Search Results

3D Smile Predict aligners are indicated for the alignment of teeth in patients with permanent dentition (i.e. all second molars) during orthodontic treatment of malocclusion.

The 3D Smile Predict aligner system is a series of prescription-only clear plastic removable aligners intended to incrementally move a patient's teeth from an initial state to a different end state using a software-generated sequence of intermediate states. 3D Smile Predict Aligners sequentially reposition teeth by way of continuous gentle force.

A digital or traditional mold impression of the patient's teeth is provided by a dental health professional (e.g. orthodontist or dentist). From the digital data created of the patient's teeth, specialized orthodontic CAD/CAM software is used to develop treatment plans. Using the software, dental technicians design a series of intermediate models corresponding to each stage of treatment, gradually realigning the patient's teeth according to the dental health professional's prescription.

The prescribing doctor reviews and approves the model scheme and treatment plan before the molds/models are produced.

Once approved, the specialized orthodontic software is used to generate standard format 3D files which are used to physically produce each model/mold in the treatment plan for aligner fabrication. 3D Smile USA, Inc produces the aligner trays by thermoforming a plastic sheet over each model in the treatment plan. The aligner trays are sent to the dental health professional who then delivers them to the patient, confirming fit and design of the first stage of treatment. Additional trays are used sequentially by the patient to gradually move the teeth to the desired position. The dental health professional monitors treatment from the moment the first aligner is delivered to when the final aligner is finished and treatment complete. The aligners are held in place by pressure and can be removed by the patient at any time.

The provided text describes the 3D Smile Predict aligner system, which is a medical device. The document is primarily a 510(k) summary, aimed at demonstrating substantial equivalence to a predicate device. As such, it does not contain information about acceptance criteria or a study proving the device meets specific performance acceptance criteria for an AI/algorithm-driven component, as would be expected for a device with an AI/algorithm performing diagnostic or treatment planning functions independently.

The "3D Smile Predict" device is an orthodontic aligner system. While it uses "specialized orthodontic CAD/CAM software" to develop treatment plans and "interpret a series of images that represent intermediate teeth states," the crucial aspect is that "The prescribing doctor reviews and approves the model scheme and treatment plan before the molds/models are produced." This indicates that the software is a tool for the dental professional, and the final decision-making power rests with the human expert.

Therefore, the acceptance criteria and detailed study proving the device meets acceptance criteria regarding AI/algorithm performance are not present in this document because the AI/algorithm component is not making independent, unreviewed decisions. The document focuses on demonstrating substantial equivalence of the aligner system (the physical device and its manufacturing process) to a predicate device, not on validating the performance of an AI model that would otherwise require such rigorous testing.

Based on the provided text, here's what can be extracted (and what is explicitly not present):

1. A table of acceptance criteria and the reported device performance:
* Not provided in the context of AI/algorithm performance.
* The document mentions "All measurements were within the acceptance criteria" for a manufacturing validation demonstrating dimensional accuracy of 3D printed molds and aligner adaptation. However, no specific numerical acceptance criteria or reported values are presented in a table for this.

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective):
* Not applicable for AI/algorithm performance testing as described in a traditional AI regulatory submission.
* The document refers to a "manufacturing validation" but doesn't specify sample sizes for that process or for any data used by the software's internal algorithms beyond the general description of "a scan of a PVS impression, physical or a digital scan."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):
* Not applicable/not provided. The "ground truth" for the treatment plan is the "prescribing doctor's instructions" and their "review and approval" of the software-generated plan. This is a human-in-the-loop system, not an AI making independent diagnoses or treatment recommendations requiring external ground truth validation for its output.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
* Not applicable/not provided.

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, an MRMC study was not done or required. The software is described as a tool that generates a plan for review and approval by a dental professional, not as an AI providing assistance that demonstrably improves human reader performance that would necessitate such a study.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
* No, a standalone performance study was not done or required. The device clearly states that the "prescribing doctor reviews and approves" the treatment plan. Therefore, it is explicitly a human-in-the-loop system, not a standalone AI diagnostic or treatment planning tool.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
* The ground truth for the "treatment plan" is the prescribing doctor's clinical judgment and approval. The software generates "intermediate teeth states" based on a "final, provisional treated state" (which itself is likely defined by a dental professional), and the doctor "reviews these images and has the option to reject or request modifications... prior to approving it."

8. The sample size for the training set:
* Not provided. Given the description of the software as a CAD/CAM tool for generating treatment plans that are then approved by a doctor, it's not described as a deep learning or AI model trained on a large dataset in the way a diagnostic AI would be. If there is an underlying algorithm, its training set size is not disclosed.

9. How the ground truth for the training set was established:
* Not provided. Similar to point 8, the document does not describe the training methodology or ground truth establishment for any "training set."

In summary: The provided document is a 510(k) submission focused on demonstrating substantial equivalence for an orthodontic aligner system. It describes the software as a CAD/CAM tool facilitating treatment planning that requires definitive human review and approval. Consequently, the rigorous testing and reporting related to AI/algorithm performance (acceptance criteria tables, standalone performance, MRMC studies, ground truth establishment for AI training/testing) that would be expected for a device with an autonomous or high-risk AI component are not present because they are not applicable to the described functioning and regulatory pathway of this particular device.

Ask a specific question about this device