Enhanced puncture event detection for teleoperated needle insertion robotic system

Lumbar puncture is a commonly used minimally invasive procedure in clinical practice that serves not only for disease diagnosis but also for therapeutic interventions and anesthesia during lower limb surgeries. However, the diagnostic, therapeutic, or anesthetic outcomes of current lumbar puncture procedures heavily rely on the physician's personal experience, resulting in prolonged procedure times and repeated exposure to harmful radiation for practitioners. In this paper, a novel teleoperated needle insertion robotic system for lumbar puncture surgery was proposed. This compact robotic system allows for augmented needle-tissue interaction force feedback and mimics the manual gestures performed by a physician, including needle feed angle adjustment and delicate insertion depth control. An enhanced puncture event detection scheme based on an error backpropagation neural network was devised to automatically detect transitions during needle insertion. To validate the effectiveness of the proposed system and detection scheme, experiments were conducted using laboratory-made artificial human tissues and a professional epidural learning pediatric simulation manikin.