Neural Networks and Artificial’s Intelligence’s Algorithms in the Tasks of Diagnostics Foreign Bodies in the Wound Channel of Patients

Oleksandr Salenko; Viktor Cherniak; Vadym Orel; Bohdan Salenko

doi:10.20535/2521-1943.2026.10.1.344614

Authors

Oleksandr Salenko National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” , Ukraine https://orcid.org/0000-0002-5685-6225
Viktor Cherniak Taras Shevchenko National University of Kyiv , Ukraine https://orcid.org/0009-0005-5280-542X
Vadym Orel Kharkiv National University of Internal Affairs , Ukraine https://orcid.org/0000-0002-8775-3253
Bohdan Salenko Yuriy Kondratyuk Poltava Technical University , Ukraine https://orcid.org/0009-0004-4026-6767

DOI:

https://doi.org/10.20535/2521-1943.2026.10.1.344614

Keywords:

wound diagnosis tools, foreign bodies, noise emission, neural networks and artificial intelligence, determination

Abstract

The paper considers a device in which information about the presence of a foreign body a fragment in a wound is obtained using a special compact sensor: a mechanical elastic flexible rod, which is placed in a protective soft tube and connected to a sealed chamber, one of the walls of which is a membrane capable of converting the vibrations of the rod into sound waves, which, propagating in the chamber, are recorded by a microphone. Mounted in a convenient housing, the sensor is connected to sound frequency filters (up to 15 kHz) via an amplifying link and allows you to obtain a noise signal, based on which you can detect the contact of the rod with a foreign body during its movement in the patient's wound. The difference in the spectrum patterns suggests that such a device can be used not only to detect foreign bodies, but also to identify them.
The use of neural networks and artificial intelligence algorithms significantly improves the accuracy of foreign body detection, especially for small objects and those whose rheological properties are similar to the patient's tissue. Thus, with the appropriate adjustment of the device, the accuracy of foreign body diagnosis is currently up to 65–95 % for bodies larger than 3.5–5.0 mm; higher accuracy is observed when detecting bodies with pronounced elastic properties.
It has been shown that the use of this device allows obtaining information about the presence of a foreign body (due to the appearance of separate clearly defined signal peaks), while the absence of foreign bodies gives a picture of noise emission distribution close to the Gaussian distribution; about the type of foreign body. The shape of the pattern allows one to assess the type of foreign inclusion (more elastic bodies have more pronounced frequency spikes); to estimate the approximate size of the body: larger bodies are in contact with the probe for a longer time as it moves.

References

V. Chorna et al., “The characteristics of injuries from various types of weapons depend on the individual’s location at the moment of the explosion”, Ukraine, Nation’s Health, No. 2, pp. 113–121, 2024, doi: https://doi.org/10.32782/2077-6594/2024.2/19.
K. Faguy, “Imaging foreign bodies”, Radiologic Technology, Vol. 85(6), pp. 655–682, 2014.
Philip J. Withers et al., “X-ray computed tomography,” Nature Reviews Methods Primers, 18, Feb. 2021, doi: https://doi.org/10.1038/S43586-021-00015-4.
K. S. Tupikov, S. V. Shkel, V. M. Orel, O. F. Salenko, and V. A. Cherniak, “Vysokotekhnolohichni innovatsii v medychnii praktytsi, shcho riatuiut zhyttia ta povertaiut zdorovia poranenym pid chas viiskovykh dii,” in Proc. 6th Int. Sci. and Practical Conf. Science and Innovation of Modern World, London, United Kingdom, Feb. 23–25, 2023, pp. 163–171.
V. A. Stoyan, Modelyuvannya ta іdentifіkatsіya dinamіki sistem іz rozpodіlenimi parametrami, Kyiv: Kyiv unіversitet, 2003. 187 p.
M. Maqbool, “Computed Tomography,” An Introduction to Medical Physics, pp. 221–262, Nov. 2017, doi: https://doi.org/10.1007/978-3-319-61540-0_8 .
V. A. Tymoniuk and E. N. Zhyvotnova, Biophysics, Kyiv: Professional, 2007.
O. Salenko et al., “The possibility of detecting non-x-ray fragments in the body of the wounded by the contact method”, Mech. Adv. Technol., Vol. 7, No. 3 (99), pp. 337–349, Dec. 2023, doi: https://doi.org/10.20535/2521-1943.2023.7.3.294822.
K. Zhiannu, M. Baldan and A. Molde, Viyskovo-Polyova surgery. The work of surgeons for the minds of sharing resources during times of violent conflicts and other situations of violence, 2013, 389 p.
A. Kapoor, A. Gulli, S. Pal, and F. Chollet, Deep Learning with TensorFlow and Keras: Build and deploy supervised, unsuper-vised, deep, and reinforcement learning models, 3rd ed. Birmingham, UK: Packt Publishing Ltd., 2022.
H. Bakır and R. Bakır, “Evaluating the robustness of yolo object detection algorithm in terms of detecting objects in noisy environment”, Journal of Scientific Reports-A, (054), pp. 1–25, 2023, doi: https://doi.org/10.59313/jsr-a.1257361.
Indrajit Mandal, P. Santosh Kumar Patra and Dipti Prava Sahu, “An Expert System for Diagnosis Of Human Diseases,” Inter-national Journal of Computer Applications, Vol. 1, No. 13, 2010, doi: https://doi.org/10.5120/279-439
A. H. Ashraf et al., “Weapons detection for security and video surveillance using CNN and YOLO-v5s”, Comput. Mater. Con-tin., Vol. 70, No. 4, pp. 2761–2775, 2022, doi: https://doi.org/10.32604/cmc.2022.018785.
S. Pavlov et al., “Healing features of experimental injuries of soft tissues that contain foreign bodies in the form of fragments of military personnel uniforms,” BMJ Mil. Health, Vol. 169(e1), pp. e59–e63, 2023, doi: https://doi.org/10.1136/bmjmilitary-2020-001666.
V Cherniak, O Salenko, K Karpenko and V Pryiemska, “Surgical treatment of a gunshot wound with a foreign body: using the sensitivity of the tactile sensor,” in SWorld-Ger Conference proceedings, 2025, pp.78-89.