IEC 62359: Ultrasonics Field Characterization — Thermal and Mechanical Indices for Diagnostic Safety

Diagnostic ultrasound is one of the most widely used medical imaging modalities, valued for its real-time capability, portability, and—most importantly—its excellent safety record. However, ultrasound energy is not entirely benign: it can heat tissue through absorption and can cause cavitation through mechanical pressure variation. To ensure that diagnostic ultrasound remains safe across all applications—from fetal imaging to transcranial Doppler—the international community developed IEC 62359.

IEC 62359, titled “Ultrasonics – Field characterization – Test methods for the determination of thermal and mechanical indices related to medical diagnostic ultrasonic fields,” defines the standardized test methods for calculating the two key safety indices displayed on every diagnostic ultrasound system: the Thermal Index (TI) and the Mechanical Index (MI).

📋 The Safety Indices — TI and MI

Before IEC 62359, ultrasound safety was assessed through a complex set of intensity parameters (SPTA, SPPA, etc.) that required expert interpretation. The TI and MI simplify this by providing two dimensionless numbers that any operator can use to assess relative risk.

Index	Symbol	Physical Meaning	Formula Basis	Typical Display Range
Thermal Index	TI	Ratio of emitted power to power required to raise tissue temperature by 1°C	TI = W / W_deg	0.1 – 6.0
Mechanical Index	MI	Likelihood of cavitation (gas bubble formation) in tissue	MI = p_r(f_c) / √f_c	0.1 – 1.9

Three variants of TI exist to cover different scanning conditions:

TIS (Soft Tissue TI): For abdominal, cardiac, and other soft-tissue imaging where bone is not in the beam path
TIB (Bone TI): For applications where ultrasound impinges on bone at or near the focus (e.g., fetal imaging in the second/third trimester)
TIC (Cranial TI): For transcranial applications where the beam passes through bone at the surface

💡 Engineering Insight: The MI formula uses the derated peak rarefactional pressure p_r divided by the square root of the center frequency f_c. The 1/√f dependency is physically significant: it reflects the fact that at higher frequencies, the rarefactional half-cycle is shorter, giving less time for cavitation nuclei to grow to resonant size. This means that higher-frequency transducers intrinsically have lower MI for the same pressure amplitude—a built-in safety margin for high-resolution imaging.

🔬 Measurement Methods — From Hydrophone to Index

The core of IEC 62359 is the definition of precise, reproducible measurement methods for determining TI and MI from acoustic output measurements. The standard specifies the entire measurement chain:

Hydrophone Measurement

Acoustic output is measured using a calibrated hydrophone (typically a membrane hydrophone or needle hydrophone with a sensitive element diameter of 0.2–1.0 mm) scanned through the ultrasound field in three dimensions. The hydrophone captures the temporal pressure waveform at each spatial location.

Measurement Parameter	Symbol	Condition	Derived Index
Peak rarefactional pressure (derated)	p_r	Attenuation 0.3 dB/cm-MHz	MI = p_r / √f_c
Output power	W	Total acoustic power from transducer	TI = W / W_deg
Center frequency	f_c	From power spectrum at point of peak p_r	Used in both MI and TI
Beam cross-sectional area	A_aprt	At -6 dB from peak intensity	Used in TIS calculation
Derated intensity spatial peak	I_SPTA.3	Derated at 0.3 dB/cm-MHz	Used in TIB, TIC

⚠️ Critical Derating Assumption: The standard specifies a derating coefficient of 0.3 dB/cm-MHz to estimate in-situ exposure through tissue. This coefficient represents a conservative average of soft tissue attenuation. However, for specific applications like neonatal cephalic imaging (where the skull is thin or absent) or ophthalmic imaging (where the fluid-filled eye has lower attenuation), the standard 0.3 dB/cm-MHz may not be appropriate. The standard allows for alternative derating values if clinically justified, but this must be documented in the operator’s manual.

📊 Broadband Transducer Considerations

Modern diagnostic ultrasound systems increasingly use broadband transducers that transmit pulses spanning several octaves (e.g., 2–12 MHz). These transducers enable harmonic imaging and coded excitation but present challenges for the measurement methods in IEC 62359, which were originally developed around narrowband pulses.

Center Frequency Ambiguity

The MI formula requires a single center frequency f_c. For broadband pulses, the definition of center frequency can significantly affect the MI value. IEC 62359 specifies that f_c should be calculated as the arithmetic mean of the -6 dB frequencies of the power spectrum at the point of peak rarefactional pressure. For highly broadband pulses, different methods (temporal zero-crossing, spectral centroid, or -6 dB bandwidth mean) can yield f_c values differing by 20% or more, leading to corresponding MI variations.

f_c Determination Method	MI for 4 MHz Nominal Transducer	Relative Difference
Spectral centroid	1.25	Reference
-6 dB bandwidth mean	1.18	-5.6%
Temporal zero-crossing	1.32	+5.6%
Peak frequency	1.40	+12%

✅ Best Practice for Broadband Transducer Testing: Use the spectral centroid method (as recommended by later editions of IEC 62359) for f_c determination. This method is the most robust against noise and best represents the frequency weighting of the cavitation mechanism. Always document the center frequency determination method in your test report—regulatory reviewers will check this.

🔧 Thermal Index Models — 1D and 3D

IEC 62359 defines three models for calculating the thermal index, depending on whether the ultrasound beam is scanned (imaging mode) or stationary (Doppler/pulsed-wave mode):

1-D Model (Stationary Beam)

For non-scanning modes (spectral Doppler, pulsed wave Doppler), the standard uses a one-dimensional model assuming heat deposition along the beam axis. The temperature rise is proportional to the acoustic power divided by the beam width at the focus. This model is conservative for stationary beams because it ignores lateral heat diffusion.

3-D Model (Scanned Beam)

For B-mode imaging where the beam is scanned across the tissue, the heat is distributed over a larger volume. The 3-D model accounts for this by using the scanned beam area rather than the stationary beam cross-section. The TIS in scanned mode is typically 3–10 times lower than the equivalent stationary beam TI.

🚨 Important Clinical Implication: When switching from 2D imaging mode to spectral Doppler mode, the TI can increase dramatically—often by a factor of 5–10—because the beam becomes stationary and all acoustic energy is concentrated on one tissue volume. This is why the ALARA (As Low As Reasonably Achievable) principle is particularly important for Doppler examinations. Operators should minimize spectral Doppler exposure duration, especially when the TI exceeds 1.0.

❓ Frequently Asked Questions

Q1: What is the maximum safe limit for TI and MI?

There is no absolute “safe limit” defined in IEC 62359 itself, but the US FDA track 3 limits and the IEC 60601-2-37 collateral standard specify: MI ≤ 1.9 for all modes, TIS ≤ 1.0 (scanned)/≤ 2.0 (non-scanned), TIB ≤ 1.5, TIC ≤ 1.0 (for ophthalmic applications; ≤ 0.23 for fetal imaging in some jurisdictions). These limits are based on epidemiological evidence and are considered conservative.

Q2: Can TI and MI be calculated from simulation rather than measurement?

Yes, for design-phase assessment. IEC 62359 allows the use of validated computational models (e.g., k-Wave, Field II, or finite-difference time-domain simulations) for estimating TI and MI during transducer development. However, final compliance testing for regulatory submission must use physical hydrophone measurements on production-representative devices. The simulation uncertainty must be quantified and included in the declared values.

Q3: How does the hydrophone bandwidth affect MI measurement accuracy?

Hydrophone bandwidth is critical. The rarefactional pressure of a broadband pulse contains high-frequency components that a limited-bandwidth hydrophone may attenuate, leading to an underestimation of p_r and therefore MI. IEC 62359 requires the hydrophone bandwidth to extend to at least the fifth harmonic of the fundamental frequency. For a 5 MHz transducer, the hydrophone must have useful response to at least 25 MHz. Membrane hydrophones (e.g., Precision Acoustics, Onda) typically meet this requirement; some needle hydrophones may not.

Q4: What is the difference between IEC 62359 and AIUM/NEMA UD-2?

IEC 62359 is the international standard used primarily outside the United States, while AIUM/NEMA UD-2 (the “Output Display Standard”) is the equivalent US document. Since the 2010s, these standards have been harmonized to a large degree, with only minor differences in specific calculation methods and reporting requirements. The FDA accepts compliance with either standard for 510(k) submissions.