Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In Neuroscience letters

To explore the clinical significance of pre-diagnostic curettage under hysteroscopy in the diagnosis of endometrial carcinoma, evaluate the clinical staging, and optimize the function of hysteroscopy computer information system after artificial intelligence algorithm optimization, 200 patients who were diagnosed as endometrial carcinoma in our hospital between June 2015 and July 2019 were included and divided into a hysteroscopy segmental curettage group (101 patients) and a segmented curettage group (99 patients). For patients in both groups, the histopathological examination was performed, and the abdominal lavage fluid was collected. The artificial intelligence algorithm was utilized to optimize the hysteroscopy computer information system. The results showed that in the hysteroscopy segmented curettage group, the coincidence rate of postoperative pathological diagnosis was 97%. In the segmented curettage group, the coincidence rate of pathological histological types diagnosed before and after the surgery was 83.8%. There was a statistically significant difference between the two groups (P < 0.01). The accuracy of the preoperative diagnosis of cervical involvement in the hysteroscopy segmented curettage group was 94%, which was significantly higher than 80% of the segmented curettage group. There was a statistically significant difference between the two groups (P < 0.05). In the hysteroscopy segmented curettage group, the pathological grading of 31 patients had a coincidence rate of 80.6% before and after the surgery. In the segmented curettage group, 45 patients underwent histopathological grading before the surgery. The coincidence rate of pathological grading before and after the surgery was 75.6%. There was no statistically significant difference between the two groups (P > 0.05). The positive rate of patients in the early stage of surgery-pathology reached 16.7%, which was different from the surgery-pathology staging, with statistical significance (P < 0.05), but no statistical difference with other factors (P > 0.05). The type of surgery and the amount of blood loss during surgery were positively correlated with nerve injury in the two groups of patients, with statistical significance (P < 0.05). The imaging quality of hysteroscopy computer information system based on AI algorithm was improved significantly. In conclusion, the curettage under hysteroscopy can accurately diagnose pathological types and clinical staging of endometrial carcinoma. The greater the amount of blood loss during the surgery is, the more likely the nerve injury will occur. The artificial intelligence algorithm was utilized to optimize the hysteroscopy computer information system. Which accurately analyzed the disease-diagnosing images and provided a diagnosis basis for the later treatment of endometrial carcinoma.

Xia Zhiyong, Zhang Liping, Liu Shengfeng, Ran Wei, Liu Yujuan, Tu Jihong

2020-Jun-27

artificial intelligence, computer information, endometrial carcinoma, hysteroscopy, nerve injury