The integration of AI has led to significant advances in medical robotics, with autonomous surgical robots increasingly being used for different surgeries. AI in medical robotics has also improved diagnostics and the outcomes of rehabilitation therapies. This article discusses the importance and application of AI in medical robotics.
This image was created with the assistance of DALL·E 3
Role of AI in Medical Robotics
In the healthcare industry, the role of AI has been specifically transformative in medical robotics. The fusion of AI in medical robotics has led to the development of autonomous surgical robots, which can assist surgeons in performing precise surgical procedures, leading to improved accuracy and patient outcomes and reduced invasiveness.
AI-powered medical robots can perform delicate surgical procedures, reach inaccessible areas of the human body, and make precise incisions, which minimize the risks involved with surgeries. AI can generate three-dimensional (3D) images as a reference for surgeons during surgeries by analyzing extensive patient data. Additionally, AI enables surgeons to make crucial decisions by providing analytics in real-time, leading to optimal results.
AI-powered telesurgery/remote surgery allows surgeons to perform their procedures on patients from any location without being present physically. Thus, AI-powered telesurgery can eliminate the existing geographical barriers to healthcare access by leveraging the capabilities of AI and virtual reality, which ensures the delivery of quality surgical care to remote regions.
In recent years, advancements in AI have led to the development of autonomous surgical robots, which can autonomously perform several simple surgical tasks to reduce the surgeon's workload and allow them to focus on complex tasks. Moreover, AI algorithms can analyze a patient's unique anatomy and disease characteristics to develop personalized surgical plans. Potential complications can be predicted, and the most effective treatment approach can be suggested by AI using machine learning (ML).
AI-powered robots can be used in rehabilitation therapies to assist patients in regaining mobility and performing exercises continuously. Diagnostics can also be improved by AI in medical robotics through the analysis of patient symptoms and data and medical images for enhanced accuracy and personalized treatment plans.
Benefits of Using AI in Medical Robotics
Enhanced Patient Outcomes: AI in medical robotics can improve patient outcomes through less invasive procedures, quicker recovery times, fewer complications, shorter hospital stays, and reduced overall cost of healthcare.
AI-powered robotic surgeries are primarily minimally invasive as these procedures lead to smaller incisions compared to conventional surgeries, resulting in a lower risk of post-surgery infection and less pain. Patients experience quicker recovery times due to the minimally invasive nature of AI-powered robotic surgeries, which implies less time spent in the hospital and a quicker return to their normal lives. Shorter hospital stays reduce the load on healthcare facilities and staff and allow them to provide service to more patients.
Additionally, AI-powered robotic surgery decreases the potential of surgical complications by reducing the risks due to human errors during surgeries. Fewer complications, shorter hospital stays, and quicker recovery times significantly reduce overall healthcare costs, which is beneficial for both patients and the healthcare system.
Reduced Surgeon Fatigue: Surgical procedures are often physically and mentally demanding as they require several hours of intense focus. AI-powered robotic systems can alleviate the burden by significantly reducing the physical and mental stress associated with complex surgical procedures.
AI allows surgeons to maintain their cognitive sharpness and focus through lengthy procedures by taking over several aspects of the operation. Thus, AI in medical robotics enables surgeons to perform tasks more efficiently, leading to better surgery outcomes.
Expanded Access to Surgical Care: AI in medical robotics can potentially democratize access to surgical care by enabling telesurgery on a large scale and increasing the availability of surgical expertise.
AI expands the availability of specialized surgical interventions by enabling surgeons to perform surgeries remotely, which is specifically beneficial for regions with an acute shortage of surgical specialists. Thus, AI in medical robotics promotes health equality by providing quality surgical care to everyone, irrespective of location.
Key Applications
Incorporating graphical processing units into ML tasks can improve the prediction of outcomes and diagnostics for several ophthalmological conditions, such as retinal conditions and glaucoma. ML can be utilized to predict the mortality outcomes after spinal tumor surgery. Transsphenoidal/pituitary gland surgery has been successfully performed using AI, telesurgery, and realistic motion tracking.
Six AI models have been employed to determine the preoperative risk of death in individuals with congenital heart defects undergoing cardiac surgery, considering their body mass index, arterial oxygen saturation index, and height. An AI model using coronary computed tomography angiography has been implemented successfully to profile the perivascular fat attenuation index and predict the cardiac risk in patients with acute myocardial infarction, patients who experienced major adverse cardiac events from earlier angiography, and patients who had undergone cardiac surgery.
In the orthopedic patient cohort, an AI-assisted system can be utilized for postoperative follow-ups. Studies demonstrated that the AI method performs similarly to the manual follow-up method based on patient feedback and cost-effectiveness.
Limitations of AI in Medical Robotics
ML and other AI analyses are highly dependent on data, and the accuracy and types of available data limit their outputs. Thus, the patterns recognized by the AI or its predictions are susceptible to systematic biases in clinical data collection. Additionally, AI cannot effectively determine causal relationships in data for clinical implementation despite the advances in causal inference and cannot provide an automated clinical interpretation of its analyses.
Moreover, the application of ineffective and inaccurate AI/ML systems can lead to iatrogenic harm affecting several surgical patients, while a single error of a surgeon can only affect a single patient, which is another major limitation. These inadvertent outcomes must be considered before deploying an AI/ML system in operation theaters. Specifically, extensive simulation and validation, systematic debugging, and audit must be performed when an AI/ML algorithm is deployed into surgical and clinical practice.
Recent Developments
The da Vinci System introduced by Intuitive Surgical is one of the leading examples of AI in medical robotics. This robotic-assisted surgical system provides enhanced 3D visualization to surgeons, greater precision, and improved dexterity. The system has been utilized in millions of minimally invasive surgeries worldwide, including general, gynecologic, urologic, and cardiac surgeries.
The Hugo™ Robotic-Assisted Surgery (RAS) System launched by Medtronic is another prominent example of AI-powered medical robotics that leverages AI to offer insights to surgeons in real-time and assists them in performing complex surgical procedures with enhanced control and precision.
The Butterfly Network has been developed as a pocket-sized, handheld AI-powered ultrasound device. Medical images of the entire body can be captured using the device, which can then be analyzed by AI to offer diagnostic insights. Integrating medical robotics and AI makes medical imaging more affordable and accessible.
The EndoPAR system, developed by the Technical University of Munich, Germany, is a ceiling-mounted experimental surgical platform that can autonomously execute knot-tying tasks using recurrent neural networks (RNNs) based on a database of 25 expert trajectories.
RNNs can approximate all dynamic systems and can be utilized for implementing sequence-to-sequence mappings that require memory, similar to the set of trajectories involved in suture knottying. Recently, an innovative endovascular surgery (ES) robot was used to evaluate a convolutional neural network (CNN)-based framework to navigate the ES robot based on the skill of a surgeon.
The CNN-based method demonstrated the capability of adapting to various situations while attaining a similar success rate in average operating time compared to known standards. Specifically, the robotic operation maintained a similar level of operating force and displayed a similar operating trajectory compared to manual operation.
The Smart Tissue Autonomous Robot (STAR) can be employed to perform robot-assisted supervised autonomous surgery in different soft tissue surgical tasks, such as ex vivo linear suturing of a longitudinal cut along the suspended intestine length, in vivo end-to-end anastomosis of porcine small intestine, and ex vivo end-to-end anastomosis.
References and Further Readings
Loh, E., Nguyen, T. (2021). Artificial intelligence for medical robotics. Endorobotics, 23-30. https://doi.org/10.1016/B978-0-12-821750-4.00002-5
Ozmen, M.M., Ozmen, A., Koç, Ç.K. (2021). Artificial Intelligence for Next-Generation Medical Robotics. Digital Surgery. https://doi.org/10.1007/978-3-030-49100-0_3
The Rise of Medical Robotics: How AI is Changing the Surgical Landscape [Online] (Accessed on 09 October 2023)