Search Results

Regulation Number

Type

Panel

Ulthera System (UC-1 Control Unit PRIME)

Reference & Predicate Devices

K233996

Intended Use

The Ulthera System is intended to apply focused ultrasound energy to the body to achieve temporary changes in the physical appearance of the skin.

The Ulthera® System is indicated for use as a non-invasive dermatological aesthetic treatment to:

Lift the eyebrow
Lift lax submental (beneath the chin) and neck tissue, which can also affect the appearance of lax tissue in the submental and neck regions
Improve lines and wrinkles of the décolleté
Improve the appearance of skin laxity on the abdomen, anterior arms, and posterior arms

Ensure proper coupling of the transducer to the skin
Confirm appropriate depth of treatment such as to avoid bone

Device Description

AI/ML Overview

The provided document is a 510(k) summary for the Ulthera System, seeking clearance for expanded indications for aesthetic use on the abdomen, anterior arms, and posterior arms.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the information provided:

1. A table of acceptance criteria and the reported device performance

The document doesn't explicitly state quantitative acceptance criteria in a typical "numeric threshold" format. Instead, the acceptance is based on demonstrating "clinically significant skin improvement" and an acceptable safety profile, comparable to the predicate device. The performance is reported qualitatively based on the outcomes of the reviewed clinical studies.

Acceptance Criteria (Implicit)	Reported Device Performance (Summary from Document)
Safety	"From a safety perspective, there were no unexpected or severe adverse events (AEs), and all observed AEs resolved without long-term sequalae."
Effectiveness (Abdomen)	Three clinical studies (n=54 subjects) demonstrated "clinically significant skin improvement up to 180 days post-treatment via a variety of endpoints, including blinded clinician photographic assessment, physician and subject Global Aesthetic Improvement Scores (GAIS), patient satisfaction surveys, and other validated clinical endpoints."
Effectiveness (Arms)	Five clinical studies (n=113 subjects) demonstrated "clinically significant skin improvement up to 180 days post-treatment via a variety of endpoints, including blinded clinician photographic assessment, physician and subject Global Aesthetic Improvement Scores (GAIS), patient satisfaction surveys, and other validated clinical endpoints."
Functionality (Non-clinical)	"In all instances, the subject device functioned as intended." (Biocompatibility, Electrical Safety, Software V&V)
Substantial Equivalence	The device has the "same general intended use and principle of operation as the predicate device. The minor changes in device hardware and software do not raise different questions of safety or effectiveness and do not affect clinical functionality or performance specifications."

2. Sample sizes used for the test set and the data provenance

Test Set Sample Size: The document does not describe a traditional "test set" in the context of an AI/algorithm performance study. Instead, it references a literature search of existing clinical studies supporting the expanded indications.
- Abdomen: 54 subjects across 3 unique clinical studies.
- Arms (anterior and posterior): 113 subjects across 5 unique clinical studies.
- Total: 167 subjects across 7 unique clinical studies (some studies likely included both abdomen and arms, though this is not explicitly broken down for unique subjects per region).
Data Provenance: The document states "a comprehensive literature search was performed." This implies the data were
- Retrospective: The studies were already completed and published, not prospectively designed for this 510(k) submission.
- Country of Origin: Not specified in the provided text.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

The concept of "experts establishing ground truth" as typically seen in AI performance studies (e.g., radiologists annotating images) is not directly applicable here. The "ground truth" for the effectiveness of the Ulthera system is based on:

Blinded clinician photographic assessment: This indicates that medical professionals (clinicians) evaluated photographic evidence, likely pre- and post-treatment, without knowing which was which. The number and qualifications of these clinicians are not specified.
Physician Global Aesthetic Improvement Scores (GAIS): Physicians assessed improvements. The number and qualifications of these physicians are not specified.
Subject Global Aesthetic Improvement Scores (GAIS): Patients themselves provided self-assessments.

4. Adjudication method for the test set

The document does not detail an adjudication method for the clinical outcomes. While "blinded clinician photographic assessment" and "physician GAIS" suggest clinical evaluation, there's no mention of how discrepancies between multiple assessors were resolved (e.g., 2+1, 3+1). If multiple clinicians assessed the same case, their consensus or a defined adjudication process is not described.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done

No, an MRMC comparative effectiveness study was not conducted. This submission is for a device that delivers energy for aesthetic treatment, not an AI diagnostic algorithm that assists human readers in reading cases. The clinical data reported focuses on the device's direct effect on patients, not on how an AI component improves human interpretation.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done

The Ulthera System is a therapeutic device with an imaging component. It is not an AI algorithm, and therefore, a standalone performance study in the context of an AI system would not apply. The device's imaging function allows visualization for proper coupling and depth confirmation, but it's not a diagnostic AI system with an "algorithm only" performance metric. The "software verification and validation testing" mentioned is for the device's operational software, not an AI diagnostic algorithm.

7. The type of ground truth used

The ground truth for effectiveness was primarily clinical and subjective-to-observer:

Blinded clinician photographic assessment: Implies visual expert assessment.
Physician Global Aesthetic Improvement Scores (GAIS): Expert (physician) subjective assessment.
Subject Global Aesthetic Improvement Scores (GAIS): Patient subjective assessment.
Patient satisfaction surveys: Patient subjective assessment.
Other validated clinical endpoints: While vaguely described, this suggests other established clinical metrics were used.

No mention of pathology (biopsy) or long-term outcomes data beyond 180 days is provided as primary ground truth.

8. The sample size for the training set

The concept of a "training set" is not applicable here as the submission is for a medical device (focused ultrasound system) with new anatomical indications, not a machine learning model developed with a training set. The existing clinical studies were used as evidence to support the expanded indications, analogous to a validation set rather than a training set for an AI/ML model.

9. How the ground truth for the training set was established

As there is no "training set" for an AI/ML model in this submission, this question is not applicable.

Ask a Question

Ask a specific question about this device

K Number

K233996

Device Name

Manufacturer

Ulthera, Inc.

Date Cleared

2024-02-22

(66 days)

Product Code

OHV

Regulation Number

Type

Special

Panel

Reference & Predicate Devices

K180623

Intended Use

The Ulthera System is intended to apply focused ultrasound energy to the body to achieve temporary changes in the physical appearance of the skin.

The Ulthera System is indicated for use as a non-invasive dermatological aesthetic treatment to:

· Lift the eyebrow

• Lift lax submental (beneath the chin) and neck tissue. which can also affect the appearance of lax tissue in the submental and neck regions

· Improve lines and wrinkles of the décolleté

· Ensure proper coupling of the transducer to the skin

· Confirm appropriate depth of treatment such as to avoid bone

Device Description

AI/ML Overview

The provided text describes a 510(k) premarket notification for the Ulthera System (UC-1 Control Unit PRIME). The submission aims to demonstrate substantial equivalence to a predicate device (Ulthera System, K180623), primarily focusing on modifications to the console.

Based on the provided text, the device is an aesthetic device (Focused Ultrasound Stimulator System for Aesthetic Use) and NOT a diagnostic device or an AI/ML-driven device that involves human readers or complex image interpretation.

Therefore, most of the requested information regarding "acceptance criteria and the study that proves the device meets the acceptance criteria" in the context of diagnostic AI/ML devices (e.g., sample size, expert ground truth, MRMC studies, standalone performance) is not applicable or not detailed in this submission.

The "study" described is a series of non-clinical performance tests validating the modified console's functionality and safety, rather than a clinical study evaluating diagnostic accuracy or reader improvement.

Here's an attempt to address the points based on the available information, noting where information is not present due to the nature of the device and submission:

Acceptance Criteria and Device Performance for Ulthera System (UC-1 Control Unit PRIME)

The 510(k) submission primarily focuses on demonstrating that modifications to the Ulthera System's console do not significantly affect its clinical functionality or performance compared to the predicate device. The acceptance criteria are essentially that the modified device functions "as intended" and remains "as safe and effective" as the predicate device.

1. Table of Acceptance Criteria and Reported Device Performance

Since this is not a diagnostic device with specific performance metrics like sensitivity/specificity, the "acceptance criteria" relate to engineering, software, and physical performance.

Acceptance Criteria Category	Reported Device Performance/Verification
Biocompatibility	User-contacting components verified in accordance with ISO 10993-1. (Performance: Verified)
Electrical Safety and EMC	Verified in accordance with IEC 60601-1, IEC 60601-1-2, IEC 60601-1-6, IEC 60601-2-37, and IEC 60601-2-62. (Performance: Compliant)
Software Verification & Validation	Conducted as recommended by FDA Guidance, "Content of Premarket Submissions Device Software Functions." A basic documentation level was used as a failure would not present a hazardous situation with a probable risk of death or serious injury. (Performance: Conducted, Passed; Functions as intended.)
Usability	Conducted in accordance with IEC 62366-1 and FDA Guidance, "Applying Human Factors and Usability Engineering to Medical Devices." Specific checks included console's portability, GUI visibility and layout, and display responsiveness. (Performance: Clinicians able to use device in representative environment; no new risks identified.)
Mechanical Testing	Display hinge of the console verified via mechanical cyclic testing. (Performance: Verified)
Packaging, Transit, & Environmental	Packaging and transit testing per ASTM D4169 and ASTM D4332. Environmental conditioning testing conducted to verify device functionality at defined operating and storage conditions. (Performance: Verified)
Ultrasound Output Testing	Acoustic reliability verified using simulated treatment over the device's service life. Power output and imaging were verified to be within specification. (Performance: Within specification, operates as intended)
Overall Comparison (Subject vs. Predicate)	The submission states: "These non-clinical performance tests of the Ulthera System demonstrated the modified console continues to operate as intended and that it is as safe and effective as its predicate for the proposed indications for use." (Performance: Demonstrates substantial equivalence, safety, and effectiveness compared to predicate)

2. Sample Size Used for the Test Set and Data Provenance

This is not a clinical study involving a "test set" of patient data in the typical sense for AI/ML. The "testing" involves engineering and software validation on the device itself. Therefore, concepts like "sample size" for patient data and "data provenance" (country, retrospective/prospective) are not applicable here.

3. Number of Experts Used to Establish Ground Truth and Qualifications

Not applicable. The ground truth for this device's performance is established by engineering specifications, regulatory standards compliance, and direct verification of the device's physical and software functions. There are no "experts" establishing ground truth in the context of interpreting medical images or clinical outcomes for a test set.

4. Adjudication Method for the Test Set

Not applicable, as there is no "test set" of cases requiring adjudication by multiple readers.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No. An MRMC study is relevant for diagnostic devices where human reader performance (with or without AI assistance) is being evaluated relative to a reference standard. This device is an aesthetic treatment system, and the submission is for console modifications, not a new diagnostic algorithm.

6. If Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

No. This is not an AI/ML diagnostic algorithm. The device has an "imaging mode" for visualization and user guidance (proper coupling, depth avoidance) but is fundamentally an energy delivery system for aesthetic treatment. Its performance is tied to its physical and software functions (power output, imaging capabilities, safety features), which were tested directly.

7. The Type of Ground Truth Used

The "ground truth" for the various tests were:

Engineering specifications and design requirements: For mechanical, electrical, and acoustic performance.
International standards (ISO, IEC, ASTM): For biocompatibility, electrical safety, EMC, usability, packaging, and transit.
FDA Guidance: For software verification and validation, and usability/human factors.
Predicate device's established safety and effectiveness: The ultimate "ground truth" for substantial equivalence is demonstrating that the modified device performs comparably to the legally marketed predicate.

8. The Sample Size for the Training Set

Not applicable. This device does not employ machine learning that would require a "training set" of data.

9. How the Ground Truth for the Training Set Was Established

Not applicable, as there is no training set for machine learning.

Ask a Question

Ask a specific question about this device

K Number

K192185

Device Name

The Cellfina System

Manufacturer

Ulthera, Inc

Date Cleared

2019-10-09

(58 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K161885

Intended Use

The Cellfina System is intended for long term improvement in the appearance of cellulite in the buttocks and thigh areas of adult females as supported by clinical data demonstrating no significant reduction in treatment benefits through five years of observation.

Device Description

The Cellfina System is intended to provide precise focal release of subcutaneous tissue for improvement in the appearance of cellulite. The system consists of a sterile, single-use, disposable kit (CK1) and an electromechanical, non-sterile, 50-use motor module (CM1).

AI/ML Overview

Here's a breakdown of the acceptance criteria and study details for The Cellfina System, based on the provided text:

Acceptance Criteria and Device Performance

1. A table of acceptance criteria and the reported device performance

Acceptance Criteria (Endpoint)	Reported Device Performance at 5 Years
Primary Endpoint: Achievement of ≥1 point average reduction in the 0-5 point Cellulite Severity Scale as determined by independent physician assessment of subject photographs taken before and (5 years) after treatment.	The mean improvement was 1.8 points (p60% of treated subjects as determined by independent physician assessment of subject photographs taken before and 5 years after treatment.
Primary Safety Result: 100% of subjects free from Serious Adverse Events (SAE) directly attributable to the Cellfina System or procedure at 5 years.	100% of subjects were free from Serious Adverse Events (SAE) directly attributable to the Cellfina System or procedure at 5 years. (Met)
Additional Measures: Mean overall rate of correct selection (likely referring to identification of improvement by independent assessors)100% of subjects had noticeable improvement by the GAIS (Global Aesthetic Improvement Scale)35% of subjects characterized as having "Much Improved" or "Very Much Improved" by GAIS78% of subjects either "Satisfied" or "Very Satisfied"No subjects reported pain in the treatment area using an 11-point numerical rating scale at the 5-year follow-up visit.	The mean overall rate of correct selection was 92.8% with a lower 95% confidence limit of 86.3%. (Achieved/Acceptable)100% of the subjects had noticeable improvement by the GAIS. (Achieved/Acceptable)35% were characterized as having Much Improved or Very Much Improved. (Achieved/Acceptable)78% of the subjects were either Satisfied or Very Satisfied. (Achieved/Acceptable)No subjects reported pain in the treatment area using an 11-point numerical rating scale at the 5-year follow-up visit. (Achieved/Acceptable)

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

Sample Size for Test Set:
- 55 subjects participated in the baseline assessment and treatment.
- Follow-up progressively decreased: 3 Day (55), 14 Day (54), 1 Month (54), 3 Month (55), 6 Month (52), 1 year (50), 2 year (52), 3 year (45), 5 year (37). The primary and secondary endpoints are reported at 5 years, so the effective sample size for the long-term evaluation is 37.
Data Provenance:
- Study Design: Prospective, multi-center, non-randomized open label, safety and effectiveness study.
- Country of Origin: Not explicitly stated in the provided text, but "multi-center" implies it was conducted at several clinical sites.

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)

The text refers to "independent physician assessment" for determining the Cellulite Severity Scale and classifying improvement. However, it does not specify the number of physicians or their specific qualifications (e.g., years of experience, specialty).

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set

The text states "independent physician assessment." It does not explicitly describe an adjudication method like 2+1 or 3+1 (where multiple readers might assess independently and then a tie-breaker or consensus is used). It implies individual independent assessments were used to determine the Cellulite Severity Scale and photographic improvement.

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, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study comparing human readers with AI assistance was not done. The Cellfina System is a physical medical device (powered surgical instrument) for treating cellulite, not an AI-powered diagnostic or assistive tool. The assessments described are of the device's clinical efficacy on patients, conducted by physicians.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

No, a standalone (algorithm only) performance study was not done. As mentioned, The Cellfina System is a physical medical device, not an algorithm or AI. The study evaluated the device's performance in patients.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

The ground truth for effectiveness was primarily established by independent physician assessment of subject photographs, utilizing a Cellulite Severity Scale and classifying improvement grades.
For safety, it was based on outcomes data regarding the absence of Serious Adverse Events (SAE) attributable to the device or procedure.
Additionally, patient-reported outcomes (GAIS, satisfaction, pain levels) were also collected as secondary measures.

8. The sample size for the training set

The provided document describes a clinical study to evaluate the performance of the Cellfina System. It does not mention a "training set" in the context of an algorithm or AI model development. Therefore, no information is available on a "training set sample size" as this device is not an AI/ML product.

9. How the ground truth for the training set was established

Since the device is not an AI/ML product, there is no "training set" or "ground truth for the training set" as would be applicable to such products. The clinical study described is for evaluating the device's clinical performance directly.

Ask a Question

Ask a specific question about this device

K Number

K180623

Device Name

Ulthera System

Manufacturer

Ulthera, Inc

Date Cleared

2018-05-04

(56 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K134032

Intended Use

The Ulthera® System is indicated for use as a non-invasive dermatological aesthetic treatment to:

· lift the eyebrow

• Ift lax submental (beneath the chin) and neck tissue; which can also affect the appearance of lax tissue in the submental and neck regions

· improve lines and wrinkles of the décolleté

The Ulthera® System in conjunction with the Ulthera® DeepSEE transducer allows for ultrasonic visualization of depths up to 8 mm below the surface of the skin. The inaging is to visualize the dermal and subdermal layers of tissue to:

· ensure proper coupling of the transducer to the skin

· confirm appropriate depth of treatment such as to avoid bone

Device Description

The Ulthera® System consists of the Ulthera® Control Unit (with system software), a handpiece with cable, and interchangeable transducers. The device produces controlled tissue coagulation below the skin surface (epidermis) within the first few millimeters of tissue (dermis) using highly focused, low-energy ultrasound deposition. The Ulthera System directs micro-focused acoustic waves to the treatment area at desired depths without affecting or requiring a secondary action to protect the skin surface. The operator may also use the device's supplemental imaging capability to visualize the treatment area and aid in assuring full/proper skin contact of the Ulthera® System transducer to the skin in the target area.

AI/ML Overview

Here's an analysis of the acceptance criteria and study information for the Ulthera System, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria	Reported Device Performance
Primary Endpoint for "lifting lax submental (beneath the chin) and neck tissue": Improvement in lift, though specific quantitative metric for "improvement" is not defined, only that it refers to a lift of ≥20 mm².	- Quantitative Assessment: 51 out of 70 patients (72.9%) had an improvement of ≥20 mm² in lift.
Qualitative Assessment from Masked Evaluators: Improvement identified qualitatively.	- Qualitative Assessment: Of the 51 patients with ≥20 mm² lift, 84.3% were identified as showing improvement by masked evaluators.
Adverse Events: Mild, short-lived, and resolved without incident. No serious adverse events (SAEs) or unanticipated adverse device effects (UADEs).	- Adverse Events: Adverse events were mild, short-lived in duration, and resolved without incident. No serious adverse events (SAEs) or unanticipated adverse device effects (UADEs) related to treatment.

2. Sample Size and Data Provenance

Test Set Sample Size: 70 patients.
Data Provenance: The document does not explicitly state the country of origin. It describes an "open-label clinical trial," which implies prospective data collection.

3. Number of Experts and Qualifications for Ground Truth

Number of Experts: The document mentions "masked evaluators" but does not specify the number of individuals.
Qualifications of Experts: Not provided. The term "masked evaluators" suggests they were independent of the treatment process.

4. Adjudication Method

The document mentions "masked evaluators" identifying improvement. It doesn't detail a specific adjudication method like 2+1 or 3+1 for resolving discrepancies, implying a single assessment per evaluator or a simple consensus if multiple evaluators were involved.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC study described for human readers with and without AI assistance. This study focuses on the device's direct clinical effect on patients. The device itself uses imaging as an aid for the operator, but the clinical trial evaluates the treatment outcome, not the efficacy of the AI component in guiding human readers.

6. Standalone (Algorithm Only) Performance Study

Not applicable/Not done. The Ulthera System is a therapeutic device that involves human operation. The performance assessed here is the clinical outcome of the treatment when performed by an operator, not a standalone algorithmic output. The "imaging" component is a visualization aid for the operator, not a diagnostic algorithm with standalone performance metrics.

7. Type of Ground Truth Used

Clinical Outcomes/Quantitative Measurement: The primary ground truth for the effectiveness of "lifting lax submental and neck tissue" was a quantitative assessment of lift (≥20 mm² improvement) and qualitative assessment by masked evaluators. Adverse events were also monitored as part of the safety profile.

8. Sample Size for the Training Set

Not applicable/Not provided. This document describes a clinical trial for an aesthetic treatment device. The "Ulthera System" itself is a device that generates focused ultrasound; it's not explicitly described as having an AI component that is "trained" in the typical machine learning sense to perform a task. If the "system software" (Version 1700) has an AI component that processes data and learns, the training set size is not disclosed in this document. The clinical study detailed is a validation study for the device's clinical efficacy for a new indication, not an AI model training or validation set.

9. How Ground Truth for Training Set Was Established

Not applicable/Not provided. As stated above, the document does not discuss the training of an AI model.

Ask a Question

Ask a specific question about this device

K Number

K161885

Device Name

The Cellfina System

Manufacturer

Date Cleared

2016-10-05

(86 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K153677

Intended Use

The Cellfina™ System is intended for long term improvement in the appearance of cellulite in the buttocks and thigh areas of adult females as supported by clinical data demonstrating no significant reduction in treatment benefits up to 3 years of observation.

Device Description

AI/ML Overview

Here's an analysis of the provided text to extract the acceptance criteria and study details for "The Cellfina System":

Acceptance Criteria and Reported Device Performance

Acceptance Criteria (Study Endpoint/Measure)	Reported Device Performance at 3 Years
Primary Endpoint: Achievement of ≥1 point average reduction in the 0-5 point Cellulite Severity Scale as determined by independent physician assessment of subject photographs.	The primary endpoint was met: The average improvement was **2.0 points (p60% of treated subjects as determined by independent physician assessment of subject photographs.
Primary Safety Endpoint: 100% of subjects free from Serious Adverse Events (SAE) directly attributable to the Cellfina System or procedure.	The safety endpoint was met: 100% of subjects were free from Serious Adverse Events (SAE) directly attributable to the Cellfina System or procedure.
Additional Secondary Measure (Rate of Correct Selection): (Implicit, likely referring to identification of affected areas/response to treatment)	The average rate of correct selection was 97% with a 95% confidence limit of 92.6%.
Additional Secondary Measure (Noticeable Improvement by GAIS):	100% of the subjects had noticeable improvement by the GAIS.
Additional Secondary Measure (Marked Improvement or Better by GAIS):	56% were characterized as having marked improvement or better.
Additional Secondary Measure (Subject Satisfaction): Percentage of subjects either satisfied or very satisfied.	93% of the subjects were either satisfied or very satisfied as evaluated by a 5-point Likert type scale.
Additional Secondary Measure (Pain Level): Percentage of subjects rating pain ≤3/10.	71% of subjects rated pain ≤ 3/10 at 3 days, and over 95% rated pain ≤ 3/10 thereafter. There were no subjects who reported pain at the 3-year follow-up.

Study Details

Sample size used for the test set and the data provenance:
- Sample Size: 55 subjects participated in the baseline assessment and treatment.
- Follow-up at 3 years: 45 subjects completed the 3-year follow-up.
- Data Provenance: The study was a "Prospective, multi-center, non-randomized open label, safety and effectiveness study." This indicates it was a prospective study, typically conducted in multiple clinical sites (country of origin is not explicitly stated, but often such trials for FDA approval are largely US-based unless otherwise specified).
- Imputation: For the 10 subjects lost to follow-up at 3 years, "last observation carried forward (LOCF)" was used for both primary and 1st secondary endpoints, and "multiple imputation" for the primary effectiveness endpoint.
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The primary and secondary endpoints were determined by "independent physician assessment of subject photographs." The exact number of independent physicians (experts) is not specified in the provided text. Their specific qualifications (e.g., "radiologist with 10 years of experience") are also not detailed, beyond being "independent physicians."
Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- The adjudication method is not explicitly described in the provided text. It only mentions "independent physician assessment." This could imply a single independent assessor or a consensus method, but the specific process (e.g., if multiple physicians' assessments were averaged or if a tie-breaking mechanism was used) is not detailed.
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 comparative effectiveness study was not done. This study evaluates a device (The Cellfina System) for treating cellulite, not an AI or imaging diagnostic tool. Therefore, the concept of "human readers improve with AI vs without AI assistance" is not applicable to this submission.
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- No, this device is a powered surgical instrument, not an algorithm. The performance detailed here is for the physical device and procedure, with human physicians performing the assessment.
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The ground truth for effectiveness was established primarily through expert assessment/consensus based on photographs ("independent physician assessment of subject photographs").
- Safety was based on outcomes data (absence of Serious Adverse Events reported).
- Subject satisfaction and pain levels were based on patient-reported outcomes.
The sample size for the training set:
- Not applicable. This study is for a medical device (a powered surgical instrument), not a machine learning model. There is no concept of a "training set" in this context. The clinical study described served as the validation and effectiveness study for the device itself.
How the ground truth for the training set was established:
- Not applicable, as there is no training set for a medical device of this nature.

Ask a Question

Ask a specific question about this device

K Number

K153677

Device Name

The Cellfina System

Manufacturer

Date Cleared

2016-02-25

(66 days)

Product Code

Regulation Number

Type

Special

Panel

Reference & Predicate Devices

K150505

Intended Use

Device Description

AI/ML Overview

The provided text is related to a 510(k) summary for the Cellfina System and discusses an extension of shelf life and labeling changes. It repeatedly refers to the predicate device (K150505) for clinical trial data and other non-clinical tests. Therefore, the details regarding acceptance criteria and the study proving the device meets those criteria are primarily found by referencing the K150505 submission, which is not fully included in this document.

However, based on the information provided in the current document (K153677) and the references within it, here's what can be extracted and inferred:

1. A table of acceptance criteria and the reported device performance

The provided document (K153677) is a Special 510(k) for a shelf-life extension and labeling changes. It states that the clinical effectiveness and safety data from the predicate device (K150505) are applicable. Thus, the acceptance criteria for the effectiveness of the device are derived from the clinical data demonstrating "long term improvement in the appearance of cellulite... up to 2 years of observation" as stated in the Indications for Use.

For the specific changes submitted in K153677 (shelf-life extension), the acceptance criteria are related to the stability and performance of the device after aging. The document states:

Acceptance Criteria (for K153677, focusing on shelf-life)	Reported Device Performance (from K153677)
CK1 disposable kit shelf life stability	"ship testing, simulated use, structural integrity, verification of physical characteristics and primary functionality on 12 month aged product demonstrates that the CK1 disposable kit is stable out to 12 months of shelf life."
Compliance with electrical safety standards	"The subject device... was tested to IEC 60601-1: 2005 + CORR. 1 (2006) + CORR. 2 (2007) and was found to be in compliance with the requirements of this standard."

Note on Clinical Performance from Predicate Device (K150505) as inferred from Indications for Use:

Acceptance Criteria (for device effectiveness as per K150505)	Reported Device Performance (Inferred from Indications for Use)
Long term improvement in the appearance of cellulite	"long term improvement in the appearance of cellulite in the buttocks and thigh areas of adult females as supported by clinical data demonstrating no significant reduction in treatment benefits up to 2 years of observation."

2. Sample size used for the test set and the data provenance

The current document (K153677) does not provide this information directly as it refers to the predicate device (K150505) for "in-vivo evaluation (clinical trial data)". For the shelf-life extension, the "test set" for performance evaluation consists of "12 month aged product" for the CK1 disposable kit. The sample size for this testing is not specified. The data provenance for the clinical trial would be available in the K150505 submission.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

This information is not provided in the K153677 document, as it defers to the predicate device (K150505) for clinical data. The clinical trial data for the original device approval would have involved assessments by qualified medical professionals (e.g., dermatologists, plastic surgeons) to determine the improvement in cellulite appearance.

4. Adjudication method for the test set

This information is not explicitly provided in the K153677 document. The adjudication method for assessing clinical outcomes (e.g., improvement in cellulite appearance) would have been specified in the clinical trial protocol referenced in the K150505 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

This is not applicable. The Cellfina System is a powered surgical instrument for treating cellulite, not an AI-assisted diagnostic or imaging device used by "readers." Therefore, a MRMC study involving "human readers" and "AI assistance" would not be relevant.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

This is not applicable. The Cellfina System is a physical device used by a physician (human-in-the-loop) to perform a surgical procedure. It is not an algorithm performing a task without human intervention.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

For the clinical effectiveness of the device (as inferred from K150505 via the Indications for Use), the ground truth for "improvement in the appearance of cellulite" would likely be based on:

Expert assessment: Clinical evaluators (e.g., physicians) making subjective and/or objective assessments of cellulite severity before and after treatment.
Patient outcomes data: Patient satisfaction and self-assessment of their cellulite appearance.
Possibly photographic evidence: Standardized photographs evaluated by blinded experts.

For the shelf-life extension (K153677), the ground truth is based on engineering and performance testing to ensure the device components meet specifications after aging.

8. The sample size for the training set

This is not applicable in the context of an algorithm training set, as the Cellfina System is a medical device for direct treatment, not an AI/ML model being trained. If "training set" refers to the subjects in the original clinical study (K150505), that information is not provided here.

9. How the ground truth for the training set was established

Not applicable for the same reasons as point 8.

Ask a Question

Ask a specific question about this device

K Number

K150505

Device Name

Cellfina System

Manufacturer

Date Cleared

2015-07-29

(154 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K134010

Intended Use

Device Description

The Cellfina System is intended to provide precise focal release of subcutaneous tissue for improvement in the appearance of cellulite. The release of subcutaneous tissue for improvement in the appearance of cellulite is a minimally invasive surgical technique by physicians using of manual surgical instruments and accessories. The Cellfina device consists of a powered cutting blade and a means for instrument guidance to control the depth, size and shape of the tissue release. The system consists of a sterile single use disposable kit and a non-sterile reusable motor module.

AI/ML Overview

The provided text describes specific acceptance criteria and performance data for the Cellfina™ System, a device intended for the improvement of the appearance of cellulite.

Here's a breakdown of the requested information:

1. A table of acceptance criteria and the reported device performance

Acceptance Criteria (Primary Endpoints)	Reported Device Performance (2-year follow-up)
Achievement of ≥1 point average reduction in the 0-5 point Cellulite Severity Scale as determined by independent physician assessment of subject photographs taken before and 2 years after treatment.	The average improvement was 2.0 points (p60% of treated subjects as determined by independent physician assessment of subject photographs taken before and 2 years after treatment.
Freedom from Serious Adverse Events (SAE) directly attributable to the Cellfina™ System or procedure.	100% of subjects were free from Serious Adverse Events (SAE) directly attributable to the Cellfina™ System or procedure. (Met)

Additional Secondary Measures Achieved:

Correct Selection by Independent Physician Assessment: Average rate of correct selection of blinded subject photographs (before and two years after treatment) was 98.1% (95% confidence limit of 94.5%).
Noticeable Improvement (GAIS) and Marked Improvement: 100% of subjects had noticeable improvement by the Global Aesthetic Improvement Scale (GAIS). 52% were characterized as having marked improvement or better by independent physician assessment.
Subject Satisfaction: 96% of subjects were either satisfied or very satisfied at the 2-year follow-up.
Subject Reported Pain: Average pain on a 0-10 numerical rating scale was 4.5 for anesthesia delivery and 3.7 for tissue release. At follow-ups, 71% of subjects rated pain ≤ 3 at 3 days, and over 95% rated pain ≤ 3 thereafter. No subjects reported pain at the 2-year follow-up.

2. Sample size used for the test set and the data provenance

Sample Size for Test Set: The document indicates that "All subjects served as their own control" and mentions endpoints being achieved for "all subjects." However, the exact number of subjects in the pivotal clinical study (test set) is not explicitly stated in the provided text.
Data Provenance: The study was a "prospective, multi-center, non-randomized open label, safety and effectiveness study." The location (country of origin) of the study centers is not specified, but it was conducted under IDE G120116.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Number of Experts: An "independent and blinded reviewer of the photographs" was used to verify effectiveness, and "independent physician assessment" determined the reduction in cellulite severity and grade improvement. The exact number of independent physicians/reviewers is not specified.
Qualifications of Experts: The text refers to "independent physician assessment." No further details on their specific qualifications (e.g., specialty, years of experience) are provided.

4. Adjudication method for the test set

The document mentions an "independent and blinded reviewer" for photographs to verify effectiveness. For the primary and powered secondary endpoints, "independent physician assessment" was used. The text does not explicitly detail an adjudication method (e.g., 2+1, 3+1) if there were multiple independent assessments or if there were disagreements. It implies a single "independent physician assessment" for each metric.

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, a multi-reader multi-case (MRMC) comparative effectiveness study focusing on AI assistance for human readers was not mentioned. The study evaluated the effectiveness of the Cellfina System itself, not an AI assistant.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

No, a standalone (algorithm only) performance study was not mentioned. The Cellfina System is a physical device that is used by physicians, and the study evaluated its direct clinical performance on patients.

7. The type of ground truth used

The ground truth was established through expert consensus/assessment of clinical data. This included:
- Independent physician assessment of subject photographs (before and 2 years after treatment) for cellulite severity and grade improvement.
- Global Aesthetic Improvement Scale (GAIS) based on independent physician assessment.
- Subject-reported satisfaction and pain levels.
- Adverse event reporting.

8. The sample size for the training set

The text does not mention a separate "training set" in the context of device development or any machine learning/AI models. The clinical study described is a pivotal trial evaluating the device's safety and efficacy.

9. How the ground truth for the training set was established

As no "training set" (in the context of AI/machine learning) is described, this question is not applicable based on the provided text. The clinical study described served as the basis for demonstrating the device's effectiveness.

Ask a Question

Ask a specific question about this device

K Number

K134032

Device Name

ULTHERA SYSTEM

Manufacturer

Date Cleared

2014-06-20

(171 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K132028

Intended Use

The Ulthera® System is indicated for use as a non-invasive dermatological aesthetic treatment to:

lift the eyebrow
lift lax submental (beneath the chin) and neck tissue
improve lines and wrinkles of the décolleté

The Ulthera System in conjunction with the Ulthera DeepSEE transducer allows for ultrasonic visualization of depths up to 8 mm below the surface of the skin. The indicated use of the imaging is to visualize the dermal layers of tissue to:

ensure proper coupling of the transducer to the skin
confirm appropriate depth of treatment such as to avoid bone

Device Description

The Ulthera® System consists of the following components:
• Ulthera® Control Unit
• Handpiece
• Transducers

AI/ML Overview

The Ulthera® System, which uses focused ultrasound for aesthetic treatment, was evaluated for an expanded indication to improve lines and wrinkles of the décolleté.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and the Reported Device Performance:

Acceptance Criteria	Reported Device Performance
Primary Endpoint: Improvement in lines and wrinkles of the décolleté as evaluated by a masked assessment of pre and post-treatment photographs. (Note: No pre-specified success criteria were established for this primary endpoint at the beginning of the trial due to issues with the initial Fabi-Bolton Scale validation).	36 out of 54 evaluable subjects (~67%) showed improvement by masked assessment of pre- and post-treatment photographs at 180 days post-treatment.
Safety: No serious adverse events (SAEs) or unanticipated adverse device effects (UADEs) related to the treatment.	No SAEs or UADEs related to treatment were reported. All adverse events were mild, with the exception of two moderate events (one of which was not device-related), and all resolved.
Treatment Efficacy at Specific Energy Levels: Efficacy achieved at Level 4 for 7-3.0 and 4-4.5 transducers, and Level 3 for 10-1.5 transducer.	Treatment efficacy was achieved at the pre-set energy levels: Level 4 for the 7-3.0 and 4-4.5 transducers, and Level 3 for the 10-1.5 transducer.

2. Sample Size Used for the Test Set and Data Provenance:

Sample Size for Test Set:
- Initial Enrollment: Up to 130 female subjects.
- Subjects Completing Protocol: 108 subjects.
- Evaluated Photo Sets (Test Set): 54 subjects. This subset was chosen as their day 180 photos were identified as "most consistent in photo quality" after excluding 54 photos with inconsistencies.
Data Provenance: The study was a prospective safety and efficacy study. The document does not specify the country of origin of the data, but the submission is to the US FDA, implying it was conducted to meet US regulatory requirements.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications:

The document mentions "masked assessment of pre and post treatment photographs" but does not specify the number or qualifications of the experts who performed this masked assessment. It also states that the initial Fabi-Bolton Scale, a published validated scale, could not be successfully validated during the clinical trial due to low kappa scores for both intra-rater and inter-rater reproducibility, leading to the change in primary endpoint to a post-hoc masked assessment.

4. Adjudication Method for the Test Set:

The document mentions "masked assessment," which implies multiple raters, but it does not explicitly state the adjudication method used to arrive at the final assessment (e.g., 2+1, 3+1, majority vote, etc.). The mention of inter-rater reproducibility issues with the Fabi-Bolton scale suggests that inter-rater agreement was a consideration, but the method for resolving discrepancies in the masked assessment is not detailed.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:

No, an MRMC comparative effectiveness study comparing human readers with and without AI assistance was not conducted or reported. This study evaluates the device's direct clinical effect on patients, not the performance of an AI algorithm assisting human readers.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done:

Not applicable. The Ulthera® System is a medical device for treatment, not an AI algorithm for diagnostic interpretation in a standalone context. The "performance data" refers to the clinical outcomes of patients treated with the device.

7. The Type of Ground Truth Used:

The ground truth for the expanded indication of improving lines and wrinkles on the décolleté was established through:

Expert Consensus (implied): "Masked assessment of pre and post treatment photographs." While not explicitly stated as "consensus," this type of assessment by presumably trained evaluators serves as the "ground truth" for visual improvement.
Clinician Global Aesthetic Improvement Scores (CGAIS): An unmasked assessment performed by clinicians.
Patient Satisfaction Questionnaires: Patient-reported improvement.

The primary endpoint shifted to the masked assessment due to issues with the initial validated scale.

8. The Sample Size for the Training Set:

This information is not provided in the document. The study described is a clinical trial to evaluate the device's effectiveness, not an AI model's development. Therefore, there isn't a "training set" in the context of machine learning. The "training" for this device would refer to the development and engineering process, which is not detailed in terms of data sets.

9. How the Ground Truth for the Training Set Was Established:

Not applicable, as there is no mention of a "training set" for an AI algorithm in this context. The study focuses on evaluating the clinical performance of the Ulthera® System treatment itself.

Ask a Question

Ask a specific question about this device

K Number

K132028

Device Name

ULTHERA SYSTEM

Manufacturer

Date Cleared

2013-12-11

(163 days)

Product Code

Regulation Number

Type

Panel