Artificial Intelligence, Machine Learning and Robotics in Healthcare — By Dr. Tunde Alaofin

In this age of innovation, robots and other intelligence devices start to move out of secured, controlled and predictable environment of our factories and industries to face the complexity and unpredictability of our daily environments. They are now co-occupants in our living rooms, offices, roads and our medical facilities. The only way to prevent catastrophes and avoid failure at the tasks they are programmed to do, especially in delicate and life or death environment like hospitals, is for robots to learn how to adapt on the go. Robots have to learn new techniques and strategies to enable them to react quickly and efficiently in their new environment. These are techniques for rapid and robust manipulation of objects.

New technologies hold a lot of promise in treating and diagnosing illnesses. When medical professionals adopt these new methods, they become more robust and rapid in their interventions. One of the technologies that can enable such patient outcomes is artificial intelligence (AI). AI refers to the study of machine algorithms to enable them to carry out cognitive functions like solving problems, making decisions, recognizing objects or reasoning. One cognitive function performed by artificial intelligence is pattern recognition and prediction. This is also called machine learning and is the basis behind making machines more adaptable. To narrow the discussion, focus in this analysis is on robot-assisted surgery. If healthcare specialists adopt artificial intelligence and machine learning, they can become more effective at dealing with difficult surgical environments and would thus save more lives.

Robots help enhance dexterity or deal with inefficiencies in the human hand during surgery. When a technologically- unaided surgeon operates on a patient, they may have problems with maintaining a steady hand. If a surgeon has tremors, they may make a mistake and cut or pierce the wrong part. Through surgical robots like “Surgeon Waldo,” a physician console and a remote patient cart can convert the doctor’s movement into steady movement (Smith, 2019). As a result, the surgeon would only operate on the exact area intended. This would eliminate unnecessary injury and blood loss. Alternatively, in case the surgeon has a weak hand that may not withstand a long procedure, these types of robots help in boosting endurance and strength. Artificial intelligence thus helps surgeons to reduce mistakes that stem from human inefficiencies. These mistakes can cause blood loss, pain, injury and sometimes even fatal outcomes.

Apart from eliminating motor errors, artificial intelligence and machine learning enable robots to detect deviations from treatment paths, thus correcting them. When surgeons want to operate, they usually have a plan. If the plan is fed into a machine, the robot can easily recognize when the surgeons have not followed it. Sometimes medics deliver non-optimal treatment because there is no machine learning (Smith, 2019). With time, this may undermine the effectiveness of the procedure. Treatment outcomes could be unsatisfactory if the intended path is not the implemented plan.

On top of treatment plan compliance, one of the most groundbreaking uses of robotics in surgery is radiation precision. If a person has tumors or cancerous lesions in sensitive parts of the body, invasive surgery may not be an option. Not only do they stand the risk of experiencing excessive damage, but they may also impair other body functions. For instance, invasive brain surgeries are risky because they can cause hemorrhages, which may lead to a stroke and, eventually, death. Other tumors in the kidney, lungs and other sensitive areas, may also lead to dilapidating outcomes if they are manipulated directly by hand. This is where machine learning and robotics come in (Malik & Rathaur, 2017). Through a technology called cyberknife (https://www.youtube.com/watch?reload=9&v=CxTSzVofgK8), people with lesions in sensitive areas can receive radiation therapy without any invasion at all. The robot already detects where the tumor is. It then targets it using the least uncomplicated path. Machine learning helps to calculate and find the best position to aim the laser beam through. This is typical of most supervised forms of machine learning. They enable the computer to predict results through an algorithm. Usually, the precision is within millimeters of the locations. As a result of the technology, some patients can have these robot-assisted surgeries done in the morning and on the same day, they can go about their daily business. Not only does that restore their health, but it does so through the least disruptive method possible.

Minimally invasive surgeries are now becoming a way to conduct many operations, and these also rely heavily on robotics and artificial intelligence. In a minimally invasive operation, the doctor cannot touch the actual body part that he wants to treat. This is often because the body part is very hard to reach, too small, or it may be too delicate. If an institution has a robotic assistant, it can still operate these areas through cameras and surgical instruments. The doctor would have access to remote hand controls in which they alter the movement of the instruments. The camera shows them exactly what or where they need to go. Artificial intelligence helps to keep the surgical tools in the correct working order (Virtua Health, 2020). The result is lower infection rates, which usually come when humans manipulate the inner parts of the body. Successful minimally invasive operations also lead to less scarring and greater chances of recovery.

While all the above benefits are real and meaningful, artificial intelligence in robot-assisted surgery has its limits. First, it cannot be regarded as a silver bullet for all delicate operations (Hashimoto et al., 2018). Even though it augments and supports decisions during operations, it is not a replacement for the human role in operations. Second, there are a lot of costs associated with such high-level applications of these technologies. Some of the robotic systems mentioned above are only available in certain institutions or parts of the world. This means that a vast majority of patients that require these interventions may not be able to access them. There are also risks associated with the use of artificial intelligence in surgeries. For instance, on the ones that conduct radiation, a machine may decide to pass radiation through a part of the brain that accounts for the person’s personality. If no human detects this machine decision, the robot may permanently change a patient’s personality merely because its algorithm told it that it was the best route. Things like human instinct and emotional intelligence cannot be replicated in a machine. Therefore, a doctor should always be on guard even when a robot does all of the procedures.

Conclusion

Artificial intelligence is critical in enabling rapid and robust manipulation of objects. This usually results in specific positive results during treatment. First, AI eliminates inefficiencies which stem from dexterity and endurance limitations. Additionally, AI keeps surgeons on the right treatment plan. Also, machine learning helps to calculate the best pathways to deliver radiation therapy. Finally, a robot helps surgeons to conduct minimally invasive surgery by guiding the surgical instruments into hard-to-reach places. The result of all these applications is more efficient medical procedures. Patients lose less blood and bear less pain. They don’t have to experience scarring or interruptions that come from tampering with their body parts. AI thus leads to better treatment outcomes.

References

Hashimoto, D., Rosman, G. & Meireless, O. (2018). Artificial Intelligence in Surgery: Promises and Perils. Annals of Surgery, 268(1), 70-76. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5995666/

Malik, A. & Rathaur, V. (2017). Overview of artificial intelligence in medicine. Journal of Family Medicine and Primary Care, 8(7), 2328-2331. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6691444/

Smith, R. (2019, Feb. 7). How Robots and AI are Creating the 21st-Century Surgeon. Retrieved from https://www.roboticsbusinessreview.com/health-medical/how-robots-and-ai-are-creating-the-21st-century-surgeon/

Virtua Health (2020). Robotic surgery. Retrieved from https://www.virtua.org/services/robotic-surgery

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s