Olympus reviewed the following literature titled "automated machine learning to predict the difficulty for endoscopic resection of gastric gastrointestinal stromal tumor." this retrospective study aimed to provide a dataset consisting of clinical and endoscopic features of patients with gastrointestinal stromal tumors (ggists) from multiple centers.A total of 555 patients were included.The proportion of patients aged < 60 years in the difficult group was 43.0%, while in the non-difficult group, it was 51.7%.The mean absolute errors being 0.021, 0.035 and 0.043.The gbm algorithm outperformed the deep neural net (dl), default random forest (drf), and generalized linear model (glm) algorithms in the validation cohort regarding auc, with a higher value of 0.894 compared to 0.881, 0.858, and 0.854.Furthermore, the accuracy values for the gbm algorithm were the highest compared to the dl, drf, and glm algorithms, with 0.935, 0.870, 0.854, and 0.878.Tumor size greater than 3.0cm was identified as the most critical predictor for difficult procedures, followed by irregular tumor shape, invasion depth beyond mp, history of alcohol consumption, and tumor location in the upper third of the stomach.The automl model based on the gbm algorithm can accurately predict the difficulty for endoscopic resection (er) of ggists before surgery.Severe intraoperative bleeding was characterized by repeated endoscopic hemostasis, a postoperative decrease in hemoglobin levels exceeding 2 g/dl, or necessitating surgical assistance.Postoperative complications included delayed bleeding, delayed perforation, and postoperative infection.Delayed bleeding was defined as clinical evidence of bleeding that occurred after endoscopic resection (er), as evidenced by hematemesis or melena, a decline in hemoglobin levels of more than 2.0 g/dl within 24 hours, or the need for endoscopic therapy.Delayed perforation was verified through x-ray or ct.Postoperative infection was determined by a postoperative body temperature exceeding 37.5°c and/or an increase in inflammatory indicators such as blood routine, crp, or calcitonin.For patients who experienced intraoperative perforation or postoperative infection, antibiotics were prescribed.Type of adverse events/number of patients [developing dataset] severe intraoperative bleeding - 14 patients postoperative complications - 58 patients [test dataset] severe intraoperative bleeding - 4 patients postoperative complications - 16 patients there is no report of any olympus device malfunction in any procedure described in this study.This literature article requires 3 reports.The related patient identifiers are as follows: (b)(4).This medwatch report is for patient identifier (b)(4).
|
This report is being supplemented to provide additional information based on the legal manufacturer's final investigation and to provide information from the author received through follow-up.The device history record was unable to be reviewed for this device since the serial and/or lot number was not provided.However, olympus only releases products to market that meet all manufacturing specifications and final product release criteria.Based on the results of the investigation, the relationship between the device and the adverse events cannot be confirmed.There was no complaint reported on the subject device.There is no evidence of an olympus device malfunction.In addition, the malfunction of the device has not been reported, and from clinical/medical evaluation and risk assessment, it is possible that the reported event is an accident, or a complication associated with a procedure using the subject device.Therefore, the root cause cannot be determined.Olympus will continue to monitor field performance for this device.
|