(105 days)
Ricoh 3D Anatomic Models are intended as physical replicas of patient anatomy to be used for diagnostic purposes in the fields of craniomaxillofacial, orthopedic, cardiovascular, neurological, gastrointestinal, genitourinary, and breast applications. The Anatomic Models are based on DICOM imaging information from a medical scanner and output files from FDA cleared software intended for the creation and output of digital files suitable for the fabrication of physical replicas. The models should be used in conjunction with other diagnostic tools and expert clinical judgement.
The subject device, each "Ricoh 3D Anatomic Model," is a patient-specific physical replica of an anatomic structure or site, produced via additive manufacturing from a user generated 3D print file. The input 3D print file is created from medical images in DICOM format that have been seqmented to a specific reqion of interest within an FDA cleared application, IBM iConnect Access (K203104). The input 3D print file is then transferred to Ricoh for production and delivery of the physical replica.
The provided text
describes the acceptance criteria and the study that proves the device meets the
acceptance criteria in the context of a 510(k) premarket notification for a 3D Anatomic
Model.
Here's the breakdown of the information requested based on the
provided text:
1. Table of Acceptance Criteria and Reported Device
Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Geometric accuracy of the physical replicas ("can be printed accurately at less than 1mm mean deviation when compared against the input digital 3D file") | Testing showed that the physical models can be printed accurately at less than 1mm mean deviation when compared against the input digital 3D file. |
| All clinically relevant acceptance criteria were met. | All clinically relevant acceptance criteria were met. (Specific criteria are not enumerated, but the broad statement |
| of meeting them is present.) | |
| Packaging adequately protects the product from damage throughout the distribution process. | Testing showed that the packaging adequately protects the product from damage throughout the distribution process. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Test Set: The document does not explicitly state the specific
sample size used for the geometric accuracy testing (e.g., number of models
tested). It refers to "testing" and "assessment" but not quantitative sample
sizes. - Data Provenance: The document does not specify the country of origin of the data
or whether the study was retrospective or prospective. It describes "bench testing"
and "simulated distribution and handling testing," which are laboratory/controlled
tests rather than patient-data-driven clinical studies in the typical sense for an
AI/imaging device.
3. Number of Experts Used to Establish Ground Truth and
Qualifications
Not applicable. For this device (3D Anatomic Model), the "ground
truth" for geometric accuracy is the input digital 3D file (the design spec).
The product is a physical replica, and its accuracy is measured against the
digital file itself, not against a human expert's interpretation of a medical
image or a medical diagnosis. The document mentions "clinical user" approval of
the model, but this is part of the workflow and quality process, not the
ground truth establishment for performance testing.
4. Adjudication Method for the Test Set
Not applicable. As noted above, the primary performance test
(geometric accuracy) compares the physical model directly to the digital input
file. There is no human interpretation or subjective assessment that would
require an adjudication method.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness
Study was Done
No. The document makes no mention of an MRMC study. This type of
study is typically performed for AI-powered diagnostic software to assess its
impact on human reader performance. The device here is a physical replica
generated from an existing cleared software's output, not a diagnostic AI
algorithm.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop
Performance) Study was Done
Yes, in a way. The "Validation Testing" described, particularly the
geometric accuracy assessment, represents a standalone performance test of the
device's manufacturing capability to produce accurate physical models based on
digital input. It assesses the output of the manufacturing process (the
physical model) against its input (the digital file) without human
intervention being a variable in the measurement of geometric accuracy itself.
7. The Type of Ground Truth Used
The ground truth for the device's performance (geometric accuracy)
is the input digital 3D file from which the physical replica is printed.
The physical model is compared directly to this digital file using measurement
techniques to determine deviation.
For the purpose of the device's function (being a physical replica
for diagnostic purposes), the underlying "ground truth" of the patient's
anatomy would stem from the original DICOM imaging information, which is then
segmented by FDA cleared software to create the 3D print file. However, the
performance assessment of this specific device (the physical model itself)
uses the digital file as its reference.
8. The Sample Size for the Training Set
Not applicable. This device is a manufactured physical product (a 3D
print) based on a digital file, not an AI/machine learning algorithm that
requires a training set. The performance data relates to the manufacturing
accuracy and physical integrity of the printed models, not to an algorithm's
ability to learn from data.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as there is no training set for this device.
§ 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).