Objective. This study aims to determine the methods of percutaneous release procedures in the lumbar ligamentum flavum (LF) under ultrasound guidance by acupotomy and provide an anatomical basis for intrusive treatment of lumbar disc herniation and lumbar spinal canal stenosis. Methods. Twelve cadavers including 4 females and 8 males aged 60 to 90 years (73.42 +/- 14.57 years), without formalin fixation, were selected. Guided by an ultrasound transducer, we punctured acupotomy to release lumbar LF in L3/L4, L4/L5, and L5/S1 segments. In the transverse-axis approach, the probe was placed transversely, while in the longitudinal-axis approach, the probe was placed longitudinally. The depth of needle penetration (A), the distance between the puncture point and spinous process (B), and the distance between the puncture point and sacral cornu (C) were measured on cadavers, and the depth of needle penetration (U-A), the distance between the puncture point and spinous process (U-B), and the angle for acupotomy (D) on ultrasound images were also measured. Statistical analyses were carried out using SPSS. Paired sample t-tests and homogeneity of variance tests and one-way analysis of variance (ANOVA) were performed. The Pearson correlation coefficients and linear correlation coefficients were calculated for the data obtained from ultrasound and cadaver measurements. Results. No obvious blood vessels and nerves were observed in the puncture path, and the spinal dura was intact. There was no statistical difference between the left and right side measurements obtained from the ultrasound images and the cadavers. The penetration depth in the transverse-axis approach was less than that in the longitudinal-axis approach, and the angle of the needle in the transverse-axis approach was greater than that in the longitudinal-axis approach. The measured data for the transverse-axis approach for L3/L4, L4/L5, and L5/S1 segments showed that there were no differences in the needle angle, the depth of needle penetration, and the distance from the spinous process to the puncture point among the three segments. There was a strong correlation between the depth of needle penetration and the distance from the spinous process to the puncture point on the ultrasonic images and the cadavers on the path of acupotomy. Linear equation A = 2.02 + 0.83 * U-A, R-2 = 0.352; B = 1.37 + 0.71 * U-B, R-2 = 0.252, where A/B refers to the data measured on the cadavers and U-A/U-B refers to the data measured on the ultrasound images. Conclusion. In this study, ultrasound guidance was applied, which better guaranteed the safety and feasibility of acupotomy therapy. Before performing the treatment, the depth of needle penetration in the human body can be determined by measuring the distance between the needle point and the target position on the ultrasound image. Under ultrasound guidance, the transverse-axis approach has a smaller puncture depth and greater puncture angle than the longitudinal-axis approach. Hence, this study believes that the transverse-axis approach is safer for the clinical application of ultrasound-guided LF acupotomy lysis.