This proof-of-concept correlated events per unit distance with the material structure. By puncturing foams with varying air pocket concentrations and analyzing the amplitude events in the resulting vibroacoustic waveforms, it was shown that the distribution of these events reflect key structural features of the material, offering a potential new method for real-time event-based tissue differentiation.

In this study, the temporal peak distribution of vibroacoustic signals was analyzed by examining the number of peaks per traveled unit distance. The analysis of foam structure revealed a correlation between the number of air pockets/10 mm and the number of amplitude peaks detected in the signal. Specifically, Foam 1, which had an average of 21.7 air pockets/10 mm, produced a significantly higher number of peaks/10 mm compared to Foam 2, which had an average of 13.6 air pockets/10 mm. This correlation was consistently observed through both visual counting and signal analysis, underscoring the significant impact of material structure on the vibroacoustic response. These findings show that temporal peak patterns provide valuable insights into material properties akin to how the human sense of touch interprets fine textures.

Consequently, the influence of different puncture speeds on the vibroacoustic signal was investigated. A robot arm was used to puncture a foam phantom at two constant speeds: 10 mm/s and 20 mm/s. The results showed that faster puncture speeds produced more peaks within a shorter time frame, which was expected, as the needle punctured the air pockets more quickly. This was especially observed at the beginning of each puncture, where the compression and rapid relaxation of the foam resulted in an interval with a higher density of peaks. At this point, the needle passed the air pockets faster than with the average speed setting but with unknown exact speed and therefore, this interval was removed for further analysis. Similarly, other studies observed an increased root-mean-square velocity (RMS velocity) with higher movement speeds [18]. The RMS velocity represents the energy of a system that is related to the frequency and amplitude of vibrations [18]. Therefore, a higher number of distinct peaks per distance also contributes to a higher RMS velocity. When the number of peaks was calculated relative to the distance traveled rather than the time, the puncture speed did not significantly alter the number of events per unit distance. This supports the hypothesis that the events are directly related to the number of air pockets encountered and thus correspond to the material structure. The stability of these results across varying speeds highlights the reliability of vibroacoustic signals for real-time applications, provided the movement speed is known.

Apart from different puncturing speeds, another important variable in clinical settings is the type of needle used, as its bevel shape can influence the needle's interaction with the material and thus FIV. The Quincke style needles used in this study are frequently used for spinal anesthesia and lumbar punctures featuring a sharp cutting bevel that facilitates easy penetration through skin, tissue, and ligaments. Therefore, those conventional needles are often referred to as “cutting needles” [24]. However, Quincke needles are known to often cause postdural puncture headache (PDPH) by creating tension on the meninges aroused by the hole created in the dural tissues [25]. An alternative is offered by so-called non-cutting needles with a blunt pencil point tip, such as the Sprotte or Whitacre models. Those atraumatic needles are designed to spread tissues rather than cut them, therefore significantly reducing the risk of PDPH [24, 25]. Since vibroacoustic signals arise from friction between the needle and tissue, it is hypothesized that a different needle tip, especially when changing from a cutting to a non-cutting model, could alter the signal profile, specifically reducing the number of peaks/10 mm. Even in synthetic foams, non-cutting needles are likely to displace rather than puncture fibers and air pockets, resulting in fewer signal events. This hypothesis merits further experimental validation.

While this study demonstrated the feasibility of using information from vibroacoustic signals to differentiate materials, several limitations that must be addressed to fully realize the potential of this approach. The experiments were confined to synthetic foam materials, limiting the generalizability of the findings. While the proof of concept demonstrated the viability of peak number and distribution as suitable audio features for foam differentiation, broader validation across a wider range of materials, especially biological tissues, is essential. Most tissues are a heterogeneous composition of different cell types, and therefore, it remains to be investigated which structural or cellular features are most suitable for biological tissue characterization.

In this study, punctures were performed at two constant speeds (10 and 20 mm/s) using a 22G Quincke needle at a 90° angle with a robotic puncture setup. This standardization ensured reproducibility but does not reflect clinical variability in puncture angle, speed, and needle type. To address this limitation, we are developing a video-based method to analyze the speed of robotic and manual needle insertions using calibration markers retrospectively. However, real-time application requires further development. Although only a single needle type was used, the clip-on sensor is compatible with a wide range of needles, including larger diameters. While resonance characteristics may vary across needle types, core material features should remain constant. The analyzed materials were only 15 mm thick, but in clinical scenarios, relevant structures often lie at greater depths (e.g., the peritoneum). Although signal intensity may be attenuated with increasing puncture depth, it is hypothesized that characteristic features remain. This is supported by previous work that successfully recorded vibroacoustic data from a 150 mm Veress needle and biopsy needles [20]. Future work should examine these factors systematically and define detection limits under realistic conditions.

In practice, the ability to detect subtle material transitions can support clinicians in tasks such as optimizing the taking of good biopsy samples, particularly in situations where margins between healthy and malignant tissue are poorly defined. The core idea of this current approach is to detect those transitions between structurally distinct tissues by finding changes in peak patterns. Another potential application is the more accurate navigation and placement of the needle in challenging interventions. For example, in peritoneum puncture, it is not clearly indicated when the target site is reached. By augmenting and interpreting the vibroacoustic signal that is created when entering the peritoneal cavity, the safety and speed of the procedure can be enhanced [20]. Furthermore, real-time feedback from vibroacoustic signals could assist clinicians in adjusting needle trajectory, avoiding non-target areas, or confirming that critical zones have been reached. To work toward this integration, future research should build on this proof of concept and extend the investigations on vibroacoustic properties to more diverse materials, primarily focusing on biological tissues such as fat, muscle, or tumors.

The foams in this study were not intended to directly replicate specific biological tissues, but rather to represent structural variation in a simplified and controlled environment, and to demonstrate how vibroacoustic peak patterns depend on material structure. The air-filled pockets in the foams serve as examples of microstructural features that are hypothesized to be found in biological tissues as well. For instance, the alveolar cavities of the lung are frequently described as foam-like structures, and, in the context of fluid-filled compartments, the alternating architecture of hepatocytes and sinusoids in the liver provides a comparable structural pattern [26, 27].

Similar to tissues, which are comprised of a heterogenous mixture of different cell types, vibroacoustic signatures are likely composed of multiple audio features summing up to a more complex audio signature. Due to that, determining and assessing the utility of peak number and distribution as one part of this signatures is a critical next step for adopting this technology into clinical practice. Identifying reliable and specific audio features is essential for comprehensively describing distinct material patterns. Additionally, exploring the effects of environmental factors, procedural variables, and needle variants on vibroacoustic responses will enhance the robustness of the technology. Developing standardized protocols for consistent signal detection across clinical conditions will pave the way for integrating vibroacoustic sensing into medical tools to enhance surgical precision through improved differentiation of tissue types.

In conclusion, the findings in this proof of concept demonstrate that vibroacoustic signals hold information to differentiate between materials based on their structural compositions. This proof of concept demonstrated a direct correlation between the frequency of detected amplitude events in vibroacoustic signals and the material structure, laying a foundation for further exploration into real-time event detection and tissue differentiation during minimally invasive procedures. Further research has the potential to refine the method and expand its application to more complex and clinically relevant materials.