One conventional approach to monitoring heart function is transesophageal echocardiography (TEE), in which a probe containing an ultrasound transducer is inserted into the esophagus and positioned so that the transducer is near the patient's heart. The probe is then used to capture video images of the heart in real time, which are typically displayed on a video monitor. Conventional TEE probes typically measure between 10-15 mm in diameter (for adults). Because of this large diameter, conventional TEE often requires anesthesia, can significantly threaten the airway, and is not well suited for long-term monitoring of the heart. More recently, smaller TEE probes that permit long term monitoring and eliminate or reduce the need for anesthesia have been developed, as disclosed in US2005/0143657 (application Ser. No. 10/996,816, filed Nov. 24, 2004).
Under certain circumstances, operating a TEE probe at a high frame rate can cause the probe's temperature to increase by an undesirable amount. This can be particularly problematic with smaller probes, where less surface area is available for dissipation of the applied power.
To keep the temperature of the ultrasound probe down, the probe is operated at a low frame rate (with correspondingly low heat generation) for the vast majority of time. Probe operation is only switched to the high frame rate temporarily at times when high temporal resolution is needed, and the probe is only operated at the high frame rate for a short period of time. During that time, a burst of images with high temporal resolution is obtained. After capturing the short burst of images, the frame rate is cut back, which reduces the generation of heat.
All other things being equal, the temperature rise (ΔT) of an ultrasound probe is proportional to the frame rate of the images. Heat management in ultrasound probes may therefore be implemented by switching the image-capture frame rate between two modes: a first mode with a high frame rate, for use at times when that high frame rate is required for the imaging task at hand, and a second mode with a low frame rate for use at times when that high frame rate is not required. Optionally, a third mode with an even lower frame rate (or a frame rate of zero) may be used to allow the probe to cool down faster after it is operated in the first mode.
For TEE imaging of the heart, a frame rate of 50 frames per second (fps) is suitable because that frame rate is fast enough for high speed applications (e.g., visualizing which portion of the heart is contracting late in a cardiac resync therapy application), yet slow enough for “speckle” noise from the blood to re-randomize (thereby reducing the impact of speckle noise on the image). However, with a small TEE probe (e.g., with a transducer with a surface area on the order of 200 mm2), operating the probe at 50 fps continuously can cause an undesirable increase in the probe's temperature.
In these applications, however, continuous operation at the high frame rate can be avoided by operating the probe at a lower frame rate (e.g., on the order of 25 fps), during the portion of the procedure when the probe is being positioned. Preferably, this lower frame rate is slow enough such that ΔT drops to the point where the probe can be safely operated continuously. This mode of operation is referred to herein as the “continuous” mode. Then, after the probe is positioned at its desired location, a clip of images are captured at the higher frame rate (e.g., on the order of 50 fps) for a short period of time (e.g., 3 seconds). This mode of operation is referred to herein as the “burst” mode. After the high speed burst, operation returns to the continuous mode. The changing frame rate in this mode of operation is depicted by the lower trace 12 in
Optionally, a low frame rate “cool down” mode may be added after the burst mode to speed up the cooling of the probe, as illustrated in
In one example, the temperature rise of a probe operating continuously at 24 fps was about 3.2° C. (measured with a tissue phantom). Since the thermal time constant of ultrasound probes is relatively large, a three second high speed burst does not cause a significant instantaneous increase in temperature. As long as the high speed bursts are relatively short (e.g., less than five seconds, and more preferably on the order of three seconds) and the cool-down periods are sufficiently long, ΔTMAX will never be too much higher than the ΔT that is associated with continuous operation.
All other things being equal, power and thus relative heat generation is directly proportional to frame rate. Thus, operating at 50 fps will generate 50/24=2.08 times as much heat as in the 24 fps normal mode, and operating in a 10 fps cool-down mode will generate 10/24=0.41 times as much heat as in the 24 fps normal mode. The cool down period is preferably sufficiently long and sufficiently close enough in time to the high speed burst so as to compensate for the entire temperature rise caused by operating at the high frame rate. This can be accomplished by keeping the total frame count below the frame count associated with the normal continuous frame rate. For example, in a system operating continuously at 24 fps, there would be 288 frames in a 12 second interval. In contrast, doing a three second burst at 50 fps followed by a nine second cool-down at 10 fps would result in a lower frame count of (50×3)+(10×9)=240 frames in 12 seconds. As a result, the temperature at the end of the burst and cool-down cycle should be lower than the steady-state temperature for the continuous mode. In another example, doing a three second bust at 50 fps followed by a seven second cool-down at 10 fps, followed by another similar burst and cool-down would result in a lower total frame count (440) than continuous operation at 24 fps for 20 seconds (which yields in a frame count of 480).
The above-described techniques are particularly useful to achieve compliance with the FDA's ALARA (“as low as reasonably achievable”) principle for heating. This is particularly important for applications in which the probe remains installed in the patient for long periods of time (e.g., over 6 hours) to minimize negative consequences from long term heating by only capturing short burst at a high frame rate as required, and otherwise running at a low frame rate. Note that a 3 second burst is sufficiently long to capture two complete cardiac cycles when the heart is beating at 60 bpm.
The above-described techniques are also useful for keeping the temperature rise below the maximum temperature rise of 6° C. recommended under IEC60601, and more preferably below 4° C.
The controller 44 control sends commands to the ultrasound engine 46 via a suitable interface (e.g., a digital bus) to change the frame rate in accordance with the approaches described above in connection with
The user interface 42 permits the user to request initiation of the burst mode at any desired time, and may be implemented using any of a variety of approaches that are well known to persons skilled in the relevant arts. Examples include mechanical switches (including but not limited to pushbuttons and foot-operated switches) or virtual switches (e.g., on a touch screen).
The control surface for the user interface may be located on the ultrasound console or on the probe itself.
Note that a frame rate of 50 fps provides 20 ms time resolution. So does a sequence of averages of frames, frame 1 averaged with frame 2, frame 2 averaged with frame 3, frame 3 averaged with frame 4, etc. 20 ms is long enough for speckle from blood to “re-randomize” that is, longer than a speckle residence time, reducing the coefficient of variation of the average of an ensemble of resolution cells in the cavity (dominated by speckle which follows a Rayleigh distribution or electronic noise) in a compounded frame (compounded from two successive frames) to 71% of the similar coefficient of variation in a single frame.
Optionally, pulses with a duration of very few cycles (e.g., 1-3 cycles) may be used to excite the transducer. Since shorter pulses (in the time domain) have a relatively wide bandwidth (in the frequency domain), and attenuation in tissue is higher for higher frequency components, the low frequency components of the wide-bandwidth pulse can provide improved penetration (e.g., on the order of 10-12 cm) using a center frequency on the order of 6 MHz, without sacrificing resolution. Experimentation showed that using a two cycle pulse 30 (see
Other strategies for reducing temperature rise that may also be used to reduce power include the following: (a) making the dimensions of the transducer as large as possible consistent with a small diameter probe (b) making the frequency as low possible consistent with adequate resolution; (c) designing for thermal conductivity to spread the heat out to reduce ΔT at the hottest point on probe, e.g., as described in application Ser. No. 11/534,403, filed Sep. 22, 2006, which is incorporated herein by reference; (d) keeping the duty cycle low, consistent with acceptable image quality; and (e) using signal processing to improve the image, e.g., as described in application Ser. No. 10/997,059, filed Nov. 24, 2004, which is incorporated herein by reference.
Note that while the above-described techniques are explained in the context of TEE, the same techniques may also be use in other ultrasound imaging contexts, e.g., with probes that is inserted into cavities other than the esophagus, or probes that are applied to external surfaces of a body. Numerous modifications to the above-described embodiments will be apparent to those skilled in the art, and are also included within the purview of the invention.
This application claims the benefit of U.S. provisional application 60/941,480, filed Jun. 1, 2007.
Number | Name | Date | Kind |
---|---|---|---|
6056691 | Urbano et al. | May 2000 | A |
6083168 | Hossack et al. | Jul 2000 | A |
6542846 | Miller | Apr 2003 | B1 |
6663578 | Salgo | Dec 2003 | B1 |
6709392 | Salgo | Mar 2004 | B1 |
7678048 | Urbano et al. | Mar 2010 | B1 |
20030028113 | Gilbert | Feb 2003 | A1 |
20030045795 | Bjaerum | Mar 2003 | A1 |
20060165179 | Feuer et al. | Jul 2006 | A1 |
20070083121 | Hastings | Apr 2007 | A1 |
Number | Date | Country |
---|---|---|
1614386 | Jan 2006 | EP |
WO 2006023984 | Mar 2006 | WO |
WO 2007000680 | Jan 2007 | WO |
Number | Date | Country | |
---|---|---|---|
20080298654 A1 | Dec 2008 | US |
Number | Date | Country | |
---|---|---|---|
60941480 | Jun 2007 | US |