Over the past few decades, there has been a significant evolution in medical devices in the healthcare industry. Medical devices are an important aspect of healthcare as they help diagnose, monitor, and treat a wide range of medical conditions. The use of medical devices helps to improve the accuracy and effectiveness of medical treatment, leading to better patient outcomes. The medical device industry is constantly evolving and innovating, with new technologies and devices being developed all the time. This promotes advancement in the medical field and guarantees that patients have access to the newest procedures and equipment. Medical devices must undergo post-market surveillance (PMS) to guarantee their continued efficacy, performance, and safety after being on sale. It assists in detecting long-term hazards, uncommon adverse events, and device malfunctions that pre-market trials could overlook (1). By facilitating real-time signal detection, automated analysis of sizable datasets, and early detection of safety trends from adverse event databases, electronic health records (EHRs), and user input, artificial intelligence (AI) improves PMS (2). A more resilient and adaptable safety ecosystem can be achieved by using machine-learning models to forecast device failures, enhance vigilance reporting, and assist regulatory decisions (3).

AI can completely change the way we manage the operational facets of healthcare delivery, diagnose illnesses, customize therapies for each patient, and track health status in real time. AI-powered diagnostic technologies, for example, are capable of precisely analyzing medical images and frequently spotting details that human eyes might miss. Patient outcomes are greatly impacted by the earlier and more precise diagnoses that result from this precision. Similar to this, therapy customization is a step toward genuinely individualized medicine as AI algorithms can search through enormous databases to find trends and forecast which medicines will work best for particular patient profiles. Furthermore, wearable technology and remote monitoring systems provide ongoing patient health oversight, facilitating prompt treatments and lowering readmissions to hospitals. This is just one example of how AI is being used in patient monitoring. From appointment scheduling to hospital workflow optimization, AI can improve healthcare delivery efficiency and patient satisfaction (4–6).

Due to its distinct features, including data reliance, algorithmic bias, performance drift, and changing behavior in real-world scenarios, PMS is crucial for AI-driven medical devices. AI-enabled systems, such as diagnostic imaging tools or predictive algorithms, are susceptible to changes in patient demographics, disease prevalence, or clinical workflows that can cause data drift, which can impair accuracy over time and jeopardize patient safety and efficacy (7). The benefit-risk profile of the device is maintained throughout its lifecycle thanks to PMS, which allows for continuous monitoring to proactively detect these problems, identify developing dangers, subgroup underperformance, or latent biases that might not be apparent during pre-market validation.

This necessity is emphasized by regulatory bodies: With regard to AI/ML-enabled devices, the U.S. Food and Drug Administration (FDA) uses a total product lifecycle (TPLC) strategy, which mandates that manufacturers use PMS plans to monitor performance in the real world, control drift, and reduce hazards after approval (8). Quality assurance and multi-site assessments are facilitated by tools for identifying input/output changes and performance variances (7). To ensure continued safety, detect unidentified side effects, and address systematic misuse, PMS, including post-market clinical follow-up (PMCF), is required in the EU under the medical device regulation (MDR) and AI Act. Radiologists and providers also help to monitor AI systems in use (9). A strong PMS encourages innovation through empirical data and supports prompt remedial measures, like improved training or design revisions.

Effective PMS, in the end, facilitates the transition from controlled pre-market testing to dynamic clinical deployment, protecting patients, upholding regulatory compliance, and facilitating ongoing advancements in AI-driven healthcare technology through data-driven upgrades.

According to Kanemasa et al. (10), post-market surveillance is the continuous evaluation of a device’s performance following its release to the general public. Important information is gathered during this procedure to help identify possible flaws that might not have been noticeable during pre-market evaluations.

A post-market surveillance plan’s goal is to swiftly detect any problems with device usage or design defects, which includes the following:

Conventional approaches placed a strong emphasis on vigilance and adverse event reporting. Manufacturers kept complaint handling systems in place to record user comments, malfunctions, and failures and then examine them for patterns (15). Global vigilance initiatives made cross-border report sharing easier. Hospital-based networks, such as the medical product safety network (MedSun) in the US, which collected user facility reports, and volunteer MedWatch submissions were additional components (13). Reporting was often enhanced by retrospective examination of pre-existing data, such as medical records or claims; but formal prospective studies were uncommon unless required (for example, under FDA Section 522 for high-risk devices) (16).

Although they offer wide coverage and are reasonably priced, passive systems have some serious disadvantages. Lack of awareness, time restraints, or liability concerns frequently result in underreporting, which delays signal discovery and produces incomplete data (13). Reports frequently lack specificity, determining causality can be difficult, and long-term patterns or uncommon occurrences may go overlooked. Because it depends on spontaneous submissions, standard PMS is reactive rather than preventive, which limits the ability to identify systemic problems or new hazards early on (17).

In the past, agencies such as the FDA and EU competent authorities have constructed surveillance around these passive methods, adding recalls, corrections, and a few active studies when indications appeared. Particularly after MDR, which requires systematic, active PMS strategies, constraints led to shifts toward proactive approaches like registries and real-world evidence (RWE) (18). To sum up, traditional PMS was successful in capturing reported occurrences; but, it also brought attention to the necessity of more thorough, proactive monitoring to better safeguard public health in a changing medical device environment.

Under contemporary laws such as the FDA’s TPLC framework and the EU medical device regulation (MDR 2017/745), PMS for medical devices has moved toward proactive, methodical, and data-driven techniques. To identify hazards early, update benefit-risk evaluations, and promote advancements during the device lifecycle, these approaches prioritize active data collection, RWE, and ongoing monitoring (7, 15).

In order to comply with modern PMS, producers must actively and methodically collect data instead of depending just on impromptu reports. This includes literature studies, user surveys, comments from medical experts, publicly accessible statistics on comparable devices, and trend analysis. A proactive approach to gathering, documenting, and evaluating quality, performance, and safety data is required by the EU MDR (Article 83), allowing for comparisons with comparable devices and prompt remedial measures (18, 19). Through systematic evaluations, support activities, and controlled monitoring throughout user training, proactive elements foresee problems.

In order to verify continued safety, performance, and clinical benefit, PMCF—a proactive, ongoing task to gather clinical data from real-world use—is a crucial contemporary component (Annex XIV, Part B, MDR). By using specialized research, registries, and polls, PMCF techniques produce high-quality RWE that enhances pre-market data. PMCF for higher-risk devices guarantees performance across a range of demographics, long-term results, and the identification of new hazards (20, 21). In order to promote evidence-based updates to clinical evaluations, registries offer longitudinal data from sizable cohorts.

For thorough insights, contemporary methods use RWE from patient registries, claims data, EHRs, and device-generated data. This makes it possible to identify trends, analyze subgroups, and spot signals early on. In order to address particular issues such as algorithmic degradation, tools for AI-enabled devices track input/output changes, data drift, and performance variances across locations (7, 9). Analyzing large datasets more efficiently is made possible by statistical techniques and AI-assisted analysis.

In accordance with EU MDR, manufacturers generate structured reports that summarize data analysis, trends, and actions. These reports are called PSURs for higher-risk (Class IIa, IIb, and III) devices and PMSRs for lower-risk (Class I) devices (22). PMS is included into lifetime management by the FDA’s TPLC approach, which offers guidelines for tracking AI/ML products using predetermined metrics and flexible strategies (8). While still crucial, proactive components are added to vigilance reporting.

By encouraging early risk mitigation, design improvements, and CAPAs, these techniques enhance patient safety and regulatory compliance. Regular reviews and changes to technical documentation are part of PMS’s integration with QMS (20). Proactive surveillance fills the pre- and post-market gaps for dynamic technologies like AI, fostering innovation while reducing risks. By utilizing real-world data and cutting-edge solutions for ongoing device safety and performance, modern PMS essentially shifts surveillance from reactive to predictive.

Following its introduction to the market, medical devices are monitored for efficacy, safety, and performance through post-market surveillance, or PMS. The FDA’s TPLC framework and the EU Medical Device Regulation (MDR 2017/745) require proactive, systematic, and data-driven methods that allow for earlier risk detection and continuous improvement. In contrast, traditional approaches were primarily passive and reactive, depending on impromptu reports (15, 16).

Conventional PMS was critiqued for underreporting and sluggish reaction times, frequently overlooking uncommon or chronic problems, despite its effectiveness in capturing reported occurrences (13). The FDA’s move toward active surveillance through tools like National Evaluation System for Health Technology (NEST) and the EU MDR’s emphasis on lifecycle monitoring are two examples of how modern PMS fills these gaps through regulatory evolution, allowing for better real-world performance assessment (23). In the case of cutting-edge gadgets like AI-enabled ones, where proactive techniques identify performance changes early, this shift represents a larger commitment to patient safety (Table 1).

A comprehensive system of performance characterization, monitoring, identification, collection, reporting, and analysis of any adverse event brought on by medical devices is known as metrovigilance. India’s medical device market is worth USD 3.1 billion. But for a long time, there was no effective mechanism in place to keep an eye on the negative outcomes associated with medical gadgets (24). At first, the Drugs and Cosmetics Act of 1940 and the Rules of 1945 governed the marketed medical devices. Later in 2017, the Medical Device Rules, 2017 (25) were formulated as a unique regulation to govern the medical devices that are offered for sale in India.

By broadening regulatory monitoring beyond conventional classifications to encompass all classes of medical equipment, categorized by risk, this reform brought Indian regulation closer to international standards. Following global best practices, the regulations standardized post-market surveillance, required reporting of adverse occurrences, and Materiovigilance (device safety monitoring). Significantly, the MDR 2017 acknowledged the need for increased regulatory infrastructure and capacity building to maintain device safety while promoting innovation and market expansion, and it placed a strong emphasis on the creation of notified organizations for conformity assessment. With the creation of the Materiovigilance Program of India (MvPI) by the Indian government, this change demonstrated India’s dedication to enhancing patient safety and device quality through organized, transparent, and enforced regulatory systems (26, 27).

Medical devices in India are governed by the DCA 1940 and Rules 1945. To import, produce, sell, and distribute medical devices, the Indian government implemented the Medical Devices Rule in 2017 after consulting with the drug technical advisory board (DTAB). Notification of this regulation was sent on January 31, 2017, and it became operative on January 1, 2018 (2, 5). On July 6, 2015, the Drug Controller General of India DCG (I) introduced the MvPI at the Indian Pharmacopoeia Commission (IPC) in Ghaziabad. The main goal of this program is to educate healthcare workers about the importance of “medical device adverse events” (MDAE) (28).

Additionally, it highlights a device’s benefit-risk profile, monitors MDAE, and shares these results with all pertinent parties. Overseeing adverse occurrences of medical devices identified in the Indian public is the responsibility of IPC, the National Coordination Center (NCC) for MvPI. MvPI is regulated by the Central Drug Standard Control Organization (29), and the National Health System Resource Center (NHSRC) provides technical assistance. The Sree Chitra Institute for Medical Sciences and Technology (SCTIMST) functions as a “National Collaborating Center.” To ensure that a case is complete, 26 Medical Device Monitoring Centers (MDMCs) and Adverse Drug Reaction Monitoring Centers (AMCs) have been established. These centers examine MDAE data and forward them to the NCC (30).

A systematic PMS system is established by India’s MDR 2017 and is essential for tracking the effectiveness and safety of medical devices following their introduction to the market. The MvPI, a nationwide program run by the IPC and the Central Drugs Standard Control Organization (CDSCO), collaborates with the PMS system to coordinate vigilance and adverse event reporting across healthcare settings.

There is a need to have a strong PMS to assure continuous patient safety due to modernized global healthcare system which is evolving rapidly with respect to medical devices, from basic tools to advanced robotic systems. The FDA’s Framework, EU’s MDR and India’s Materiovigilance Program under the 2017 Medical Device Rule supports regular monitoring and obtaining approval from these bodies. PMS is being revolutionized with AI which uses machine learning, natural language processing, and deep learning which significantly improves shareholders’ confidence and patient safety by enabling quicker signal detection, precise real-world data analysis, unbiased and early risk identification.

A standardized, transparent, and AI-enhanced PMS systems requires coordinated local governance comprising manufacturers, healthcare providers, and regulators to optimize these benefits. With new medical innovations developing at a faster pace, stricter laws with AI becomes very important to maintain safe, competent, and patient-oriented healthcare.