Endometriosis is defined as the presence of functional endometrial glands and stroma outside the uterine cavity, i.e., ectopic endometrium as opposed to eutopic endometrium. Adenomyosis specifically refers to ectopic endometrial tissue located in the myometrium. Endometriosis, on the other hand, is generally used to refer to endometriosis other than adenomyosis. The pathogenesis, epidemiology, and clinical symptoms of the two are very different, and this article will describe the latter (hereafter referred to as endometriosis), focusing mainly on magnetic resonance imaging diagnosis. The main clinical symptoms of endometriosis include infertility and pain, mostly in women of childbearing age, with an average age at diagnosis of 25-29 years, 80% combined with chronic pelvic pain and 59% combined with infertility. The overall incidence is approximately 5-10%, with only 5% occurring in postmenopausal women. Some atypical symptoms are associated with the site of the lesion [1]. Pathologically, it can present as a lesion that can only be observed microscopically or as a largely visible ectopic endometrial lesion. The gross pathological presentation of endometriosis depends on the course of the disease and the depth of lesion penetration. Implanted ectopic endometrium can be either a punctate lesion or a small stellate patch, usually less than 2 cm in diameter. The amount of pigmentation increases with the age of the lesion. Initially, the lesions are white, yellow or red in color and then progress to a more mature blue or brown color [2]. These brownish petechial changes are described as “gunpowder spots”. The implanted endometrium can become swollen, congested, and even bleed with the menstrual cycle. Mature ectopic endometrium initiates an inflammatory response that leads to hematoma mechanization, fibrosis, and adhesion formation. Extensive adhesions can distort the normal anatomy of the pelvis and even close the rectal fossa of the uterus. The most commonly involved site is the ovary, but essentially all organs of the pelvis can be involved. Microscopically, endometriosis consists of endometrial glands, stroma, and occasionally contains smooth muscle fibers. As with the in situ endometrium, ectopic endometrial lesions respond to circulating hormones, producing secretory phase changes in the second half of the menstrual cycle and a metaphase response during pregnancy. Bleeding from these lesions results in an inflammatory response with tissue cell infiltration followed by iron-containing heme deposition. II. Clinical diagnosis and staging of endometriosis: The findings of endometriosis have certain characteristics. Tenderness of the uterosacral ligament and the rectal fossa of the uterus is often present. Sometimes thickened ligaments or nodular changes and masses in the rectovaginal compartment can be palpated. If the ovaries are involved, tenderness or masses in the adnexal region may be present. In pelvic adhesions, the pelvic organs are fixed and the uterus is often fixed in a posterior tilted position. However, many patients fail to detect abnormalities on physical examination [1]. Laparoscopy is the standard method for diagnosing endometriosis and allows for staging. Typical findings include implanted endometrium, ectopic endometrial cysts, and adhesions. The correct diagnosis is demanding on the operator and severe adhesions may even separate the pelvis, making laparoscopy impossible. There are various stages of endo, but the 1985 American Fertility Society’s modified staging method (r-AFS) is the most popular. It is based on laparoscopic observation and is scored according to the size of the ovarian and peritoneal lesions, the degree of adhesions, and the closure of the recto-uterine sulcus. However, this staging fails to take into account the pleomorphic and functional status of the lesion, i.e., active (red lesions) versus inactive (white lesions), especially as it does not express the two important clinical facts of pain and infertility [3]. And a considerable number of units do not perform laparoscopy, so some corrections are needed. III. Imaging evaluation of endometriosis Currently ultrasonography is the most commonly used imaging modality in patients with endometriosis [4]. However, it is only valuable for endometriosis cysts and does not allow the examination of endometrial implants and adhesions. Magnetic Resonance Imaging (MRI) has been shown to be more specific than other noninvasive imaging modalities [5]. It provides a larger field of view than ultrasonography and shows the adhesions of the lesion to the surrounding anatomical structures more clearly, making it a valuable adjunctive test for evaluating the occupancy of the adnexal region and can be chosen as a means of problem solving. 1. Technique: MRI examination should use pelvic-specific coils. An array of coils on the surface can provide a high signal-to-noise ratio, thereby improving spatial resolution and the display of anatomical details. The imaging plane can include three standard planes (axial, sagittal, and coronal), with the sagittal plane being particularly useful for evaluating the utero-rectal fossa and rectum [6]. Pelvic MRI usually scans fat-suppressed T1-weighted images in addition to the conventional T1- and T2-weighted images. Fat suppression narrows the signal range and thus highlights tissue signal contrast. On T1-weighted images, ectopic endometrial cysts may appear as relatively uniform high signal (the same as or higher than fat). When the high signal from the surrounding fat is suppressed, the lesion can be shown more clearly. Fat-suppressed T1-weighted images are valuable in the evaluation of endoheteropathy, improving the sensitivity of MRI in the qualitative diagnosis of small lesions and thereby excluding fat-containing lesions such as dermatomal cysts [7]. Contrast enhancement is not very helpful in the evaluation of endoheteropathy. On enhancement scans, the cystic wall of endometriosis enhances in various forms and cannot be distinguished from other benign or malignant lesions [8]. Moreover, normally enhancing parametrial tissue may be mistaken for an endometriosis lesion, leading to a false-positive diagnosis. In contrast, when ovarian cancer is suspected, enhancement scans are still meaningful. 2. MRI presentation of basic lesions: The three basic lesions of endometriosis include abdominal ectopic endometrial implants, ectopic endometrial cysts, and adhesions [9]. (1) Implanted endometrial lesions: are endometrial epidermis and stroma implanted in the peritoneal plasma membrane, initiating an inflammatory response and recurrent bleeding. Sites of abdominal involvement include the ovaries, uterine ligament, rectal fossa, peritoneal convolutions above the uterus, fallopian tubes, rectum and bladder. The implanted endometrium behaves very differently depending on the duration of bleeding and the degree of degradation of blood products. The lesions are often small and have large signal variability. Their signal is often close to that of normal endometrium, i.e., long T1 and long T2 signal, but they may also show low or high signal on T1 and T2 images. Smaller endometrial lesions are generally more difficult to visualize on MRI, which is a limitation of MRI application in this area. Deep pelvic lesions are described later. (2) Ectopic endometrial cysts of the ovary (i.e., chocolate cysts) contain brownish jelly-like material surrounded by fibrous walls of varying thickness. They are often multiple and bilateral. A common and important MRI feature is the presence of “shadowing” (i.e., loss of signal within the lesion) on T2-weighted images [6, 9]. This “shadow” reflects the chronic nature of ectopic endocysts and can be distinguished from other blood-containing lesions. The blood component within these cysts is the result of cyclic bleeding over the course of adult life. The contents of these chronic lesions are very viscous and contain blood degradation products, including very high concentrations of iron and proteins. At high concentrations, interconnection between proteins occurs, resulting in a decrease in T2 relaxation time. All of these factors produce a “shadow”. The shadows appear as vague stratification changes on the T2 image or as complete signal loss, which varies with the concentration of the blood component and can be highly variable. In contrast, it is often a uniform high signal on T1-weighted images. When there is acute hemorrhage it can show low signal on both T1 and T2 images, while old hemorrhage can show high signal on both T1 and T2 images. The low-signal ring around an ectopic endocyst is the result of the combination of the fibrotic cyst wall and macrophages containing iron-containing heme [5], which shows low signal on both T1 and T2-weighted images. Despite the diverse morphology and signal of ectopic endothelial cysts, Togashi et al [4] found that the diagnosis of ectopic endothelial cysts can be basically concluded when the cysts are high signal on T1 images and shadows are visible on T2 images. And when multiple high-signal cysts were present on T1-weighted images, the diagnosis of ectopic endothelial cysts could be basically determined regardless of the signal on T2 images. In this study, the overall sensitivity, specificity, and accuracy of MR imaging for the diagnosis of endothelia reached 90%, 98%, and 96%, respectively. These cysts contain blood components that vary in timing and concentration and therefore in presentation, and lesions that do not show high signal on T1 images are often difficult to differentiate from other adnexal masses. Other lesions that show high signal on T1 images include dermatomal cysts, mucinous cystadenomas, and hemorrhagic masses. Dermatomal cysts can be distinguished from ectopic endothelial cysts by the presence of chemical shift artifacts and suppression of signal on fat-suppressed images. Mucus-containing lesions can be high signal on T1 images, but their signal intensity is much less than that of fat or blood. The most difficult to differentiate is ovarian corpus luteum hemorrhage, which has an MR presentation similar to that of endoheterotrophy. Differentiating points can include the fact that hemorrhagic cysts are often single-atrial, whereas ectopic endometrial cysts are more multiatrial and occur bilaterally. Moreover, hemorrhagic cysts are not shadowed on T2 images and fade over time [10]. MR or ultrasound follow-up review can confirm the diagnosis. Ovarian cancer may occasionally have internal bleeding. Masses with solid components, internal compartments, and larger size often suggest the possibility of malignancy. 3) Adhesions are the most common and major complication of internal heterogeneity, and this is the issue that imaging examinations focus on. On MR examination, adhesions can sometimes show up as pin-like low signal strips that blur the organ boundaries. Posterior tilt of the uterus and ovaries, angulation of the bowel, elevation of the posterior vaginal vault, compartmentalization of pelvic fluid and hydrocele in the fallopian tubes, and loss of the interface between the ovaries and surrounding anatomical structures are signs that suggest the presence of adhesions [1, 5]. However, imaging is often difficult to understand the extent and severity of the adhesions, and laparoscopy is required for a definitive diagnosis. The value of MRI in the diagnosis of deep pelvic endometriosis: deep endometriosis is defined as a subperitoneal ectopic endometriotic lesion in the pelvis that penetrates the peritoneum to a depth of more than 5 mm [11-12]. The uterosacral ligament is the most frequently involved site, followed by the rectum and bladder. Preoperative diagnosis of deep endometriosis is important for the planning and scoping of surgery, but physical examination, ultrasound or even laparoscopy often fail to meet the diagnostic requirements due to the limitations of the scope. MRI is particularly helpful in the visualization of deep endometriosis lesions, especially in the uterosacral ligament and the vagino-rectal septum.Kinkel et al. described the visualization of deep endometriosis lesions confirmed by surgical pathology by MRI and concluded that T2-weighted images of MRI could detect 100% of endometriosis invasion of the uterosacral ligament. The diagnostic strategy they used was the presence of nodular changes in the proximal uterosacral ligament with a thickness of more than 9 mm [12]. In contrast, Bazot et al. in a recent large prospective study found that the diagnostic strategies of limited thickening, bilateral asymmetry, and morphologic irregularities on the uterosacral ligament with a thickness less than 9 mm had a higher diagnostic specificity compared to thickness measurements alone [13]. MRI can also effectively detect lesions at the utero-rectal fossa and bladder, showing its superiority especially when the utero-rectal fossa is occluded by the lesion [14]. However, it does not show well for rectal involvement. Some investigators have recommended preexamination water enemas, intravenous drugs to inhibit bowel motility, and the use of endorectal coils to improve imaging [11-12]. With technological advances, especially the adoption of endoluminal coils and array coils, the reliability of MRI for evaluating deep endorectal heteropathy has continued to improve. In addition, MRI can monitor the response of endoheteropathy to treatment, as well as predict the efficacy of treatment beforehand. Moreover, MRI may show endoheteropathy invading nerves (e.g., sciatic nerve endoheteropathy) and abdominal wall lesions [5]. In conclusion, MRI has a high value in the diagnosis of endometriosis, especially ectopic endometrial cysts of the ovary and deep endometriosis, and is low operator dependent, produces visual imaging data, and thus plays an increasingly important role in the diagnosis of endometriosis, especially in preoperative evaluation.