Application Research On Intelligent Auxiliary Diagnosis And Treatment For Li-ALS

Liver disease has always been a serious threat to human health.Artificial liver support system is an effective treatment for liver disease.The preoperative diagnosis of artificial liver and the monitoring of surgical process parameters are of great significance to the effectiveness of artificial liver treatment.Preoperative diagnosis includes the analysis and diagnosis of images and text information based on electronic medical records.Accurate analysis of liver CT images can be of great help in the selection of artificial liver surgery.This article combined the diagnosis and treatment business,using artificial intelligence method for the auxiliary diagnosis of liver CT images.The monitoring of parameters during artificial liver surgery includes various pressure values,peristaltic pump speed,fluid accumulation,temperature,blood ammonia,and patient physiological parameters.The timely monitoring of these artificial liver operating parameters facilitates the advancement of the treatment process.There are no technical solutions for automatically recording and monitoring these operating parameters in existing artificial liver devices currently on the market.Therefore,it is of great research value and application prospects to collect and monitor the operating parameters of the advanced Li-ALS.On the one hand,it can guarantee the patient's life safety in the course of treatment,and on the other hand,it provides the data basis for the improvement of the treatment process.The currently reported methods for auxiliary diagnosis of liver CT images have many deficiencies in the preoperative diagnosis of artificial liver,such as low accuracy of liver and tumor segmentation and the time-consuming problem.In this paper,around the intelligent auxiliary diagnosis and treatment of Li-ALS,a network model for automatic segmentation of the liver based on Fully Convolutional Networks was designed.Multiple experiments fine-tuned the network structure and hyper-parameters.In this paper,the original data is preprocessed,and the input data is a dual-channel input composed of the original image and the window width-adjusted image.The result is a 95.3%Dice score.Based on the segmentation of liver,this paper further optimized the network model and algorithm details,and used the method of introducing category equalization items,background filling,and dilated convolution to segment liver tumors in CT images.The result is an 88.2%Dice score.In this paper,the 3D convolutional neural network was used to classify the tumors in liver CT images.The effect of its application were analyzed The basis for the follow-up study was laid.Based on the study of the algorithm,combined with the characteristics of Li-ALS and my previous experience in the development of Li-ALS,an intelligent auxiliary diagnosis and treatment system based on Li-ALS was designed and implemented in this paper.The system includes the CT image cloud-assisted function and Li-ALS data monitoring function.