Preoperative precision and non-invasive identification of ovarian cancer subsectives

Ovarian cancer is a malignant tumor derived from the ovary skin, according to the pathogenesis and tissue origin, can be divided into type I and type II ovarian cancer, type I ovarian cancer growth is slow, mostly early, better prognosis; Preoperative non-invasive precision identification of Type I and Type II helps ovarian cancer patients choose future treatment options and improve prognostics.because of the complex morphology of type I and type II ovarian cancer and its clinical characteristics, it is only identified by the clinician's naked eye, and the subjectivity is strong and the diagnostic accuracy is low. In recent years, imaging histology based on quantitative image analysis and artificial intelligence technology has developed rapidly, which can establish the link between tumor image and tumor tissue pathology, and is widely used in preoperative non-invasive assessment of tumor, which provides a new way of thinking for precise non-invasive identification of type I and type II ovarian cancer.Recently, Gao Xin team of Suzhou Institute of Biomedical Engineering Technology of Chinese Academy of Sciences, in collaboration with Qiang Jinwei of Jinshan Hospital affiliated with Fudan University, jointly carried out the first large sample study of ovarian cancer multi-center based on MRI imaging histology, based on non-invasive machine learning model of type I and type II ovarian cancer, and used visualization technology to identify key areas for the first time in imaging. A total of 294 patients with ovarian cancer (including 143 patients with type I and 151 patients with type II) were included in the study, and patient multi-parameter MRI image data (including T2WI-FS, DWI, ADC, CE-TIWI) were collected. The team extracted high-volume imaging features from patient tumor areas, screened features and modeled them by imaging histology, and the results showed that the imaging histology models built by the team were able to identify type I and type II ovarian cancer with an average accuracy of 83%.Given that CE-T1WI scanning requires injections of contrast agents and that some patients are allergic to contrast agents, the team used only three sequences of T2WI-FS, DWI and ADC to build a lightweight model with an average accuracy of 81% and no significant decline in diagnostic performance, indicating that patients do not need to perform CE-T1WI scans when clinical examination is not necessary. Visual results show that the key areas for identifying patients with type I and TYPE II ovarian cancer are located in the tissue loose area or at the junction of reality and cysticity (Figure 2), and the findings are expected to assist in the positioning of frozen pathological slices in surgery, thus reducing sampling errors. Based on the work of identifying good and malignant ovarian tumors in the early stage, this study is a step closer to realizing the subtype differentiation of ovarian malignant tumors and promoting the process of automatic diagnosis of ovarian tumors. (Suzhou Medical Institute)