Experimental setup for data acquisition of needle insertions into an ex-vivo porcine phantom

Experimental setup for data acquisition of needle insertions into an ex-vivo porcine phantom

Source publication
Article
Full-text available
Purpose Percutaneous needle insertion is one of the most common minimally invasive procedures. The clinician’s experience and medical imaging support are essential to the procedure’s safety. However, imaging comes with inaccuracies due to artifacts, and therefore sensor-based solutions were proposed to improve accuracy. However, sensors are usually...

Contexts in source publication

Context 1
... same experimental setup used in [9] is employed in this work ( Fig. 1) to acquire audio and force datasets at different insertion velocities. For the audio acquisition, a stethoscope connected to a microphone attached to the needle's proximal part (18G, 200 mm length biopsy needle-ITP, Germany) via ...
Context 2
... Fig. 10 an example of event detection in I F c and the corresponding audio indicator signal intensity inside the window is shown. Two significant events were detected in the force, and then two windows are generated for analyzing the audio ...
Context 3
... means that the related information between force and audio is located mainly in the low-frequency dynamics. Figure 11 shows the histograms of the Pearson coefficients between the three force indicators and the audio indicator for the four tested insertion velocities. The histograms confirm the results shown in Table 2: for the three insertion velocities, the curvature is the force indicator that better correlates with the audio indicator. ...
Context 4
... analysis made in Fig. 11 is confirmed by the cumulative histograms displayed in Fig. 12, where it is possible to observe that for the four tested velocities, more than 70% of the correlations between the force curvature indicator I F c and the audio indicator involve coefficients higher than 0.6. Concerning the derivative force indicator I F d , more than 60% ...
Context 5
... analysis made in Fig. 11 is confirmed by the cumulative histograms displayed in Fig. 12, where it is possible to observe that for the four tested velocities, more than 70% of the correlations between the force curvature indicator I F c and the audio indicator involve coefficients higher than 0.6. Concerning the derivative force indicator I F d , more than 60% of the recordings, regardless of the insertion velocity, ...
Context 6
... the force intensity indicator I F i seems to have low values compared to the other two indicators, the coefficient values of more than 50% of the recordings are over 0.5, which can be considered high if we consider the completely different nature between audio and force sensors. Figure 13 gives us an idea of how the force curvature indicator and the audio indicator, which present high or middle-high Pearson coefficient value (displayed in percentage in the figure), looks like. The figure shows eight recordings taken from the implemented database; two recorded It is important to point out in Fig. 13 that the higher the velocity, the smoother is the force signal. ...
Context 7
... audio and force sensors. Figure 13 gives us an idea of how the force curvature indicator and the audio indicator, which present high or middle-high Pearson coefficient value (displayed in percentage in the figure), looks like. The figure shows eight recordings taken from the implemented database; two recorded It is important to point out in Fig. 13 that the higher the velocity, the smoother is the force signal. This is because the force change with the needle insertion speed. When the needle insertion velocity increases, the force decreases [8]. When the velocity is low, tiny tissue breakage of internal structures results in peaks in the force signal. When the velocity is higher, ...
Context 8
... decreases [8]. When the velocity is low, tiny tissue breakage of internal structures results in peaks in the force signal. When the velocity is higher, there is no reaction of force for these structures, and it is possible only to identify reactions in the force resulting from significant punctures. This property can also explain what happens in Fig. 11 where the higher the insertion velocity, the higher the percentage of recordings having high correlation values between audio and force indicator since only the main puncture events are ...
Context 9
... a high percentage of recordings present correlations between force and audio indicators over 0.5, there are cases (see Fig. 11) where low correlations are obtained. Fig- ure 14 displays two examples of low correlations (ρ < 0.4) for velocities of 3 mm/s and 10 mm/s. These low correlations may have various reasons. First, a force signal peak can sometimes occur with no tissue rupture, which is always necessary to create an audio excitation. Another possible ...
Context 10
... it should be noted that recordings displaying low correlation in the histogram of Fig. 11 are not necessarily the same for each force indicator. A correlation between audio and two different force indicators can result in different correlation coefficients. This can occur because different force indicators can enhance different characteristics of the force signal that may or may not be correlated with the audio envelope. ...
Context 11
... of Fig. 11 are not necessarily the same for each force indicator. A correlation between audio and two different force indicators can result in different correlation coefficients. This can occur because different force indicators can enhance different characteristics of the force signal that may or may not be correlated with the audio envelope. Fig. 12 Cumulative histograms of the Pearson coefficients (in terms of percentage of nb. of recordings) for all comparisons between the force indicators and audio indicator, and for the four tested velocities Event-to-event correlation Figure 15 shows the scatter plots of the maximal values of the detected events for force and audio indicators ...
Context 12
... can occur because different force indicators can enhance different characteristics of the force signal that may or may not be correlated with the audio envelope. Fig. 12 Cumulative histograms of the Pearson coefficients (in terms of percentage of nb. of recordings) for all comparisons between the force indicators and audio indicator, and for the four tested velocities Event-to-event correlation Figure 15 shows the scatter plots of the maximal values of the detected events for force and audio indicators (as explained in Sect. 3.1.3) ...