
Mosaic Clinical Technologies Surpasses 100,000 Daily Radiology Reports with AI-Assisted Mosaic Reporting Platform
Mosaic Clinical Technologies, a healthcare technology company focused on improving radiology reporting workflows, has announced that radiologists using its Mosaic Reporting application are now completing more than 100,000 examinations per day. The milestone highlights the growing adoption of the company’s AI-assisted reporting technology and its expanding role in supporting radiologists across large-scale clinical practices.
Mosaic Reporting is part of the company’s MosaicOS platform, which was developed to address longstanding inefficiencies and administrative challenges associated with traditional radiology reporting systems. By redesigning the reporting process around natural-language dictation and automated report structuring, Mosaic aims to reduce the time radiologists spend managing documentation and allow them to focus more directly on interpreting medical images.
The application was first used in clinical practice in April 2025 before being commercially launched to hospitals and health systems in June 2026. Since its initial clinical deployment, the technology has expanded across Radiology Partners Clinical Services (RP Clinical Services), one of the largest radiology practice organizations in the world. Mosaic Reporting now supports approximately 75% of the reports signed by RP Clinical Services radiologists on a typical day.
Radiologists have reported more than 14 million examinations using the platform to date. The latest milestone of over 100,000 daily reports reflects the scale of its implementation and the increasing number of clinicians incorporating the technology into their regular reporting activities.
Expansion Across Radiology Practices
More than 2,200 radiologists working across 74 RP Clinical Services-affiliated practices have transitioned from legacy reporting tools to Mosaic Reporting. The transition includes practices moving away from established systems such as PowerScribe 360, which is scheduled to be phased out.
Replacing established reporting software across a large clinical organization can involve substantial operational challenges. Radiologists often have different dictation habits, reporting preferences and specialty-specific requirements. A replacement platform must accommodate these differences while maintaining consistent workflows and supporting high examination volumes.
Mosaic designed its reporting application to address these requirements through a more flexible, natural-language-based approach. Rather than requiring radiologists to follow rigid dictation commands, the application allows them to describe imaging findings in their preferred order and wording.
Radiologists can dictate naturally without speaking punctuation instructions or formatting commands. The software then organizes the dictated information and places individual findings into the appropriate sections of a structured radiology report. It also drafts the report’s impression, which the radiologist can review and modify before finalizing the document.
This approach is intended to reduce the administrative work associated with arranging report content and navigating multiple software functions. By automating parts of the formatting process, Mosaic Reporting seeks to make documentation more efficient while preserving the radiologist’s control over the final clinical interpretation.
The platform’s patent-pending workflow is central to this process. It converts natural-language descriptions into organized report sections, helping radiologists maintain their preferred reporting styles without having to manually structure every element of the document.
Natural-Language Dictation Reduces Reporting Complexity
Traditional radiology reporting applications can require clinicians to follow specific commands and formatting conventions when dictating findings. These requirements may interrupt the interpretation process and create additional administrative steps.
Mosaic Reporting takes a different approach by allowing radiologists to describe findings naturally. The system processes the dictated information and places it into the appropriate sections of the report. This is designed to reduce the need for repeated corrections, manual formatting and unnecessary navigation between different parts of the application.
The company believes that minimizing these interruptions can help radiologists maintain their attention on the medical images they are reviewing. It may also support more consistent reporting practices across large organizations where clinicians have different working styles.
The application is designed to integrate into the radiologist’s existing reporting process rather than requiring a separate documentation workflow. Its growing use across RP Clinical Services demonstrates its deployment at a substantial scale, with thousands of radiologists using the same technology to report a high volume of examinations.
According to Mosaic, the platform now processes approximately 75% of the reports signed by RP Clinical Services radiologists on a typical day. This adoption level represents a significant change in the organization’s reporting infrastructure and demonstrates the company’s ability to implement its technology across multiple affiliated practices.
Automated Follow-Up Recommendations Support Clinical Consistency
In addition to workflow improvements, Mosaic Reporting is designed to support clinical quality through the integration of evidence-based Best Practice Recommendations (BPRs).
Radiology reports sometimes identify findings that require additional monitoring, clinical assessment or follow-up testing. Examples include abdominal aortic aneurysms and lung nodules. Appropriate follow-up can be important for ensuring that potentially significant findings receive attention from referring physicians and other members of the care team.
When a radiologist describes a finding that calls for follow-up, Mosaic Reporting automatically adds the relevant recommendation established by the practice. The radiologist can then review the recommendation and confirm or modify it before completing the report.
The BPR functionality is intended to reduce variability in how follow-up recommendations are communicated. By incorporating practice-specific, evidence-based guidance directly into the reporting workflow, the application can help ensure that relevant next steps are included in the report.
This feature may also help referring physicians understand what action should be considered after an imaging examination. Rather than requiring radiologists to locate and manually insert recommendation language, the software makes the applicable guidance available within the report.
The company views this capability as an important part of its broader effort to improve radiology workflows. Its strategy extends beyond accelerating documentation to supporting more standardized communication and helping practices incorporate established clinical recommendations into routine reporting.
Integration with AI-Drafted Radiology Reports
Mosaic Reporting was also developed to support the increasing use of artificial intelligence in radiology, including reports drafted by vision-language models (VLMs).
Mosaic expects AI-drafted reports to become a standard part of radiology workflows as the technology advances. To prepare for this development, the company designed its application to allow findings generated by investigational VLMs from Cognita Imaging, Inc. (Cognita), Mosaic’s advanced AI laboratory, to flow directly into the reporting environment.
This integration allows AI-generated findings to appear within the report for radiologists to review, edit and sign. The information can be incorporated without requiring clinicians to open separate application windows or copy and paste content between systems.
The integrated workflow is intended to reduce friction between AI-assisted image analysis and the radiologist’s reporting process. Instead of treating AI-generated findings as a separate output, Mosaic Reporting provides a unified environment in which the radiologist can assess the information and determine how it should be reflected in the final report.
The company reported that, through October 22, 2026, its investigational VLMs had drafted nearly 8 million examinations for radiologist review under a research protocol. The models were drafting an average of more than 37,000 examinations per day during that period.
These figures demonstrate the scale at which Mosaic and Cognita are evaluating AI-supported reporting workflows. However, the VLM-generated findings remain subject to radiologist review under the research protocol described by the company. The radiologist retains responsibility for reviewing, editing and signing the final report.
Mosaic recently announced quality and efficiency findings from its VLM research, reporting improvements in detection and reporting efficiency. The research forms part of the company’s broader efforts to evaluate how vision-language models can support radiologists and potentially improve the integration of AI into clinical reporting.
The company’s approach combines AI-generated information with a reporting application designed around the needs of practicing radiologists. This integration is intended to make it easier for clinicians to assess AI-generated findings while maintaining oversight of the final clinical documentation.
Company Leadership Highlights Adoption and Scalability
Commenting on the milestone, Mike Peresie, president of Mosaic Clinical Technologies, emphasized the pace of adoption since the platform’s initial clinical use in 2025.
“The pace we are moving at is quite staggering. In spring 2025, only about 40 RP Clinical Services radiologists were the first to read on Mosaic Reporting,” Peresie said. “Today, our radiologists signed more than 100,000 reports on it.”
He also highlighted the questions that radiology practices and health systems typically consider when replacing their reporting software. These include the complexity of implementation, the ability of the application to support large-scale demand and whether the technology can accommodate the different reporting styles used by radiologists throughout a typical working day.
According to Peresie, the adoption of Mosaic Reporting by more than 2,200 radiologists, who collectively complete over 100,000 reports per day, demonstrates the platform’s ability to address these operational requirements.
The company views the milestone as evidence that its reporting technology can support a large and diverse radiology workforce. Its expansion across 74 affiliated practices also reflects the importance of scalability when introducing new software into established clinical environments.
Radiologists Describe Workflow Benefits
Radiologists using the application have also highlighted the potential benefits of reducing interruptions during image interpretation.
Vikram Krishnasetty, MD, of Columbus Radiology, said Mosaic Reporting allows him to keep his attention on the images while reducing the need to switch between different software functions.
“Mosaic Reporting lets me keep my eyes on the images. Less context switching makes my day easier and lets me take care of more patients,” Krishnasetty said.
He also described the value of having guideline-based follow-up recommendations incorporated directly into the reporting process. When an imaging finding requires additional attention, the recommendation is already available for review, allowing the radiologist to confirm or modify it before the report is completed.
“When a finding calls for follow-up, the guideline-based recommendation is already there for me to confirm or change, so the referring physician gets clear next steps,” he added.
These comments reflect two of the principal capabilities Mosaic is emphasizing: reducing administrative interruptions during reporting and making relevant follow-up guidance easier to incorporate into clinical documentation.
Mosaic to Demonstrate Technology at RSNA 2026
The announcement comes less than two weeks before the annual meeting of the Radiological Society of North America (RSNA), scheduled for November 29 through December 3, 2026, at McCormick Place in Chicago.
Mosaic plans to demonstrate Mosaic Reporting at booth 5304 in the South Hall AI Showcase. The company will use the event to introduce its reporting technology to radiology practice leaders, hospital executives, health system representatives and other attendees interested in developments in radiology workflows and artificial intelligence.
Practice and health system leaders will be able to arrange demonstrations, review the booth presentation schedule and access the company’s latest research through the Mosaic RSNA Event Hub.
The event will provide an opportunity for attendees to examine how the platform combines natural-language dictation, automated report structuring, clinical follow-up recommendations and AI-generated findings within a single reporting environment.
Advancing AI-Assisted Radiology Workflows
Mosaic’s achievement of more than 100,000 daily reports marks a significant stage in the company’s expansion of AI-assisted radiology reporting. With more than 2,200 radiologists using the application across 74 RP Clinical Services-affiliated practices and over 14 million examinations reported through the technology, the company has demonstrated substantial adoption within a large radiology organization.
The platform’s development reflects a broader effort to address the administrative demands of radiology reporting while preparing for the growing role of artificial intelligence in clinical workflows. Its natural-language dictation capabilities, automated report structuring and integrated Best Practice Recommendations are designed to support radiologists throughout the reporting process.
At the same time, the integration of investigational vision-language models from Cognita demonstrates Mosaic’s efforts to connect AI-generated findings directly with the radiologist’s existing documentation workflow. The company’s reported research activity, including nearly 8 million AI-drafted examinations through October 22, 2026, highlights the scale of its ongoing evaluation.
As Mosaic prepares to showcase its technology at RSNA 2026, the company is positioning Mosaic Reporting as a platform designed to support both current reporting requirements and emerging AI-assisted approaches. Its latest milestone underscores the importance of scalable implementation, clinician oversight and workflow integration as radiology organizations consider the next generation of reporting technology.
About Mosaic Clinical Technologies, Inc.
Mosaic Clinical Technologies, Inc., a wholly owned subsidiary of Radiology Partners, Inc., develops advanced technology and proprietary vision-language AI models for medical imaging, built in-house in collaboration with radiologists in the Radiology Partners Clinical Services practice. Its MosaicOS platform brings diagnostic tools, cutting-edge AI and radiology workflows together in a cloud-native operating system designed to support radiologists and the healthcare organizations they serve. Learn more at MosaicClinical.ai and connect with us on LinkedIn and YouTube.
About Radiology Partners, Inc.
Radiology Partners, Inc., through RP Clinical Services and our managed practices, together with Mosaic Clinical Technologies, Inc., provides radiology services and advanced technology to more than 3,400 hospitals and other healthcare facilities. Our network includes more than 4,000 radiologists who interpret more than 55 million imaging studies each year. Learn more at radpartners.com and connect with us on LinkedIn, X, Instagram and YouTube.
About Cognita Imaging, Inc.
Cognita Imaging, Inc., Mosaic Clinical Technologies, Inc.’s advanced AI lab, develops proprietary vision-language foundation models and other generative AI capabilities for radiology. Our work is focused on helping radiologists deliver more accurate, efficient care and expanding access to high-quality healthcare. We seek to elevate the performance of all radiologists to the level of the best expert subspecialists and, longer term, to empower radiologists with digital imaging information not easily perceptible to the human eye.



