Advantages of Ultrasound Imaging Diagnosis include non-invasive, painless, and free from ionizing radiation, with high sensitivity and resolution, supporting repeated diagnoses. It is currently one of the best clinical examination methods for real-time observation of the internal structure of the heart. In clinical applications, ultrasound imaging diagnosis is becoming increasingly important.

1. Structural Composition

The ultrasound imaging diagnostic device mainly consists of a main unit (including software), a display, a probe, and accessories (such as a puncture frame, printer, ECG leads, etc.). The probe mainly consists of a transducer, a transmitter, and a connector (which may include a controller).

2. Classification

● Structural types: trolley-type, portable, and handheld (mobile phone type)

● Functions: B-mode ultrasound, color Doppler ultrasound

● Application areas: general ultrasound, cardiac ultrasound, obstetric and gynecological ultrasound, and anesthetic intervention ultrasound

● Target users: human and veterinary ultrasound imaging diagnostic devices

● Probe structures: convex array (micro-convex, large convex), linear array, phased array

● Diagnostic areas: ophthalmology, cardiology, abdomen, cranial, intravascular, and pediatrics

● Application methods: external probes, internal probes, puncture biopsy probes

● Array elements: single-element, multi-element

3. Risk Classification

In China, the risk classification for ultrasound imaging diagnostic equipment in medical device registration management includes Class II and Class III (based on the registered probe). According to the classification directory of medical devices “Announcement on the Release of the Medical Device Classification Directory (2017 No. 104),” ultrasound soft tissue cutting and hemostasis surgical equipment falls under code 06 “Medical Imaging Devices,” with the primary product category being 07 ultrasound imaging diagnostic equipment.

4. Registration Unit Division

Based on different working principles, devices can be divided into analog devices and digital devices (devices using digital beamforming technology). Analog and digital devices should be registered separately according to different registration units. The same registration unit should select typical products based on the coverage of product risk and technical indicators. Typical products should represent the safety and effectiveness of other products within the same registration unit and should prioritize the most complex structure, most complete functions, highest risk, and most comprehensive technical indicators. If the main technical indicators within the same registration unit cannot cover each other, then multiple models should be considered as typical products.

5. Relevant Applicable Standards for Products

Safety GB 9706.1-2020 Medical Electrical Equipment Part 1: General Requirements for Basic Safety and Essential Performance

Safety GB 9706.237-2020 Medical Electrical Equipment Part 2-37: Special Requirements for Basic Safety and Essential Performance of Ultrasound Diagnostic and Monitoring Equipment

Electromagnetic Compatibility YY 9706.102-2021 Medical Electrical Equipment Part 1-2: General Requirements for Basic Safety and Essential Performance Parallel Standard: Electromagnetic Compatibility Requirements and Testing

Performance GB 10152-2009 B-mode Ultrasound Diagnostic Equipment

Performance YY 0767-2009 Color Doppler Ultrasound Imaging System

Performance YY/T 0593-2015 Transcranial Doppler Blood Flow Analyzer

Performance YY/T 1279-2015 Three-Dimensional Ultrasound Imaging Performance Testing Methods

Performance YY/T 1480-2016 Performance Testing Methods for Ultrasound Elastic Imaging Devices Based on Acoustic Radiation Force

Performance YY/T 0108-2008 Ultrasound Diagnostic Equipment M-Mode Testing Methods

Performance YY/T 1142-2013 Performance Testing Methods for Ultrasound Echo Doppler Diagnostic Equipment

Footswitch YY/T 1057-2016 General Technical Conditions for Footswitches

Environment GB/T 14710-2009 Environmental Requirements and Testing Methods for Medical Electrical Equipment

Biological Evaluation GB/T 16886.1-2022 Biological Evaluation of Medical Devices Part 1: Evaluation and Testing in the Risk Management Process

Biological Evaluation GB/T 16886.5-2017 Biological Evaluation of Medical Devices Part 5: In Vitro Cytotoxicity Testing

Biological Evaluation GB/T 16886.10-2017 Biological Evaluation of Medical Devices Part 10: Tests for Irritation and Skin Sensitization

6. Key Indicators Related to Technical Standards

GB 10152 – 2009 Deviation of working frequency from nominal frequency within ±15%

GB 10152 – 2009 Detection depth meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Lateral/axial resolution meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Blind area meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Slice thickness meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Lateral/longitudinal geometric position accuracy meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Perimeter and area deviation within ±20% or meets the indicators published by the manufacturer in the random documents

GB 10152 – 2009 M-mode performance indicators meet the indicators published by the manufacturer in the random documents

GB 10152 – 2009 Three-dimensional reconstruction volume calculation deviation within ±30% or meets the indicators published by the manufacturer in the random documents

YY 0767 – 2009 Color Doppler blood flow mode detection depth should not be less than the published value in the random documents

YY 0767 – 2009 Coherence test between color and B-mode images should show no significant misalignment between the simulated blood flow color image and the grayscale image of the pipeline

YY 0767 – 2009 Blood flow direction recognition ability test should show color change in the blood flow graph when the blood flow direction is changed

YY 0767 – 2009 Spectral Doppler mode detection depth test should not be less than the published value in the random documents

YY 0767 – 2009 Flow velocity measurement error test should meet the indicators published by the manufacturer in the random documents

YY 0767 – 2009 Sampling area cursor position accuracy test should show no significant spectral signal outside the pipeline

YY/T 0593 – 2015 Ultrasound working frequency deviation should not exceed ±10%

YY/T 0593 – 2015 Flow velocity measurement range should meet the indicators published by the manufacturer in the random documents

YY/T 0593 – 2015 Flow velocity measurement error should not exceed ±20%

YY/T 0593 – 2015 Working distance should meet the maximum and minimum working distances published by the manufacturer in the random documents

YY/T 0593 – 2015 Distance gating error should not exceed the error value published by the manufacturer

YY/T 1279 – 2015 Ultrasound output power should meet the indicators published by the manufacturer in the random documents

YY/T 1279 – 2015 Detection depth (three-dimensional) should meet the indicators published by the manufacturer in the random documents

YY/T 1279 – 2015 Blind area (three-dimensional) should meet the indicators published by the manufacturer in the random documents

YY/T 1279 – 2015 Lateral/axial resolution (three-dimensional) should meet the indicators published by the manufacturer in the random documents

YY/T 1279 – 2015 Geometric position accuracy (three-dimensional) should meet the indicators published by the manufacturer in the random documents

YY/T 1279 – 2015 Volume measurement should not exceed ±30%

GB 9706.237 – 2020 Acoustic field testing NMPA and CE have no restrictions on MI, TI values, FDA has restrictions on MI, TI, and I based on usage “limited value requirements

7. Considerations for NMPA Registration (For Reference Only)

First Point: Product Classification: Class II or Class III?

Check if it has AI diagnostic functions (such as automatic fetal heart rate measurement/tumor grading); Class II: pure imaging devices (like B-mode ultrasound machines); Class III: diagnostic-grade AI software etc. (such as benign or malignant judgment of breast nodules); avoid pitfalls: if the main unit + AI software combination, register according to the highest category etc.

Second Point: Registration Path

Hardware Related: Probe frequency range (do not write “high frequency,” write clearly 2-18MHz); electromagnetic compatibility report (2023 new regulations: must use YY 0505 new standards)

Software Related: Core: algorithm white-boxing ( reviewers are picky about this) image processing algorithms (such as filtering and noise reduction) should be commented line by line; AI functions should provide proof of training data source (hospital ethics approval) etc.

Network Security: conduct attack testing report (simulate hacker intrusion)

Clinical Evaluation: comparison with similar products (saves time and money), find already certified reference products? Prove your probe accuracy ≥ reference products (key: comparison of acoustic output parameters); clinical trials (time-consuming and costly): Class III with AI must do? Sample size ≥ ? cases; find non-cooperative hospitals (reviewers may not trust data from manufacturer-associated hospitals?) etc.

Third Point: System Assessment

Design and development documentation: traceability from requirement documents to code testing records (focus on change records) etc.

Production site: monitoring records of temperature and humidity in the probe workshop, software burning isolation, etc.

Fourth Point: Review Phase

Raise high-frequency red flags: AI function data bias not explained (for example, training data is all from top-tier hospitals, can it be used in grassroots settings?); safety values of acoustic output lack verification; clinical report missing statistical items.

Fifth Point: Post-Certification Attention

Changes must be reported: change probe supplier? Redo biocompatibility testing ? software upgrades? Follow change process etc.

Mock inspections: simulate sudden inspections by the drug administration every year, randomly retrieve production records, hacker attack and defense testing etc.

8. Important Notes

a） For ultrasound imaging diagnostic devices using pulsed wave signals, acoustic field testing needs to confirm synchronization signals; otherwise, acoustic field scanning cannot be performed; for continuous wave signal acoustic field scanning, synchronization signals are not required.

b） Performance indicators are determined by product technical requirements, and performance indicators must meet or exceed the requirements in GB 10152 standards.