(107 days)
The UltraShape System delivers focused ultrasound energy that can disrupt subcutaneous adipose tissue (SAT) for lipolysis (breakdown of fat) to provide a non-invasive a desired aesthetic effect. It is intended for non-invasive reduction in abdominal circumference and fat reduction in the flanks and thighs.
The UltraShape System is comprised of multiple components, including the control unit and two ultrasonic transducers. The UltraShape System selectively targets subcutaneous adipose tissue via focused ultrasound for the purpose of non-invasive aesthetic body contouring. The transducers are electro-mechanical devices that convert an electrical signal into mechanical (acoustical) energy. The operating parameters of the UltraShape System achieve selective disruption of adipose tissue without damaging neighboring tissues such as blood vessels, nerves, or muscle. The UltraShape System has two treatment modes available, the Single Focus Deep Mode and the Super Mode.
- Table of Acceptance Criteria and Reported Device Performance:
| Acceptance Criteria (Primary Effectiveness Endpoint) | Reported Device Performance |
|---|---|
| Blinding reviewers, when shown randomized pairs of thigh photographs, correctly identify both pre- versus post-treatment and treated versus untreated thighs in at least 80% of cases. (Based on visual assessment of randomized photographs of before (baseline) versus after (16 weeks following last treatment), as well as treated versus untreated thighs.) | Met endpoint: Blinded reviewers correctly identified (by agreement between the blinded reviewers) both the pre-/post-treatment photographs as well as the treated/un-treated thighs in 81% (30 of 37) subjects. Furthermore, even in the Intent-to-Treat population, with worst case imputation where all missing data were treated as failures, 64% of the subjects were successes. |
-
Sample Size Used for the Test Set and Data Provenance:
- Sample Size: 47 subjects (all female).
- Data Provenance: Prospective, single-arm, self-controlled, multicenter clinical study conducted at 3 U.S. sites.
-
Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications:
- Number of Experts: 2 blinded reviewers.
- Qualifications: The document does not specify the qualifications of the blinded reviewers (e.g., radiologist with X years of experience). It only states they were "blinded reviewers."
-
Adjudication Method for the Test Set:
- The device performance indicates that the primary endpoint was met by "agreement between the blinded reviewers." This implies a consensus-based adjudication method, likely 2 out of 2 agreement, for the visual assessment of photographs.
-
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. The study design was a prospective, single-arm, self-controlled study rather than a comparative effectiveness study comparing AI with and without human assistance. The primary endpoint involved human reviewers assessing photographs, but it was for standalone device effectiveness, not to measure human improvement with AI assistance.
-
Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study:
- No, this study was not a standalone (algorithm only) performance study. The primary effectiveness endpoint relied on assessment by "blinded reviewers" of photographs, which inherently involves human-in-the-loop performance for evaluation.
-
Type of Ground Truth Used:
- The primary ground truth for the effectiveness endpoint was expert consensus (agreement between the two blinded reviewers) based on visual assessment of photographs of pre- and post-treatment thighs, as well as treated vs. untreated thighs.
- Secondary endpoints used objective measurements like ultrasound and caliper measurements for fat thickness reduction, and investigator and subject satisfaction assessments.
-
Sample Size for the Training Set:
- The document describes a clinical study to evaluate the device, not an AI or algorithm, so there is no mention of a "training set" in the context of machine learning. The clinical study itself served to demonstrate the device's performance in humans.
-
How the Ground Truth for the Training Set Was Established:
- Not applicable, as this is not an AI/machine learning device that would require a separate training set with established ground truth. The study evaluated the direct effects of the UltraShape System on human subjects.
§ 878.4590 Focused ultrasound stimulator system for aesthetic use.
(a)
Identification. A Focused Ultrasound Stimulator System for Aesthetic Use is a device using focused ultrasound to produce localized, mechanical motion within tissues and cells for the purpose of producing either localized heating for tissue coagulation or for mechanical cellular membrane disruption intended for noninvasive aesthetic use.(b)
Classification. Class II (special controls). The special control for this device is FDA's “Class II Special Controls Guidance Document: Focused Ultrasound Stimulator System for Aesthetic Use.” See § 878.1(e) for the availability of this guidance document.