AI revolutionizes the use of medical needles

Imagine a doctor trying to reach a cancerous nodule deep inside the patient's lungs. Deep inside the patient's lungs, hidden behind a peas-sized target hidden behind a beating maze, and behind critical blood vessels and airways that shift with each breath. If you are off course, the main artery can puncture holes, and if it is deficient, the cancer will be completely missing and spread without treatment.

This is faced in high stakes that real-life doctors face in thousands of procedures every day. There, accuracy is important, and tasks are complicated by invasive or sensitive anatomical disorders. Will artificial intelligence (AI) and robots help address these challenges and improve patient outcomes?

The new era of “AI Guidance” is dawn in medicine,” says Ron Arobitz, a well-known professor at Lawrence Grossberg in the Department of Computer Science. “A robot with advanced AI can assist doctors, automate specific tasks, enable unprecedented levels of accuracy, and make complex procedures safer and more effective.”

A recent article in Science Robotics formalizes the concept of AI guidance on medical needle-based procedures and outlines the extent of AI guidance for each component. The author defines four components of AI guidance. Anatomy perception, planning instrument movement, perception of instrument state, and performing instrument movement during procedures. Robots with AI guidance can improve the accuracy and accuracy of needle guidance than human doctors, while also enabling the use of state-of-the-art needle designs that can bend to pilot the body.

This paper, “Medical Needles in the Hand of AI: Advances towards Autonomous Robotic Navigation,” was written by Alterovitz and by Janine Hoelscher of Clemson University and Alan Kuntz of the University of Utah. Hoelscher and Kuntz previously completed their PhD in Computer Science from UNC and completed Alterovitz as advisors.

The era of AI guidance

The doctors relied on imaging guidance using x-ray images, computed tomography (CT) scans, and magnetic resonance (MR) images to visualize the patient's anatomy and planned needle paths prior to the procedure. This advancement, dating back to the discovery of x-rays in the late 1800s, has allowed for safer access to points within the body.

Recent advances in AI allow us to move forward. AI can automatically analyze images, identify targets and obstacles, calculate safe trajectories, and even autonomously manipulate robotic needles around sensitive tissue deep inside the body. One such example demonstrated a medical robot shown by a team of researchers from the University of North Carolina Chapel Hill University, Vanderbilt University and the University of Utah. This demonstrates superior performance over physicists using only traditional tools in living tissues that are clinically relevant to needles autonomously. The authors detail this transformative shift from imaging guidance to AI guidance by helping AI perceive patient anatomy, track progress in procedures, plan equipment movements, and carry out those movements.

“For decades, imaging guidance has helped doctors better plan and implement medical procedures,” says Altovitz. “AI guidance is moving further towards making procedures safer and less invasive.”

This article formalizes the concept of AI guidance. This leverages AI to improve physician performance and creates a component of higher levels of robotic autonomy. In this article, we define the four components of AI guidance as follows:

1. Anatomy Perception

2. Planning equipment movements

3. Perception of musical instrumental state

4. Perform equipment movements during the procedure

Each of these four components can provide their own AI guidance.

• Eyes-on/Handson-Providing assistance where doctors perform tasks with AI

• Eyes-on/Hands-off- AI is performing tasks, doctors are ready to monitor AI and intervene in unusual circumstances

• Eye Off/Handoff – AI performs tasks and doctors only intervene when requested by AI

• Complete AI Guidance – Where AI performs tasks perfectly

Finally, this article will divide the current system into these categories and discuss research challenges to allow for a higher degree of AI guidance. In particular, researchers highlight important barriers to widespread clinical adoption, including the need to ensure safety, operation within a regulatory environment, the development of intuitive physician-AI interfaces through each degree of AI guidance, and the ability to integrate the necessary technologies into all aspects of clinical workflows.

While acknowledging the many challenges that still need to be overcome, Alterovitz expressed his excitement for the future of AI guidance in medical procedures.

“The AI ​​and robotics breakthroughs continue to allow for increased AI guidance and robotics automation for medical procedures,” says Alterovitz. “AI and robotics can provide doctors with new tools and make challenging procedures safer and more effective.”