Search Results

The Digital Color Doppler Ultrasound System is a general-purpose ultrasonic imaging instrument intended for use by appropriately trained healthcare professionals in hospital for evaluation of Fetal, Abdominal, Pediatric, Small Organ (breast, testes, thyroid), Cephalic (neonatal and adult), Transrectal, Trans-vaginal, Peripheral Vascular, Cerebral Vasculoskeletal (Conventional and Superficial), Cardiac (pediatric and adult), Trans-esoph (Cardiac), Laparoscopic, OB/Gyn and Urology.

This SonoScape P60 Series Digital Color Doppler Ultrasound System is an integrated preprogrammed color ultrasound imaging system, capable of producing high detail resolution intended for clinical diagnostic imaging applications.

The basic principle is that system transmits ultrasonic energy into patient body and implements post processing of received echoes to generate onscreen display of anatomic structures and fluid flow within the body.

This system is a Track 3 device that employs a wide array of probes that include linear array, convex array phased array and etc.

This system consists of a mobile console with touch screen and keyboard control panel, power supply module, color LCD monitor and optional probes.

This system is a mobile, general purpose, software controlled, color diagnostic ultrasound system. Its basic function is to acquire ultrasound data and to display the image in B-Mode (including Tissue Harmonic Image), M-Mode, TDI, Color-Flow Doppler, Pulsed Wave Doppler, Continued Wave Doppler, Power Doppler, Direct Power Doppler, or the combination of these modes, Compound, 3D/4D, Elastography and Contrast.

Here's a breakdown of the acceptance criteria and the study details for the S-FetusAI feature of the Sonoscape Digital Color Doppler Ultrasound System (P60 Series), based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Clinical Performance:

• Precision of Sp (spinal cord cone end positioning). | S-FetusAI has a good consistency with the manual measurement.
  The Sp (spinal cone end positioning) precision was basically equal between the trial group and control group. |

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

• S-Fetus-Recognition Test Set:
  • More than 200 images for each of the 14 standard sections.
  • Total: 3717 images.
• S-Fetus-Measure Test Set:
  • More than 250 images for each of the 9 standard sections.
  • Total: 2929 images.
• Data Provenance: A mix of retrospective and prospective data collection in clinical practice. The test sets were collected at three hospitals in China.

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

• Initial Ground Truth Drawing: Five OB/GYN experts with more than 3 years' experience.
• Cross-checking: Two experts with more than 20 years' experience.
• Re-checking: Four experts with more than 30 years' experience.
• Clinical Performance Study (Adjudication of results): Two Readers from an independent evaluation group, consisting of qualified professionals (senior experts who had been involved in prenatal ultrasound examination).

4. Adjudication Method for the Test Set

The ground truth was established through a multi-level expert consensus process:

1. Initial Drawing: 5 OB/GYN experts.
2. Cross-checking: 2 senior experts.
3. Re-checking: 4 very senior experts.
   This indicates a robust, multi-stage consensus-based adjudication, effectively a form of "expert consensus with multiple layers of review."

For the clinical performance testing, a separate "independent evaluation group" with 2 senior experts evaluated the section recognition results and precision of Sp from both the trial (AI-assisted) and contrast (manual) groups.

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

• A clinical performance testing was conducted, which involved comparing the S-FetusAI function (Trial group) with manual recognition and measurement by investigators (Contrast group). This can be considered a comparative effectiveness study.
• Research Group Composition (Contrast Group): Investigators (physicians with more than 3 years of specialized training in obstetric ultrasound examination; total 7 investigators) performing manual recognition and measurement.
• Effect Size: The study focused on demonstrating non-inferiority rather than a specific improvement percentage for human readers with AI assistance. The conclusion states that "the recognition precision of the standard sections of S-FetusAI was not inferior to that of the manual scan." This implies that the AI-assisted system performs at least as well as, or comparably to, manual recognition. It does not explicitly quantify how much human readers improve when using AI, but rather that the AI's standalone performance (for recognition) and consistency/precision (for measurement) are comparable to manual methods.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, a standalone performance testing was done for the S-FetusAI feature for both S-Fetus-Recognition and S-Fetus-Measure. The results (90.01% accuracy for recognition and 85% accuracy for measurement) are reported as entirely algorithmic performance against established ground truth.

7. The Type of Ground Truth Used

The ground truth used was expert consensus. It was established by multiple OB/GYN experts with varying levels of experience, going through a process of initial drawing, cross-checking, and re-checking. Any images not meeting inclusion/exclusion criteria were excluded from the ground truth set.

8. The Sample Size for the Training Set

The document explicitly states: "The test set used to test the standalone performance of S-FetusAI are independent of the training data set and tuning data set". However, the sample size for the training set is not provided in the given text.

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

While the document implies that training and tuning datasets were used, the specific method for establishing their ground truth is not mentioned in the provided text. It only details the ground truth establishment for the test set.

Ask a specific question about this device