐MD investigators. (2024). Clinical Evidence for High-Risk Ce-Marked Medical Devices for Glucose Management: A Systematic Review and Meta-Analysis. Diabetes Obesity And Metabolism, 26(10), 4753–4766.

Califf, R. M., Hernandez, A. F., & Landray, M. (2020). Weighing the Benefits and Risks of Medical Devices: The Importance of Real-World Evidence. JAMA, 323(6), 495–496.

Chernyshov, P. V., Zouboulis, C. C., Tomas‐Aragones, L., et al. (2018). Quality of Life Measurement in Acne: Position Paper of the EADV Task Forces. Journal of the European Academy of Dermatology and Venereology, 32(2), 194–208.

Drummond, M., Sculpher, M., Claxton, K., Stoddart, G., & Torrance, G. (2015). Methods for the Economic Evaluation of Health Care Programmes (4th ed.). Oxford University Press.

Elshikh, M. S., Alwahibi, M. S., Alexiou, A., Papadakis, M., et al. (2024). Soil Pollution: An Agricultural and Environmental Problem with Nanotechnological Remediation Opportunities and Challenges. Discover [Journal Name]. Springer.

Esteva, A., Robicquet, A., Ramsundar, B., et al. (2019). A Guide to Deep Learning in Healthcare. Nature Medicine, 25(1), 24–29.

European Commission. (2020). MDR - Medical Device Regulation (EU) 2017/745.

European Medicines Agency (EMA). (2021). Reflection Paper on the use of Artificial Intelligence (AI) in the Medicinal Product Lifecycle.

FDA. (2021). ARtificial Intelligence and Machine Learning in Software as a Medical Device.

Food & Drug Administration, HHS. (2018). Medical Devices; Exemption from Premarket Notification, Class II devices, over-the-counter denture repair kit. Federal Register, 83(50), 11144–11145.

Galli, V., Golia D'Augè, T., DI Pierro, F., et al. (2024). Safety and Efficacy of a class II Medical Device Based on Plant Extracts in Vulvar and Vaginal Atrophy. Minerva Obstetrics and Gynecology, 76(4), 343–352.

Garrison, L. P., Neumann, P. J., Willke, R. J., et al. (2018). A Health Economics Approach to US Value Assessment Frameworks. Value in Health, 21(2), 161–165.

Guney, T., & Yazici, A. R. (2018). 24-Month Clinical Evaluation of Different Bulk-Fill Restorative Resins in Class II restorations. Medical Devices (Auckland), 83(108), 25910–5.

Hoshino, I. A., Fraga Briso, A. L., Bueno Esteves, L. M., et al. (2022). Clinical Trial of Class II Restorations using Flowable Bulk-Fill Resin Composites: 4-year follow-up. Clinical Oral Investigations, 26(9), 5697–5710.

IMDRF. (2022). Real-World Evidence (RWE) Working Group Documents.

Kitonakis, N., & Kontis, A. (2008). The Determinants of Greek Foreign Direct Investments in Southeast European countries. Southeast European and Black Sea Studies.

Kuo, C., & Liu, X. (2020). The Integration of AI and Machine Learning in Medical Device Clinical Evaluations. Medical Device Regulatory Review, 16(2), 99–112.

Liakou, A. I., & Zouboulis, C. C. (2015). Links And Risks Associated with Psoriasis and Metabolic Syndrome. Psoriasis (Auckland), 5, 125–128.

Liakou, A. I., Theodorakis, M. J., Melnik, B. C., et al. (2013). Nutritional Clinical Studies in Dermatology. Journal of Drugs in Dermatology, 12(10), 1104–1109.

Liakou, C. I., & Plytas, M. (2025). Patient Access to High-Risk Devices. International Journal of Engineering Technology Research and Management, 9(3), 23.

Liakou, C. I., & Plytas, M. (2025). Telemedicine in Europe: Economic and Business Implications. International Journal of Innovative Research in Computer and Communication Engineering, 13(3), 2484–2493.

Liakou, C. I., & Plytas, M. (2025). The Relationship Between Leadership and Organizational Performance: A Case Study About Pfizer Pharmaceutical Inc. International Journal of Innovative Research in Computer and Communication Engineering, 13(3), 2460–2466.

Liakou, C. I., Papadakis, M., & Plytas, M. (2025). Clinical Evaluation of Immunotherapy Vaccines as Medical Devices. International Journal of Engineering Technology Research and Management, 9(3), 32.

Liakou, C. I., Papadakis, M., & Plytas, M. (2025). The Challenges for Manufacturers Under the EU MDR. International Journal of Engineering Technology Research and Management, 9(3).

Makady, A., van Veelen, A., Jonsson, P., et al. (2017). Using Real-World Data in HTA: A comparative Study. PharmacoEconomics, 35(1), 29–40.

McKendrick, P., & Turner, L. (2021). Post-Market Surveillance and Artificial Intelligence. Journal of Medical Device Regulation, 13(4), 233–241.

Mehta, K. S., & Dureja, H. (2018). Role of Artificial Intelligence in Medical Device Regulation. Regulatory Affairs Journal, 12(3), 45–53.

Morley, J., Machado, C. C. V., Burr, C., et al. (2020). The Ethics of AI in Health Care: A Mapping Review. Social Science & Medicine, 260, 113172.

Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019). Dissecting Racial Bias in an Algorithm. Science, 366(6464), 447–453.

Pantelidi, K., Agiasofitou, E., Marnelakis, I., et al. (2020). Quality of life in Hidradenitis Suppurativa Treated with Adalimumab. Experimental Dermatology, 29(Suppl 1), 50.

Pathmanathan, P., Aycock, K., Badal, A., et al. (2024). Credibility Assessment of in Silico Clinical Trials. PLoS Computational Biology, 20(8), e1012289.

Petaniti, E., Tzamaria, D., Liakou, C. I., & Plytas, M. (2025). Early-stage Medical Technology Assessment. International Journal of Science and Technology, 16(1), 1–9.

Porter, M. E. (2010). What is Value in Health Care? New England Journal of Medicine, 363(26), 2477–2481.

Rajkomar, A., Dean, J., & Kohane, I. (2019). Machine Learning in Medicine. New England Journal of Medicine, 380(14), 1347–1358.

Rotsiamis, N., Rompoti, N., Liakou, A., et al. (2025). Inflammation, Depression, and QoL in Hidradenitis Suppurativa. Archives of Dermatological Research, 317(1), 1–9.

Sendak, M. P., D’Arcy, J., Kashyap, S., et al. (2020). Translating ML products into Healthcare Delivery. Emj Innovations, 4(1), 3–10.

Sklavounos, D., Edoh, A., & Plytas, M. (2017). EWMA and CUSUM Charts for R2L Intrusion Detection. 2017 Cybersecurity and Cyberforensics Conference, 25–30.

Sorenson, C., Drummond, M., & Kanavos, P. (2008). Ensuring value for Money in Health Care. World Health Organization.

Tarricone, R., Boscolo, P. R., & Armeni, P. (2020). What Type of Evidence is Needed to Assess Medical Devices? The European Journal of Health Economics, 21(5), 701–710.

Topol, E. J. (2019). High-Performance Medicine: The Convergence of Human and AI. Nature Medicine, 25(1), 44–56.

Tsagkaris, C., Papadakis, M., et al. (2023). Is it About Time to Develop Social Surgery? American Journal of Surgery.

Tzamaria, D., Petaniti, E., Liakou, C. I., & Plytas, M. (2025). Leveraging AI and Computer Science in Primary Care. International Journal of Science and Technology, 16(1), 1–15.

U.S. Food and Drug Administration (FDA). (2021). Class II Special Controls Guidance Documents. Silver Spring (MD): FDA.

Wiens, J., & Shenoy, E. S. (2019). Machine Learning for Healthcare: Epidemiological shift. Clinical Infectious Diseases, 69(8), 1300–1303.