Abstract
Epidural anesthesia is a pain relief treatment mainly used for pregnant women during delivery. The procedure is described as an injection of anesthetic fluid into an epidural space using a hemodynamic Tuohy needle. The correct identification of epidural space is crucial for successful anesthesia, but due to the blindness of the procedure, the existing methods can fail with potentially severe consequences for patients. Since the tissue preceding the epidural space has high density, the strain significantly changes when the needle reaches the proximity of the epidural space, and this change can be used as an indicator. Distributed fiber optic sensors, working with an optical backscatter reflectometer, can improve the scenario by measuring the strain along each section of the epidural needle. This work goes beyond the epidural space identification: distributed strain data detect strain events leading to a potential failure to reach the epidural space before the procedure is finalized. The experiments have shown that the penetrations along different paths (with different speeds, with misalignment, and with rotation) result in different strain patterns. The proposed method has been validated on phantoms reproducing the epidural space, in different insertion conditions.