The emergence of real-world imaging data (RWiD) is causing a rapid change in the innovation landscape of medical devices. The focus is on how RWiD demonstrates device performance, patient safety, and regulatory compliance, rather than on why device manufacturers are a critical component of healthcare. The process has undergone genuine changes as a result of its integration.

In the past, limited imaging evidence and controlled clinical trials were used to evaluate medical devices. Real-world data (RWD) can include datasets beyond electronic health records (EHR), such as results of pathology tests, diagnostic imaging, genomic, and/or other “omic” data. Regarding RWiD, a robust dataset can be made accessible for novel clinical studies, allowing for early-stage analytics, feasibility assessments, and exploratory analyses.

Digital imaging, advanced data management systems, and integration of AI/ML have facilitated widespread use of large collections of high-quality images linked to clinical outcomes. Device manufacturers are under pressure from organizations like the FDA to demonstrate product safety and effectiveness in the real-world clinical practice as opposed to being described in idealized studies. This change puts real-world imaging data at the center of surveillance and medical device innovation.

Why are medical device companies shifting their focus to RWiD? 

While imaging technologies are not new, the realization and incorporation of RWiD with clinical patient data elements is becoming a priority for medical device companies. The reasons for this shift are the following: 

Development & validation

The current health innovation landscape revolves around careful development and true validation of products that support patient recovery and improve patient outcomes. RWiD is important for training and validating AI algorithms to maintain stable performance across patient populations, including underrepresented populations. Large, multi-site imaging datasets enable medical device companies to link patient outcomes to device usage and identify target and high-risk patient populations. Leveraging datasets that include both patient clinical and imaging data decreases the chances of device failures or incorrect diagnoses in real-world settings substantially.

Reducing bias & enhancing equity

RWiD often reflects real-life diversity of patients, helping device makers and AI developers identify, mitigate, and actively reduce bias from the earliest phases of product development. This keeps devices from failing or even harming underrepresented groups, and advances equity in diagnostic and patient care outcomes.

Readiness for regulation

Integrating imaging and clinical data reinforces clinical trials, expedites regulatory approval, and increases confidence in device safety and efficacy in multi-center registries. RWiD can assist with producing thorough and credible submissions to regulatory bodies (FDA, EMA, TGA). It allows post-market surveillance, as required by the MDR, demonstrating continuous evidence of device safety and performance through Post-Market Clinical Follow-up (PMCF) with real-world data.

Improved trial efficiency

RWiD uses and restores genuine real-world clinical scenarios with both broadly and unrestricted patient populations, providing the generalizability with both their data set and data outcome variability. RWiD effectively combines radiomic data from multiple modalities, aggregating patient outcomes early in trials while also advancing biomarker indicators throughout the pharmaceutical lifecycle.

Quicker timelines for R&D

Access to RWiD can accelerate medical device R&D by giving innovators access to shorter recruitment times, large, diverse, and longitudinal imaging cohorts. Compared to traditional datasets, medical device innovators can lower costs and time to market by quickly testing hypotheses, simulating real-world performance, and even conducting embedded pragmatic trials.

Using device and supply chain data to enhance real-world insights

The integration of RWiD with supply chain and EHR data provides insight into patterns of device utilization and patient outcomes. This integration connects the gap to show not just the theoretical performance of devices but real-world effectiveness. This integrated analysis allows for retrospective studies and prospective studies that aim to understand and improve device design and adherence to device usage protocols.

Real-world imaging data applications for medical devices 

Risk stratification for AI in radiology cohorts

The accuracy of AI models used for risk stratification in clinical settings can be greatly improved by using RWiD. For example, a top imaging analytics platform was able to stratify lung cancer risk by using millions of real-world CT scans. The model was initially trained on a small trial dataset that lacked the diversity of real-world cases. Inclusion of diverse patient demographics and imaging circumstances in datasets allowed the AI model to identify early-stage nodules and risk factors that may have been missed in traditional clinical trials. By applying data, clinicians were able to make earlier interventions and provide better outcomes with more personalized care.

Regulatory submission with selected clinical and imaging information

Clinical data and real-world imaging can speed up the regulatory approval process for medical devices. A medical device company obtained FDA approval for a new diagnostic device using multi-site imaging data and clinical records from various populations. In contrast to the controlled environments of clinical trials, this large data set demonstrated the efficacy and safety of the device. By providing real-world evidence in their submission, the company was able to accelerate the FDA approval process and gain approval for an expansion of the device's labeling. This example demonstrates how real-world imaging data can provide stronger regulatory submissions and a faster timeline for market access.

Real-world imaging information for a thorough assessment of device performance

A comprehensive picture of a device's performance in clinical practice can be obtained by combining real-world imaging data with clinical notes and patient outcomes. For example, a medical device manufacturer linked imaging data from multiple sites with patient EHRs and long-term outcomes to demonstrate the true value of their product. Payers and regulators received strong evidence of the device's safety and effectiveness in practical situations thanks to this "360-degree" evidence. By supporting post-market surveillance and securing reimbursement, the integration of imaging with clinical notes ensured continued safety monitoring and wider device adoption.

Segmed’s advantage in providing MedTech businesses with regulatory-grade image data

In-circumstance imaging data has become a competitive, operational, and regulatory requirement for today's medical device companies, instead of simply a "nice-to-have." Leading innovators are distinguished by their access to high-quality RWiD, like Segmed's, in which imaging technology, AI/ML, and regulatory expectations are evolving together. These datasets offer a unique way to create safer, better, and innovative health tech products for patients across the globe.

RWiD, a unique source of data that offers visual data providing patient safety, treatment effectiveness, and device performance data, is now critical in the medical device industry. RWiD makes it possible to detect conditions earlier, evaluate treatments thoroughly, and provide high-resolution, objective evidence.

At Segmed, we offer access to over 100 million high-quality, regulatory-grade imaging datasets from diverse geographies across multiple continents. Our extensive, tokenized datasets help MedTech companies enhance post-market surveillance, maintain regulatory compliance, and derive actionable insights that optimize device safety and performance.

‍Connect with us to find out how our high-quality, diverse, tokenized imaging datasets can support a myriad of research and development for medical devices R&D.