The Development of Artificial Intelligence for Robot-assisted Surgeries

Introduction

Artificial intelligence (AI) refers to the examination of algorithms that enable machines or systems to reason and carry out intellectual functions, such as pattern recognition and problem-solving functions (Hashimoto et al., 2018). Over the past decade, there have been various developments in the medical field, one of which is the application of artificial intelligence for medical procedures. Furthermore, AI has been a significant technological innovation that supports medical clinical decision-making processes. Though its application is still in its early stages, it has been successfully applied imaging and diagnosis, precision medicine, and genomics in both pre-surgery and intra-surgery procedures (Tsui, 2020).

Supervised algorithms such as atlas-based learning approach have been developed to enhance surgical procedures such as minimally invasive operations. Also, in a study on Deep Convolutional Neural Networks, Krizhevsky et al. (2012) suggested the application of ImageNet, which is a major neural network, for surgical procedures. The researchers emphasise that such deep leaning algorithms can support surgical procedures and improve visibility during surgical operations. Additionally, deep leaning methods like the Deep Convolutional Neural Network (DCNN), whereby several convolutional layers are programmed, have facilitated the use of robotically learned information-driven descriptors to be used for pre-operative procedures such as imaging examination (Zhou et al., 2019). This paper highlights the benefits of implementing AI technology into medical procedures, particularly robot-assisted surgeries, and highlights certain features  of such implementation,

2. An Overview of Robotics in Medicine

During robotic-assisted surgeries, the clinical devices are not handled by the surgeon; rather, robotic machines  direct these devices. The robotic machines have various motors and sensors implemented into them, which enhance surgeons’ capabilities when rotating and moving around during minimally invasive procedures. By offering simulated tactile movements, robots improve doctor’s deftness and eye-hand management, which improves surgical outcomes. One of the earliest surgical robots is the da Vinci SP robot, which serves as a console during telesurgeries (Tonutti et al., 2016). This system has been widely used for minimally invasive surgeries and various other surgeries, including abdominal and heart surgeries. It has also been approved by the Food and Drug Administration (FDA) (Tonutti et al., 2016).

The da Vinci system has a master console that has various controllers. It also comprises of the slave console with four robotic hands and 3D camera. Doctors manage the slave console and the robotic hands during surgical procedures. The hands could have other tailored devices implemented on them, such as articulated actuators that can be used to move endoscopes or other similar devices during operations. This improves the outcomes of robotic minimally invasive surgeries like prostatectomy (Marcus et al., 2014). A similar robotic system is known as the i-Snake surgical robot, which enables hand motions with eight degrees of freedom because it has four articulating joints. Various cameras, devices, and imaging fibres can be transmitted through its tubular body, while the system remains flexible and can be controlled by a surgeon (Tonutti et al., 2016). The applications of these kinds of robots during surgical procedures are further discussed in the next section.

3. Applications of AI for Robot-Assisted Surgical Procedures 

3.1     Telesurgery

Telesurgery utilises robotic and wireless networking technologies and enables doctors to conduct remote surgeries on their patients. The first telesurgery was conducted in 2001 with the use of a robotic device known as ZEUS, which was successfully used to conduct a laparoscopic cholecystectomy in France  (Xu et al., 2015). The use of robots for telesurgery has aided in resolving issues related to scarcity of surgeons, physical remoteness of urgent and valuable surgical care, substantial financial challenges, and required lengthy-distance travel (Choi et al., 2018). Also, according to Choi et al. (2018), this technology is beneficial to both patients and surgeons, as it improves technical precision during surgical procedures.

Though still in its early stages of development, tele-neurosurgical robot-assisted has been applied in surgical procedures. In a study on robotic skull base surgery, O’Malley and Weinstein (2007) performed a skull base surgical procedure with the use of a trans-oral method. Similarly, Wirz et al. (2015) discussed the trans-sphenoidal surgery on a hypophysis tumour, and the physicians stated that the use of the robotic system was identical to a standard endoscope. However, a standard telesurgery system had a significant downside because it does not offer tactile information during surgeries, and the surgeon must depend on visual feedback (Stark et al., 2012). The technology that facilitates the transfer of tactile data to a surgeon is referred to as haptic feedback. This is further discussed in the next sub-section.

3.2 Haptics Feedback in Robotic Systems

A significant factor considered during the application of surgical robots is haptic feedback. Haptics refer to feelings like pressure, force, and heat, which could often be challenging to measure and signify in robotic systems during minimally invasive procedures or other surgeries (Tonutti et al., 2016). Haptic feedback would aid a surgeon in feeling the uniformity of tissues and the pressure in stitches, thereby preventing injury to the delicate tissues, or cutting of patients’ stitches during a surgery (Choi et al., 2018). Therefore, the lack of haptic feedback could reduce a surgeon’s ability to determine the right level of pressure required  during surgical procedures. More recent surgical robots have included tactile and haptic feedback, which has enhanced the outcome of surgical robot procedures.

One of the first telesurgery systems that had integrated haptic feedback was the Telelap Alf-x system, which was developed in 2015. Being able to offer haptic feedback during surgical procedures aided in reducing the usual time spent on experimental cholecystectomy by an hour (Stark et al., 2015).  The Telelap Alf-x system also had an eye-trailing technology that would stop the robotic hands when a surgeon’s eyes were not focused on its console or screen. This further enhanced the outcome of the robot-assisted surgeries. Subsequently, Su et al. (2017) developed an innovative MRI-focused telesurgery tool that surgeons could utilize for percutaneous surgeries and offered haptic feedback to surgeons through differing ranges of pneumatic pressure. On the other hand, handheld robotic systems have haptic feedback integrated into them via numerous tactics, such as auditory signals and vibrations. These tactics have been used for surgical procedures like endoscopy with enhanced outcomes (Choi et al., 2018).

3.3     The Development of the Visual Feedback System

Shenai et al. (2014) proposed the Virtual Interactive Presence (VIP), an innovative technology that enables neurosurgeons to remotely work together with a 3D console through high-definition binoculars . The visual system enables doctors to examine a combined surgical presentation of each other’s hand movements. The VIP has been successfully used for various surgical procedures, including suboccipital craniotomies, with enhanced outcomes. Shenai et al. (2015) suggest that the system is suitable for both surgical procedures and surgical instruction, as it facilitates real-time collaboration between surgeons located in different hospitals worldwide.

4. Conclusion

Artificial intelligence and the use of surgical robots highlight that these technological innovations could transform the future of surgical procedures. Indeed, the use of such robotic systems could aid in enhancing navigation, imaging, and various pre-operative and post-operative procedures. Therefore, it can be projected that in the future, advanced surgical robots  systems would be able to observe and comprehend complex environments, perform real-time decision making, and execute required responsibilities with enhanced accuracy, care, and effectiveness.

References

Choi, P. J., Oskouian, R. J. and Tubbs, R. S. (2018). Telesurgery: Past, Present, and Future. Cureus, 10 (5), pp. 1-4.

Feizi, N., Tavakoli, M., Patel, R. V. and Atashzar, S. F. (2021). Robotics and AI for Teleoperation, Tele-Assessment, and Tele-Training for Surgery in the Era of COVID-19: Existing Challenges, and Future Vision. Frontiers in Robotics and AI, 8, pp. 1-16.

Hashimoto, D. A., Rosman, G., Rus, D. and Meireles, O. R. (2018). Artificial Intelligence in Surgery: Promises and Perils. Annals of surgery, 268 (1), pp. 70–76.

Krizhevsky, A., Sutskever, I. and Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems 25 (NIPS 2012), pp. 1-9.

Leslie, S., De Castro Abreu, A.L., Chopra, S., et al. (2014). Transvesical robotic simple prostatectomy: initial clinical experience. Eur Urol, 66, pp. 321-329.

Marcus, H.J., Seneci, C.A., Payne, C.J. et al. (2014). Robotics in keyhole transcranial endoscope-assisted microsurgery. Oper Neurosurg, 10, pp. 84–96.

O’Malley, B.W. and Weinstein, G.S. (2007). Robotic skull base surgery: Preclinical investigations to human clinical application. Arch Otolaryngol Head Neck Surg., 133, pp. 1215–1219.

Shenai, M.B., Tubbs, R.S., Guthrie, B.L. and Cohen-Gadol, A.A. (2014). Virtual interactive presence for real-time, long-distance surgical collaboration during complex microsurgical procedures. J Neurosurg., 121, pp. 277–284.

Stark, M., Morales, E.R. and Gidaro, S. (2012) Telesurgery is promising but still need proof through prospective comparative studies. Journal of Gynecol Oncol., 23, pp. 134–135.

Stark, M., Pomati, S., D’Ambrosio, A., Giraudi, F. and Gidaro, S. (2015). A new telesurgical platform–preliminary clinical results. Minim Invasive Ther Allied Technology, 24, pp. 31–36.

Su, H., Shang, W., Li, G., Patel, N. and Fischer, G.S. (2017). An MRI-guided telesurgery system using a Fabry-Perot interferometry force sensor and a pneumatic haptic device. Ann Biomed Eng., 45, pp. 1917–1928.

Tonutti, M., Elson, D. S., Yang, G. Z., Darzi, A. W. and Sodergren, M. H. (2016). The role of technology in minimally invasive surgery: state of the art, recent developments, and future directions. London: The Hamlyn Centre for Robotic Surgery, Imperial College London.

Tsui, E. Y. (2020). Application of Artificial Intelligence (AI) in Surgery. [Online]. Available at: https://www.imperial.ac.uk/news/200673/application-artificial-intelligence-ai-surgery/ [Accessed 5 October 2021].

Wirz, R., Torres, L.G., Swaney, P.J., et al. (2015). An experimental feasibility study on robotic endonasal telesurgery. Neurosurgery, 76, pp. 479–484.

Xu, S., Perez, M., Yang, K., Perrenot, C., Felblinger, J. and Hubert, J. (2015). Effect of latency training on surgical performance in simulated robotic telesurgery procedures. Int J Med Robot Robotics Comput Assist Surg., 11, pp. 290–295.

Zhou, X. Y., Guo, Y., Shen, M. and Yang, G. Z. (2019). Artificial Intelligence in Surgery. Frontiers of Medicine, pp. 1-50.