Preprint

Article

Perspective of Creating a Low-Cost Medical Assistant Robot Based on the WafflePI4 Platform with a Palm Vein Pattern Scanner

Altmetrics

Downloads

Views

Comments

Valentyn Anufriiev

Olha Kravchuk,

Yevhenii Levchenko

Hanna Hlukhova^*,Elena Linnyk,

Hanna Hlukhova^*,Elena Linnyk,

Oleg Glukhov^*

This version is not peer-reviewed

Submitted:

26 March 2024

Posted:

28 March 2024

You are already at the latest version

Alerts

Abstract

Military medical service is an important component of the Armed Forces of any country [1]. In wartime, healthcare performs critical functions of medical support for military operations, providing effective and qualified care to the wounded and sick. Consequently, it is imperative to equip military hospitals with state-of-the-art materials and technical resources commensurate with the predefined requirements. In this study, we examine a series of enhancements applied to the TurtleBot 3 Waffle Pi-based medical assistant robot, tailored for deployment in military hospital settings.

Keywords:

Subject: Engineering - Electrical and Electronic Engineering

1. Introduction

Military medical facilities and the personnel who serve them play a particularly important role under martial law. Military medicine is currently a fairly developed field in Ukraine that meets all the modern trends typical of developed countries in Europe and the world. Nevertheless, given the overcrowding in hospitals, it is very difficult for staff to provide timely and appropriate assistance to all victims. After analyzing the infectious diseases typical of the military risk group (such as pulmonary tuberculosis, viral hemorrhagic fevers, etc.) [2], it was decided to develop a robotic system, using the best practices of our group’s previous studies [3,4]. It is capable of autonomously performing the typical functions of nursing staff, allowing workers to provide emergency care to soldiers who have suffered serious injuries, such as shrapnel.

2. Materials and Methods

The TurtleBot 3 Waffle Pi was chosen as a platform for modification due to: 1) open-source software; 2) portability of the design; 3) compatibility with ROS (Robot Operating System); 4) modularity. In the article, the robot platform paired with OpenCR module (power supply, spatial orientation) and the Raspberry Pi 3 Model B microcomputer (remote control, control of sensors, disinfection module, and UI-UX interface).

In this research, we propose to replace the Raspberry Pi 3 with the Raspberry Pi 4, as a strategic move designed to yield multiply performance benefits, as substantiated by prior studies [5,6]. Implemented modifications and schematic diagram of the medical robot -assistant is shown on Figure 1. The anticipated advantages encompass heightened processing capabilities and augmented data transfer speed [7,8], expedited processing of graphical data [7,8], elevated data transfer rates facilitated by the inclusion of a 1Gb Ethernet chip and multi-channel Wi-Fi connectivity compliant with 802.11 b/g/n/ac standards [9,10], and the prospect of swifter data transmission and an expanded capacity for connected devices, courtesy of BLE 5.0 technology [11]. This proposition stands to enhance the operational efficiency and overall capabilities of the robotic system under proposed research.

The OpenCR module using SLAM (simultaneous localization and mapping) for spatial navigation [12,13,14] remains unchanged. SLAM technique is using probabilistic approach which applies a probability distribution to predict the robot and landmark’s location from the generated map. The probability distribution form, P is defined as:

P(x_k, m | Z_k, U_k),

where, k – time constant, x_k – robot location, Z_k – sequence of measurements between robot and landmarks assuming one measurement per time step, U_k – sequence of robot odometry or relative motion.

The proposed mechanism comprises a matrix of capsules, each housing a specific medication, and a controlled delivery system that employs a rotary mechanism. This rotary mechanism is actuated by a servo drive, facilitating the controlled ejection of the desired medication from the respective capsule. A critical aspect of this system is the implementation of a dispensing control system, which incorporates a matrix of optocouplers [15]. This control system acts as a fail-safe, guarding against erroneous medication dispensing. By interrogating optocouplers into the system, it becomes capable of accurately verifying and confirming the correct dosage and medication identity before dispensing to the patient. Incorporating this level of precision and error reduction in medication dispensing is paramount, especially in healthcare settings where patient safety and accurate dosing are of utmost importance. This study introduces a novel approach to enhance the reliability and precision of medication dispensing, ultimately contributing to improved patient outcomes and healthcare quality.

3. Results

A palm vein pattern scanner is used to provide contactless identification. This type of biometric identification was chosen due to its high level of identification accuracy, high resistance to changes over time and injuries to the palm veins, and the absence of the need for physical contact with the scanner.

The identification system consists of a computing module, a camera without an IR filter, and an IR LED unit. The computing module is a central robot controller (Raspberry Pi 4), and the camera is a Raspberry Pi Camera module 3 NoIR. The block diagram of the device is shown in Figure 2.

The implemented device consists of: Raspberry Pi 4B (1), Raspberry Pi Camera module 3 NoIR (2), IR LEDs (3), resistors (4) and a crown battery (5) and is shown in Figure 3 a). The assembled experimental setup is shown in Figure 3 b) and c). Notably, the Raspberry Pi 4b empowers the robot to gather and transmit data across longer distances at elevated speed, all while reducing overall energy consumptions. Algorithm of the device operation [16] is shown on Figure 4, and goes by following steps:

1. camera initialization and power supply to the LEDs;

2. taking a picture of the palm;

3. image processing;

4. patient identification.

To perform image processing, we implemented the code using the Python programming language, leveraging the capabilities of the OpenCV and NumPy libraries. The sequence of image processing steps is illustrated in Figure 5, delineating the various stages involved in the computational manipulation of the image data.

In this work, we employ a Convolutional Neural Network (CNN) as a robust methodology for the identification and recognition of user’s venous patterns. A Convolutional Neural Network is a specialized architecture within the domain of neural networks, extensively employed by the processing and analysis of images and videos [17,18,19]. Its design is tailored to excel in pattern recognition tasks, showcasing inherit properties that enhance its effectiveness in discerning intricate patterns within visual data. The CNN serves as a powerful tool for the precise identification of venous patterns in the context of our study. The structure of a CNN is shown in Figure 6.

The construction of the neural network was undertaken using the Python programming language as well as for the imaging processing, wherein we utilized the TensorFlow Keras library – an open-source machine learning software framework. This framework provides a comprehensive suite of tools and functionalities for the development, training, and evaluation of neural networks, ensuring a robust and efficient implementation of our model. Figure 7 illustrates the fundamental operational principles underlying this manuscript, which is centered around the unique palm vein pattern.

The depiction in the manuscript expounds on the intricate workings of the system, elucidating how it harnesses the distinctive features within the palm vein pattern for accurate and reliable biometric identification. The graphical representation serves as a visual guide, providing a detailed insight into the underlying mechanisms that drive the functionality of the developed biometric identification system.

5. Conclusions

The scope of this research work has revealed the tremendous potential for the adaptation and enhancement of a cost-effective medical robot-assistant, poised for both military and civilian applications. The core innovation underpinning this progress revolves around the replacement of the Raspberry Pi 3b microcomputer with its more advanced counterpart, the Raspberry Pi 4b. This transition has ushered in significant advancements, particularly in the domain of computational efficiency. One of the most notable improvements stemming from this microcomputer upgrade is the remarkable acceleration in the robot’s computational capabilities. By harnessing the enhanced processing power of the Raspberry Pi 4b, our robot can perform calculations with remarkable celerity, contributing to quicker decision-making and more responsive interactions. Moreover, we have achieved an innovative milestone through the development of a unique prototype featuring a biometric personal identification system predicted on the venous pattern of a palm. This pioneering approach to biometric identification holds great promise, offering heightened security and precision in identifying individuals, particularly within healthcare contexts. In conclusion, our study underscores the transformative potential of integrating the Raspberry Pi 4b microcomputer into robotic medical assistance systems, offering advancements in computation, data transmission, and energy efficiency. Simultaneously, the introduction of a palm-based biometric identification system holds promise in reshaping healthcare. These innovations collectively represent a step forward in leveraging technology to enhance the quality and efficiency of healthcare services. The outcome of this work, ripe for further exploration and refinement, has the potential to redefine the landscape of healthcare and military assistance.

Author Contributions

Conceptualization, V. Anufriiev and O. Glukhov; methodology, O. Kravchuk and H. Hlukhova; software, C. Anufriiev and Y. Levchenko; validation, V. Anufriiev, O. Glukhov and Y. Levchenko; formal analysis, V. Anufriiev, H. Hlukhova, O. Glukhov; investigation, V. Anufriiev, O. Glukhov, Y. Levchenko; resources, H. Hlukhova; data curation, V. Anufriiev and O. Kravchuk; writing—original draft preparation, O. Glukhov, O. Kravchuk, H. Hlukhova; writing—review and editing, H. Hlukhova.; visualization, E. Linnyk; supervision, O. Glukhov.; project administration, O. Glukhov and E. Linnyk. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: H. Hlukhova and O. Glukhov are family members, but they have lived and worked in different countries for more than 6 years, and they do not have any conflict of interest for the submitted work.

References

Allied joint doctrine for medical support, NATO STANDARD (United Kingdom). https://www.gov.uk/government/publications/allied-joint-medical-support-doctrine-ajp-410.
Liu, G. D. , Wang, N., Wang, H. M., Li, X., Shao, J. J., Liu, Z. F., Jiang, M., Wang, L., Wang, Z. K., Li, M., Cao, X. Y., Wang, J., Zhang, R., & Chen, Y. D. (2021). Military medical research on internal diseases in modern warfare: new concepts, demands, challenges, and opportunities. Military Medical Research, 8(1), 20. [CrossRef]
Chekubasheva, V. , Glukhov, O., Kravchuk, O., Levchenko, Y., Linnyk, E., & Rohovets, V. (2022). Possibility of Creating a Low-Cost Robot Assistant for Use in General Medical Institutions During the COVID-19 Pandemic. У Springer Proceedings in Physics (p. 203–213). Springer International Publishing. [CrossRef]
Chekubasheva, V. A. , Kravchuk, O. A., Hlukhova, H., & Glukhov, O. V. (2022). Creating of a remote-presence robot based on the development board Texas Instruments to monitor the status of infected patients. Biosensors and Bioelectronics: X, 100215. [CrossRef]
The Raspberry Pi Foundation, "Raspberry Pi 3 Computer Model B Product Brief", Feb. 2016.
The Raspberry Pi Foundation, "Raspberry Pi 4 Computer Model B Product Brief", Jan. 2021.
Broadcom, "ARM Peripherals", BCM2837B0 datasheet, Feb. 2012.
Broadcom, "ARM Peripherals", BCM2711 datasheet, Feb. 2020.
MICROCHIP, "Hi-Speed USB 2.0 to 10/100/1000 Ethernet Controller", LAN7515 datasheet, 2014.
Broadcom, "Single Port RGMII Gigabit Ethernet Transceivers", BCM54213 datasheet, Apr. 2021.
Cypress, "Single-Chip 5G WiFi IEEE 802.11n/ac MAC/Baseband/Radio", CYW43455, Mar. 2019.
Karlsson, N., Bernardo, E., Ostrowski, J., Goncalves, L., Pirjanian, P.,Munich,M.: The vSLAM Algorithm for Robust Localization and Mapping. In: Proceedings of the 2005 IEEE International Conference on Robotics and Automation, pp. 24–29, Barcelona, ICRA (2005).
Xu.Y., Cao, J., Shmaliy, Y.S., Zhuang, Yuan: Distributed Kalman filter for UWB/INS integrated pedestrian localization under colored measurement noise. Satell. Navig. 2, 22 (2021).
Choi, K.-S., & Lee, S.-G. (2010). Enhanced SLAM for a mobile robot using extended Kalman Filter and neural networks. International Journal of Precision Engineering and Manufacturing, 11(2), 255–264. https://doi.org/10.1007/s12541-010-0029-9. [CrossRef]
Chekubasheva, V. A. (2023). Application of robotic systems in military hospitals during the period of martial law in Ukraine. Radio electronics and youth in the XXI century (p. 5–6). Kharkiv National University of Radio Electronics. https://drive.google.com/file/d/1C0zoc_xZXkpcbs2n3EmhplmJWevsTGV1/view?usp=drive_link.
Wu, W., Elliott, S. J., Lin, S., Sun, S., & Tang, Y. (2019). Review of palm vein recognition. IET Biometrics, 9(1), 1–10. https://doi.org/10.1049/iet-bmt.2019.0034. [CrossRef]
Rasheed, T. (2016). Make Your Own Neural Network. CreateSpace Independent Publishing Platform.
Yang, S. , Zhu, F., Ling, X., Liu, Q., & Zhao, P. (2021). Intelligent Health Care: Applications of Deep Learning in Computational Medicine. Frontiers in Genetics, 12. [CrossRef]
Knok, Ž. , Pap, K., & Hrnčić, M. (2019). Implementation of intelligent model for pneumonia detection. Tehnički glasnik, 13(4), 315–322. [CrossRef]

Figure 1. Block diagram of the medical robot assistant with implemented modifications.

Figure 2. Block diagram of the palm vein pattern scanner.

Figure 3. a) Schematic of the experimental setup with connected modules; b) top view of implemented experimental setup; c) side view of implemented experimental setup.

Figure 4. Algorithm of the device operation.

Figure 5. Stages of image analyzing and processing.

Figure 6. The structure of Convolutional Neural network (CNN).

Figure 7. Working principle of the biometric identification system.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.