Invasive placenta includes placenta accreta, increta, and percreta, which are classified according to the depth of uterine invasion by the chorionic villi. Invasive placenta is a major cause of maternal morbidity and mortality, primarily by postpartum hemorrhage. Placenta percreta is the most severe type which penetrates through the myometrium and invades the serosal layer of the uterus or even into adjacent organs. Placenta percreta might result in the need for an extremely difficult surgical technique requiring specific procedures such as aortic clamping or hypogastric ligation. Consequently, determining whether an invasive placenta is a placenta percreta is vital to decrease maternal morbidity and mortality. Although ultrasound is preferred in the diagnosis of invasive placenta, when ultrasound findings are suspicious or inconclusive, magnetic resonance imaging is recommended as a supplemental imaging technique. The objective of this study was to identify valuable MRI features for distinguishing placenta percreta from placenta accreta and increta.46 patients diagnosed with invasive placenta (19 cases of placenta percreta; 27 cases of placenta accreta and increta) were enrolled in this retrospective study. All MRI examinations were performed using a MAGNETOM Sonata 1.5T MR scanner (Siemens Healthcare, Erlangen, Germany) including sagittal, axial and coronal half-Fourier acquisition with single-shot turbo spin echo (HASTE) and true fast-imaging with steady-state precession (True FISP) sequence. HASTE: TR 1000.00 ms, TE 94.00 ms, flip angle 90, 256×256 matrix, 5-mm thickness with no gap, FOV 400 mm×300 mm, NEX 1, spatial resolution 1.6 mm ×1.4 mm ×5.0 mm; True FISP: TR 4.30 ms, TE 2.10 ms, flip angle 76, matrix 256×256, 5-mm thickness with no gap, FOV 400 mm×400 mm, NEX 1, spatial resolution 1.6 mm×1.8 mm×5.0 mm. All patient diagnoses were confirmed using surgical or pathological criteria. Two experienced radiologists evaluated a total of seven MRI features of 46 patients using surgical and histopathological correlations. These features were bulging placenta with cyrtoid outline, nodular protrusion in the urinary bladder wall, intraplacental dark bands, intraplacental abnormal vascularity, marked placental heterogeneity, uterine bulging and signs of placental protrusion. Inter-rater reliability was assessed using Kappa statistics. Features were evaluated using Fisher’s two-sided exact test for comparison of their capabilities for distinguishing placenta percreta from placenta accreta and increta. Sensitivity and specificity of all seven MRI features for diagnosis of placenta percreta were estimated.Inter-rater reliability was moderate or better for all the features (Table 1). Fisher’s two-sided exact test results showed that features including marked placental heterogeneity(figure 1), placental protrusion (figure 2) and bulging placenta with cyrtoid outline (figure 3-5) were significantly associated with placenta percreta (reader 1: 0.001, 0.002,＜0.001; reader 2: 0.006, ＜0.001, ＜0.001). In 6 cases of placenta percreta invading urinary bladder, 4 cases showed the nodular protrusion in the urinary bladder wall (figure 6). Sensitivity and specificity of all MRI features for diagnosis of placenta percreta were shown in Table 2.Placenta and fetus imaging is a big challenge for MR, as there is severe motion affecting the image quality, so in this study we used two fast imaging technique: HASTE and Truefisp for placenta imaging.The MRI features of bulging placenta with cyrtoid outline, marked placental heterogeneity and signs of placental protrusion are useful for distinguishing placenta percreta from placenta accreta and increta. Nodular protrusion in the urinary bladder wall is indicative of urinary bladder involvement.