Subtle Medical

Subtle Medical is a healthcare technology company that develops artificial intelligence-powered software for improving medical imaging. Founded in 2017 as a Stanford University spinout, the company focuses on making MRI, PET, and CT scans faster and clearer while helping hospitals and diagnostic imaging centres make better use of their existing equipment. Its technology uses deep learning to process medical images, reduce visual noise, improve image sharpness, and generate additional imaging contrasts. These capabilities are intended to support radiologists, imaging specialists, and healthcare providers by improving image quality and reducing the time patients spend undergoing certain examinations.

The company was founded by Dr. Greg Zaharchuk, a neuroradiologist and Stanford University professor, and Dr. Enhao Gong, a deep learning scientist. Their work brought together medical imaging expertise and machine learning research to address challenges associated with image quality, scan duration, and access to diagnostic imaging. Subtle Medical subsequently developed software products designed to work with existing imaging infrastructure rather than requiring healthcare providers to replace their scanners. Its solutions are vendor-neutral, allowing facilities to introduce image enhancement technology across compatible equipment from different manufacturers.

Subtle Medical’s product portfolio covers several imaging modalities, with MRI forming a major part of its technology offering. One of its central product packages is Subtle-ELITE, which combines SubtleHD (MR), SubtleSYNTH, and SubtleALIGN. These applications address different stages of MRI acquisition and image processing. Together, they are designed to improve image quality, shorten examination times, and reduce repetitive adjustments in radiology workflows. The package is particularly relevant to imaging centres that want to increase scanner capacity and improve patient throughput without immediately investing in additional MRI equipment.

SubtleHD (MR) is an AI-powered MRI image enhancement application that uses denoising and sharpening to improve the clarity of acquired images. It is designed to process images obtained through accelerated MRI protocols, helping radiology departments produce clearer images while reducing acquisition time. The company reports that some MRI sequences can achieve scan-time reductions of up to 80%, depending on the protocol and clinical application. This can help facilities make more efficient use of scanner schedules, create additional appointment capacity, and reduce the amount of time patients spend inside the scanner. The application is designed to work with existing MRI equipment from different manufacturers, including older systems that may not support the latest manufacturer-specific upgrades.

SubtleSYNTH focuses on generating synthetic MRI contrasts from images acquired through existing sequences. Rather than requiring every contrast to be acquired separately, the software uses deep learning to produce additional image information from available MRI data. One of its highlighted applications is the generation of synthetic STIR images for spine imaging. Subtle Medical reports that this approach can provide 100% acceleration for the supported STIR acquisition by generating the corresponding images without requiring a separate STIR scan. This can reduce examination time and help radiology departments streamline protocols. Its commercial availability and supported anatomical applications should be checked with the company, as the product’s capabilities may expand over time.

SubtleALIGN is designed to automate brain MRI alignment. It uses machine learning to reorient brain images into consistent anatomical positions, including axial, coronal, and sagittal orientations. This can reduce the need for manual adjustments and help make image comparisons more consistent, particularly when radiologists are assessing changes over time. Consistent image orientation is important in neurological examinations because differences in positioning can make longitudinal comparisons more difficult. SubtleALIGN currently supports 3D brain sequences, making it a specialised tool for compatible brain MRI workflows rather than a general-purpose image alignment application.

The company also offers SubtleHD (PET), an AI-powered image enhancement solution for positron emission tomography. PET imaging is used in several areas of medicine, including oncology, neurology, and cardiac imaging. Image quality can be affected by acquisition time and the amount of detected signal available during a scan. SubtleHD (PET) uses deep learning-based denoising and sharpening to improve the appearance of low-count PET images. The technology is intended to help imaging facilities obtain useful image quality from shorter acquisitions and make more efficient use of existing PET, PET/CT, and PET/MR systems. Subtle Medical reports that supported workflows can achieve scan-time reductions of up to 75%. The company announced FDA clearance for this product in May 2026 and states that it supports FDA-approved radiotracers.

Subtle Medical has expanded its portfolio into computed tomography through SubtleHD (CT). This application uses AI-powered image enhancement to reduce image noise and improve low-contrast detectability. Its purpose is to help radiologists assess structures that may be difficult to distinguish in noisy CT images. The technology is designed to work across different scanner types and can be used with existing CT and PET/CT equipment. It may be particularly useful for facilities operating mixed scanner fleets or older systems that do not have access to the newest manufacturer-specific enhancement technologies. SubtleHD (CT) received FDA clearance in June 2026, marking the company’s expansion into another major medical imaging modality.

Another technology associated with Subtle Medical is SubtleGAD, which is being developed for contrast-enhanced MRI. It uses deep learning to generate post-contrast imaging information from lower-dose contrast-enhanced scans. The intended goal is to reduce the amount of gadolinium-based contrast agent required while maintaining useful image quality. The company has reported research involving substantial reductions in contrast-agent dosage, including work conducted with Bayer. However, SubtleGAD should be treated as a development and clinical research technology rather than assumed to be a generally available commercial product. Healthcare providers should confirm its current development, regulatory, and availability status before considering it for clinical use.

Subtle Medical also has a partnership with Bracco Imaging involving AiMIFY. This application uses Subtle’s technology to enhance the visibility of brain lesions in contrast-enhanced MRI. It is designed to amplify the appearance of MRI contrast and help radiologists identify relevant findings. AiMIFY is distributed by Bracco, so it is best understood as a partner solution powered by Subtle technology rather than simply another standalone product sold directly under the Subtle Medical name. The partnership demonstrates how the company’s image-processing technology can be incorporated into broader medical imaging products.

A major characteristic of Subtle Medical’s approach is its focus on integrating with existing medical imaging environments. Its software is designed to work with scanners from different manufacturers and fit into established radiology workflows. This is important for hospitals and imaging centres that operate equipment from multiple vendors and want to avoid maintaining completely separate enhancement systems for each scanner brand. The company’s technical materials describe DICOM-based workflows and supporting software components for connecting imaging equipment, processing applications, and image-management environments. Its products are intended to work alongside the systems healthcare providers already use, subject to the compatibility and configuration requirements of each product.

The potential applications of Subtle Medical’s technology extend across several medical specialties. MRI enhancement can support neurological, musculoskeletal, and spine imaging, while PET enhancement can contribute to workflows used in oncology and other areas where functional imaging is important. CT image enhancement can support a broad range of examinations involving the chest, abdomen, pelvis, spine, and other anatomical regions. The technology does not replace the radiologist or independently establish a final diagnosis. Instead, it processes medical images to support image quality and acquisition efficiency within clinical imaging workflows. Qualified healthcare professionals remain responsible for interpreting results and making clinical decisions.

Subtle Medical also places emphasis on medical device quality, regulatory compliance, and responsible AI development. Its compliance programme includes a quality management system for the design, development, manufacture, installation, and servicing of medical imaging software. The company reports certifications and compliance frameworks that include ISO 13485 for medical device quality management, ISO/IEC 42001 for AI management systems, and relevant medical device regulatory requirements. It also reports SOC 2 Type II and HIPAA-related security audits. These measures are important because medical imaging software operates in environments where patient information, system reliability, and clinical safety must be carefully managed.

Regulatory clearance is an important consideration when evaluating the company’s products. Subtle Medical reports multiple FDA clearances, but the regulatory status of individual applications differs. A clearance for one product should not be interpreted as approval for every product in the portfolio or for every possible clinical use. Availability can also differ between countries because medical device registration and distribution requirements vary by market. Healthcare institutions should therefore confirm the applicable product version, intended use, regulatory status, and local availability before deployment.

Subtle Medical primarily serves hospitals, radiology departments, diagnostic imaging centres, medical imaging networks, and healthcare organisations that operate MRI, PET, or CT scanners. Its solutions are especially relevant to facilities that want to improve imaging capacity, extend the useful life of existing equipment, and reduce delays caused by lengthy examinations. The company has continued expanding its commercial reach through partnerships and international distribution arrangements. It has also reported deployment across more than 1,300 scanners in multiple countries and investment in further product development and global expansion.

The company’s business is focused on specialised healthcare software rather than consumer image editing. Its applications are developed for clinical imaging environments and require appropriate technical integration, equipment compatibility, and professional oversight. Subtle Medical is therefore most suitable for healthcare organisations with established radiology operations rather than individuals looking for a general-purpose AI image enhancement application.

Subtle Medical has also invested in clinical research and product development. Its work combines medical imaging knowledge with deep learning methods, and the company reports peer-reviewed publications, patents, and regulatory milestones associated with its technology. Research and clinical validation are particularly important in this field because an image enhancement system must preserve medically relevant information rather than simply produce visually attractive images. The value of an enhancement tool depends on how reliably it performs within its intended clinical application and how well it fits the imaging protocols used by healthcare professionals.

Pricing

Subtle Medical uses a business-to-business commercial model aimed at hospitals, radiology departments, and diagnostic imaging providers. The company does not publish standard subscription prices or a public self-service pricing plan for its medical imaging products. Prospective customers are directed to request a demonstration and contact the company for commercial information.

Pricing may depend on the selected applications, number of scanners, imaging facilities, deployment requirements, and commercial agreement. Institutions may also need to consider implementation, integration, technical support, and ongoing infrastructure costs when evaluating the total cost of adoption.

Review

Subtle Medical has a specialised product portfolio that addresses practical challenges in medical imaging rather than attempting to provide a general-purpose healthcare AI platform. Its focus on MRI, PET, and CT gives it a clear position in the radiology technology market. The availability of separate tools for image enhancement, synthetic contrast generation, and image alignment also gives healthcare organisations the opportunity to consider applications that match their particular imaging requirements.

One of its strongest advantages is its emphasis on working with existing scanner infrastructure. This can be valuable for hospitals that operate equipment from multiple manufacturers or need to improve capacity without immediately replacing expensive imaging systems. Its expansion into CT and PET also broadens the potential usefulness of the platform beyond its earlier MRI-focused offerings.

The company’s regulatory and quality-management work is another positive consideration for healthcare institutions evaluating medical device software. However, adoption still requires careful assessment of product-specific clearances, clinical evidence, compatibility, and local availability. Its enterprise-focused sales model and lack of published pricing may also make initial evaluation more difficult for smaller imaging providers.

Subtle Medical is best suited to established healthcare organisations that want specialised AI image enhancement and workflow improvements. It is not a consumer application, a general radiology information system, or a replacement for professional image interpretation. Its value depends on the clinical requirements of the facility, the performance of the selected application, and the suitability of its integration with existing imaging systems.