Ultrasound can be used to study tendon and muscle movement. However, quantization is mostly based on manual
tracking of anatomical landmarks such as the musculotendinous junction, limiting the applicability to a small number of
muscle-tendon units. The aim of this study is to quantify tendon displacement without employing anatomical landmarks,
using dedicated speckle tracking in long B-mode image sequences. We devised a dedicated two-dimensional multikernel
block-matching scheme with subpixel accuracy to handle large displacements over long sequences. Images were
acquired with a Philips iE33 with a 7 MHz linear array and a VisualSonics Vevo 770 using a 40 MHz mechanical probe.
We displaced the flexor digitorum superficialis of two pig cadaver forelegs with three different velocities (4,10 and 16
mm/s) over 3 distances (5, 10, 15 mm). As a reference, we manually determined the total displacement of an injected
hyperechogenic bullet in the tendons. We automatically tracked tendon parts with and without markers and compared
results to the true displacement. Using the iE33, mean tissue displacement underestimations for the three different
velocities were 2.5 ± 1.0%, 1.7 ± 1.1% and 0.7 ± 0.4%. Using the Vevo770, mean tissue displacement underestimations
were 0.8 ± 1.3%, 0.6 ± 0.3% and 0.6 ± 0.3%. Marker tracking displacement underestimations were only slightly smaller,
showing limited tracking drift for non-marker tendon tissue as well as for markers. This study showed that our dedicated
speckle tracking can quantify extensive tendon displacement with physiological velocities without anatomical landmarks
with good accuracy for different types of ultrasound configurations. This technique allows tracking of a much larger
range of muscle-tendon units than by using anatomical landmarks.