(103 days)
ACUSON Sequoia and ACUSON Sequoia Select
The ACUSON Sequoia and Sequoia Select ultrasound imaging systems are intended to provide images of, or signals from, inside the body by an appropriately trained healthcare professional in a clinical setting for the following applications: Fetal, Abdominal, Pediatric, Neonatal Cephalic, Small Parts, OB/GYN (useful for visualization of the ovaries, follicles, uterus and other pelvic, structures), Cardiac, Adult Cephalic, Musculoskeletal and Peripheral Vascular applications.
The system supports the Ultrasonically-Derived Fat Fraction (UDFF) measurement tool to report an index that can be useful as an aid to a physician managing adult patients with hepatic steatosis.
The system also provides the ability to measure anatomical structures for fetal, abdominal, pediatric, small organ, cardiac, transrectal, transvaginal, peripheral vessel, musculoskeletal and calculation packages that provide information to the clinician that may be used adjunctively with other medical data obtained by a physician for clinical diagnosis purposes.
ACUSON Origin
The ACUSON Origin ultrasound imaging system is intended to provide images of, or signals from, inside the body by an appropriately trained healthcare professional in a clinical setting for the following applications: Abdominal, Pediatric, OB/ GYN (useful for visualization of the ovaries, follicles, uterus and other pelvic structures), Cardiac, Transesophageal, Intracardiac, Vascular, Adult Cephalic, and Peripheral Vascular applications.
The catheter is intended for intracardiac and intra-luminal visualization of cardiac and great vessel anatomy and physiology as well as visualization of other devices in the heart of adult and pediatric patients. The catheter is intended for imaging guidance only, not treatment delivery, during cardiac interventional percutaneous procedures.
The ACUSON Sequoia, Sequoia Select, and Origin Diagnostic Ultrasound Systems are multi-purpose, mobile, software-controlled, diagnostic ultrasound systems with an onscreen display of thermal and mechanical indices related to potential bio-effect mechanisms. The ultrasound system function is to transmit and receive ultrasound echo data and display it in B-Mode, M-Mode, Pulsed (PW) Doppler Mode, Continuous (CW) Doppler Mode, Color Doppler Mode, Color M Mode, Doppler Tissue Mode, Amplitude Doppler Mode, a combination of modes. Panoramic Imaging. Contrast agent Imaging. Virtual Touch Strain Imaging (except Origin), Virtual Touch - pSWE Imaging, Virtual Touch - SWE Imaging, Custom Tissue Imaging, 3D/4D Volume Imaging or Harmonic Imaging on a Display and provide cardiac anatomical and quantitative function software applications.
The provided FDA 510(k) summary describes the Siemens ACUSON Sequoia, ACUSON Sequoia Select, and ACUSON Origin Diagnostic Ultrasound Systems, focusing on new software applications (AI Measure, AI Assist, 2D HeartAI, 4D HeartAI) that incorporate Machine Learning algorithms for cardiac imaging.
The document includes summaries of validation testing for each of these AI features.
1. Table of Acceptance Criteria and Reported Device Performance
| AI Feature | Acceptance Criteria / Performance Metric | Reported Device Performance |
|---|---|---|
| AI Assist | Overall Success Rate: Percentage of placements requiring no adjustment or only minor adjustment by the user. | 99.3% overall successful placement across all cardiac views and placement locations. For any single placement location, success is 92% or higher. |
| 2D HeartAI | Pearson Correlation Coefficient: With user edits compared to reference standard ground truth. | 0.81 or higher |
| Bland-Altman Bias: With user edits compared to reference standard ground truth. | less than 5.2 (minimal bias) | |
| Performance for Normal BMI subjects ( 25 kg/m2): Pearson Correlation Coefficient. | greater than 0.82 | |
| 4D HeartAI | Pearson Correlation Coefficient: With user edits compared to reference standard ground truth. | 0.87 or higher |
| Bland-Altman Bias: With user edits compared to reference standard ground truth. | less than 13.3 (minimal bias) | |
| Performance for Normal BMI subjects ( 25 kg/m2): Pearson Correlation Coefficient. | greater than 0.81 | |
| AI Measure | Overall Success Rate: Percentage of measurements deemed clinically acceptable (Pass or Pass with/Edit). | 89.6% overall successful performance. For any single acquisition mode, success is 88.1% or higher. |
2. Sample Sizes Used for the Test Set and Data Provenance
| AI Feature | Test Set Sample Size (Patients/Exams/Samples) | Data Provenance |
|---|---|---|
| AI Assist | 12 patients; 14 exams (12 patients, with 2 having additional data); 168 samples (16 frames x 12 cardiac views x 14 exams). | US; Retrospective (collected by a cardiac sonographer who scanned 12 cardiac views included in an Adult Echocardiogram examination). |
| 2D HeartAI | 45 exams; 90 images extracted. | US, Mexico; From 5 different institutions; Retrospective. |
| 4D HeartAI | 32 patients; 64 volumes extracted. | US, Mexico; From 5 different institutions; Retrospective (for 5Z1 (17 TTE) and 6ZT (15 TEE) transducers, with Volume rates > 13.4vps). |
| AI Measure | 32 individual patients; 1354 samples (from 392 images). | US, Mexico; From 5 institutions; Retrospective (the dataset consists of exams from across five institutions, five probes (5V1, 5Z1, 8V3, 10V4, Z6T) and different Sequoia CV Systems. The data collection protocol was standardized across all data collection sites). |
3. Number of Experts and Qualifications for Ground Truth
| AI Feature | Number of Experts | Qualifications |
|---|---|---|
| AI Assist | 3 sonographers | Expert cardiac sonographers (implied by context). |
| 2D HeartAI | 3 examiners | "Examiners" performed all manual contouring and measurements. No specific years of experience mentioned, but expertise is implied for establishing reference standard. |
| 4D HeartAI | 3 examiners | "Examiners" performed all manual contouring and measurements. No specific years of experience mentioned, but expertise is implied for establishing reference standard. |
| AI Measure | At least 3 sonographers | Expert cardiac sonographers (implied by context). |
4. Adjudication Method for the Test Set
| AI Feature | Adjudication Method |
|---|---|
| AI Assist | Consensus-based scoring: Three sonographers scored the results for each placement made by the algorithm. Success was defined as "no adjustment or minor adjustment needed," while "failure" meant a "major adjustment needed." This implies a form of expert consensus without explicit mention of conflict resolution, but rather individual scoring that collectively determined the success rate. |
| 2D HeartAI | Mean value from multiple examiners: Three examiners performed all manual contouring and measurements. Reference standard for each measurement was established by calculating the mean value from the three examiners. Variability was assessed by intraclass correlation (ICC) and inter-reader variability by Pearson correlation and Bland-Altman. This is a form of 3-expert consensus by averaging. |
| 4D HeartAI | Mean value from multiple examiners: Identical to 2D HeartAI. Three examiners performed all manual contouring and measurements. Reference standard for each measurement was established by calculating the mean value from the three examiners. Variability was assessed by intraclass correlation (ICC) and inter-reader variability by Pearson correlation and Bland-Altman. This is a form of 3-expert consensus by averaging. |
| AI Measure | Consensus-based scoring: At least three sonographers scored the results for each measurement made by the algorithm to assess success ("measurements were deemed clinically acceptable"). Similar to AI Assist, this implies a form of expert consensus without explicit mention of conflict resolution. |
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
The document does not describe a multi-reader multi-case (MRMC) comparative effectiveness study to measure the effect size of human readers improving with AI vs. without AI assistance. The studies performed focus on the standalone performance of the AI algorithms, often with user edits being factored in, but not a direct comparison of human performance with and without AI.
6. Standalone Performance Study
Yes, standalone performance studies were done for each AI feature:
- AI Assist: Evaluated the algorithm's ability to successfully place Color Box and Doppler Gate based on expert sonographer scoring of its output (99.3% success). This is a standalone assessment of the algorithm's initial placement accuracy.
- 2D HeartAI: Evaluated the accuracy of the 2D HeartAI "as measured by comparison of 2D HeartAI with user edits compared to a reference standard ground truth." While user edits are mentioned, the reported correlation coefficient and bias (0.81 or higher, bias
§ 892.1550 Ultrasonic pulsed doppler imaging system.
(a)
Identification. An ultrasonic pulsed doppler imaging system is a device that combines the features of continuous wave doppler-effect technology with pulsed-echo effect technology and is intended to determine stationary body tissue characteristics, such as depth or location of tissue interfaces or dynamic tissue characteristics such as velocity of blood or tissue motion. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.