(30 days)
Nucleus.io is a software-based image management solution and PACS to be used by radiologists, other medical personnel and patients. Nucleus.io is comprised of software modules that provide image receipt, diagnostic viewing, storage, distribution, enhancement, sharing, manipulation, and networking of medical 2D/3D images at distributed locations.
Nucleus.io is a PACS which, when integrated with standard off-the-shelf hardware/software, acts similarly to its predicate device and other industry standard PACS systems. The Nucleus System has the following primary features and functions: Zero-footprint HTML5 medical image acceptance (upload) and transfer of medical images between facilities. Easy, real-time access to images for all participants in the healthcare process, including radiologists, imaging technicians, workflow coordinators, physicians, nurses, and other patient care facilitators. High-speed, zero-footprint diagnostic review of medical images using industry-standard tools for manipulation, annotation, measurement, and comparison. Simultaneous information review with multiple parties, including radiologists and coordinators. Sharing of medical images between healthcare providers, facilities, and patients. Encrypted transmission of medical images through secured wired and wireless networks. Encrypted storage of medical images. Organization and matching of multiple medical images based on patient name, medical record number, facility, etc. Quality assurance review of studies by coordinators and preparation of information for study interpretation by radiologists. Notification of care coordinators that studies need to be transferred to client systems. HIPAA-compliant data management and LDAP security integration with external systems Management of users, roles, permissions, and organization accounts Configuration of image acceptance (upload) and transfer settings. The Nucleus.io System consists of the following primary components: Nucleus.io Image Exchange (iX)™ - Exam acceptance (upload), transfer, sharing, and management. Nucleus.io Viewer (dX)™ - Zero-footprint streaming viewer suitable for diagnostic image review of all modalities. Nucleus.io Infinite Store (iS)™ (formerly Image Store) - Image storage and archiving. Nucleus.io PACS™ - Radiology workflow and worklist features. Nucleus PaaS - Vendor-neutral Class I platform that can be incorporated as a component of other devices.
The document describes the Nucleus.io, a Picture Archiving & Communications System (PACS), and its equivalency to a predicate device, Visage PACS/CS (K082269), particularly concerning the diagnosis of mammography.
Here's an analysis of the provided information regarding acceptance criteria and the study that proves the device meets them:
1. Table of acceptance criteria and the reported device performance:
The document does not provide a formal table of quantitative acceptance criteria and reported device performance metrics in the way one might expect for an AI/ML diagnostic algorithm (e.g., sensitivity, specificity, AUC). Instead, the acceptance criteria are implicitly defined by the goal of demonstrating substantial equivalence to the predicate device, particularly for mammography diagnosis. The reported performance is framed as the availability of features that are comparable to those in the predicate device.
Implicit Acceptance Criteria (based on comparison to predicate device):
- Availability of core PACS functionalities (image receipt, diagnostic viewing, storage, distribution, enhancement, sharing, manipulation, networking).
- Compatibility with various modalities (CT, MR, US, CR, DX, NM, PET, XA).
- Ability to handle uncompressed and non-lossy compressed images for primary image diagnosis in mammography.
- Support for the same mammography-specific viewing features as the predicate (Quadrant Navigation, Next/Previous Prior Exam, Change between 2D, Projection, and 3D (Tomo) Display).
- Meeting general safety and effectiveness requirements for a Class II medical device.
Reported Device Performance (as demonstrated by feature comparison):
| Feature/Function | Subject Device (Nucleus.io) | Comparable Device (Visage PACS/CS) | Assessment |
|---|---|---|---|
| Product Name | Nucleus.io | Visage PACS/CS | Equivalent in function, different name. |
| 510(k) Number | N/A - Proposed | K082269 | |
| Manufacturer | Nucleus Health, LLC | Visage Imaging, Inc. | |
| Web Site | www.nucleushealth.io | www.visageimaging.com | |
| Overview of devices | PACS/clinical viewing/image sharing/storage/long term archiving | PACS/clinical viewing/image sharing/storage/long term archiving | Equivalent. |
| User Types | Hospital Administrative Staff, Technicians, Radiologists, Patients | Hospital Administrative Staff, Technicians, Radiologists, Patients | Equivalent, with explicit mention of patients for non-diagnostic purposes in Nucleus.io. |
| Supports DICOM image transfer | Yes | Yes | Equivalent. |
| Software based | Yes (moderate concern) | Yes (moderate concern) | Equivalent risk classification by both. |
| LAN/WAN support | Yes | Yes | Equivalent. |
| HTML/Web based image transfer | Yes | Yes | Equivalent. |
| Client Hardware | Zero-footprint (browser) minimum | Thin Client minimum | Different technology (web-based vs. thin client), but serves equivalent purpose. |
| Image streaming | Yes | No | Different technology, claimed to be an improvement (faster, safer, efficient). |
| Secure image sharing to outside locations | Yes | Yes | Equivalent. |
| Dual monitor support | Yes | Yes | Equivalent. |
| Diagnostic/Clinical viewing | Yes | Yes | Equivalent. |
| Off the shelf hardware | Yes | Yes | Equivalent. |
| JPEG/industry standard lossy/lossless compression | Yes | Yes | Equivalent. |
| Zoom/panning features | Yes | Yes | Equivalent. |
| Image flip/rotate capability | Yes | Yes | Equivalent. |
| Text annotation | Yes | Yes | Equivalent. |
| Statistical reporting | Yes | Yes | Equivalent. |
| ROI | Yes | Yes | Equivalent. |
| Window level by region | Yes | Yes | Equivalent. |
| HIS/RIS connectivity | Yes | Yes | Equivalent. |
| Off-site viewing (including teleradiology) | Yes | Yes | Equivalent. |
| Fax support | Yes | Yes | Equivalent. |
| Scalable platform | Yes | Yes | Equivalent. |
| HIPAA compliant (encryption) | Yes | Yes | Equivalent. |
| HL-7 integration | Yes | Yes | Equivalent. |
| LDAP security integration | Yes | Yes | Equivalent. |
| Archiving to cloud based or network servers | Yes | Yes | Equivalent. |
| General admin features (assign, relate studies, etc.) | Yes | Yes | Equivalent. |
| Secure log in | Yes | Yes | Equivalent. |
| Support for Mammography | Yes | Yes | Key equivalence for this submission. |
| Quadrant Navigation (Mammography Specific) | Yes | Yes | Equivalent. |
| Next/Previous Prior Exam (Mammography Specific) | Yes | Yes | Equivalent. |
| Change between 2D, Projection, and 3D (Tomo) Display (Mammography Specific) | Yes | Yes | Equivalent. |
2. Sample size used for the test set and the data provenance:
The document does not describe a clinical performance study with a specific "test set" in the context of an AI/ML algorithm's diagnostic performance evaluation (e.g., a set of mammograms for classification). Instead, the "study" is a feature-by-feature comparison and technical validation against the predicate device.
- Sample size: Not applicable in the context of a dataset for AI evaluation. The "sample" here refers to the device itself and its implemented features.
- Data provenance: Not applicable. The document focuses on the features of the device rather than testing it against a clinical dataset.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
Not applicable. This device is a PACS/viewer system, not an AI diagnostic algorithm that requires expert-established ground truth for a clinical dataset. The "ground truth" for this submission is established by the functional specifications and features of the predicate device (Visage PACS/CS K082269) and the general requirements for PACS systems.
4. Adjudication method for the test set:
Not applicable. No clinical dataset requiring adjudication was used for this submission.
5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
No MRMC study was done or described. This is not an AI-assisted diagnostic device that augments human readers but rather a platform for viewing and managing medical images.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
Not applicable. Nucleus.io is a PACS/viewer, not a standalone diagnostic algorithm.
7. The type of ground truth used:
The "ground truth" for this 510(k) submission, particularly regarding the claim for mammography diagnosis, is the functional equivalency to a legally marketed predicate device. This is demonstrated by showing that Nucleus.io possesses the same relevant features and capabilities as Visage PACS/CS, especially those related to mammography viewing and processing, and that it adheres to industry standards (e.g., DICOM, handling uncompressed/lossless images for primary diagnosis). The risk management and verification/validation processes serve to confirm that the implemented features meet specifications and do not introduce new safety or effectiveness concerns.
8. The sample size for the training set:
Not applicable. This is not an AI/ML device that requires a training set.
9. How the ground truth for the training set was established:
Not applicable. This is not an AI/ML device that requires a training set.
In summary, this 510(k) submission for Nucleus.io is primarily based on demonstrating substantial equivalence to a predicate PACS system (Visage PACS/CS) by comparing functional features, particularly those supporting mammography diagnosis. It is not an submission for an AI/ML-driven diagnostic device, and therefore the traditional metrics and study designs (e.g., test sets, ground truth establishment, MRMC studies) typically associated with AI performance evaluation are not present in this documentation. The acceptance criteria are qualitative, focusing on feature parity and adherence to general medical device safety and performance standards.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).