Background/Purpose: The water jet technique dissects tissue while sparing cord-like structures such as blood vessels. The mechanism of such tissue-selective dissection has been unknown. The novel piezo actuator-driven pulsed water jet (ADPJ) system can achieve dissection with remarkably reduced water consumption compared to the conventional water jet; however, the system's characteristics and dissection capabilities on any organ have not been clarified. The purposes of this study were to characterize the physical properties of the novel ADPJ system, evaluate the dissection ability in swine organs, and reveal the mechanism of tissue-selective dissection. Methods: The pulsed water jet system comprised a pump chamber driven by a piezo actuator, a stainless steel tube, and a nozzle. The peak pressure of the pulsed water jet was measured through a sensing hole using a pressure sensor. The pulsed water jet technique was applied on swine liver in order to dissect tissue on a moving table using one-way linear ejection at a constant speed. The dissection depth was measured with light microscopy and evaluated histologically. The physical properties of swine liver were evaluated by breaking strength tests using tabletop universal testing instruments. The liver parenchyma was also cut with three currently available surgical devices to compare the histological findings. Results: The peak pressure of the pulsed water jet positively correlated with the input voltage (R2 = 0.9982, p < 0.0001), and this was reflected in the dissection depth. The dissection depth negatively correlated with the breaking strength of the liver parenchyma (R2 = 0.6694, p < 0.0001). The average breaking strengths of the liver parenchyma, hepatic veins, and Glisson's sheaths were 1.41 ± 0.45, 8.66 ± 1.70, and 29.6 ± 11.0 MPa, respectively. The breaking strength of the liver parenchyma was significantly lower than that of the hepatic veins and Glisson's sheaths. Histological staining confirmed that the liver parenchyma was selectively dissected, preserving the hepatic veins and Glisson's sheaths in contrast to what is commonly observed with electrocautery or ultrasonic instruments. Conclusions: The dissection depth of liver tissue is well controlled by input voltage and is influenced by the moving velocity and the physical properties of the organ. We showed that the device can be used to assure liver resection with tissue selectivity due to tissue-specific physical properties. Although this study uses an excised organ, further in vivo studies are necessary. The present work demonstrates that this device may function as an alternative tool for surgery due to its good controllability of the dissection depth and ability of tissue selectivity.