|
OVERVIEW
|
|
|
The
main
advantage of pulsed Doppler ultrasound is its capability to
provide spatial information associated to velocity values.
However, in the Doppler modes, both the weak flow
signals and the large-magnitude echoes from
the
surrounded tissue (commonly referred as a clutter) are simultaneously
converted into
digital. Thus, to avoid ADC saturation, the analog gain should be set
in accordance with the clutter swing.
|
It is
widely accepted that signals
from the blood scatterers can be 40-100 dB weaker than echoes from
clutter. Consequently, to
map the flow signals into sufficient amount of bits, a high-resolution
ADC should be employed. Otherwise, the receive
channel may provide an unacceptably low flow resolution due to the
quantization noise. For instance, having a -60 dB
flow-to-clutter ratio
and a 12-Bit ADC, the flow signal resolution is only 2 Bit.
|
| High-resolution
(> 12 Bit) ADCs are more expensive, operate
at lower sampling frequency, and consume more power. To avoid its
implementation, a number of commercial ultrasound scanners
incorporates a separate analog beamformer for Doppler processing that
increases the system complexity and cost. Therefore, it would be
beneficial to develop a technology allowing acquisition of Doppler data
by a regular ADC. |
| In the
early phase of Doppler imaging, it has been proposed to move the
ADC within a feedback loop. This technique seeks to
arrange an ADC so that the large-amplitude echoes
caused by stationary and near-stationary targets are filtered out prior
to A-to-D conversion. Operating as a stationary
canceller (or wall filter), such converter incorporates an
extrapolative function by
which a predicted input signal value is compared with the actual one to
generate an error signal and the error signal is converted to
digital. This architecture is known as a Predictive
ADC. The same technique is often referred as Predictive
Coding, Differential Pulse Code Modulation (DPCM), or Delta Modulation.
|
|
|
|
| ADVANCED
DOPPLER DATA ACQUISITION |
|
|
|
We, at
EchoeScan, have further
developed the concept of
tracking/predictive ADC. The proposed
converter exhibits a feedback loop comprising a conventional ADC, a
DAC, a predictive filter, and a fixed gain
amplifier.
|
|
 |
|
The
above arrangement operates as a closed-loop high-pass
filter
whose
frequency response has a steep
transition around zero Hz and substantial flatness. According to
the original concept, the amplitude resolution of
tracking ADC is dictated by the dynamic range of the DAC.
However, implementing a high-resolution DAC per
channel would cause the same kind of problems as those associated with
high-resolution ADCs. On the contrary, the proposed architecture
utilizes ADC and DAC of the same resolution. A
novel prediction
error
filter contributes not only to cancellation of stationary
echoes but
simultaneously
boosts the weak flow signals. Thus, these signals
are moved into higher bits of the ADC, thereby
improving the SNR for Doppler modalities.
The input-referred frequency response of the predictive ADC is
shown below.
|
|
 |
|
| A
rigorous analysis and
thorough simulation of the proposed technique
reveal a 4-5 bits reduction in required word-length for a tracking ADC
comparing with a conventional counterpart. |
|
|
|
|
Home
Site
Map
Links
