As a severe malignant tumor of the digestive system,the highly invasive pancreatic cancer lacks typical preliminary symptoms. Rapid metastatic dissemination and difficulty in early-stage diagnosis preclude the chance of radical curative resection,hence resulting in a poor overall prognosis in most patients. In recent years,the wide application of the artificial intelligence(AI),represented by machine learning and deep learning,has developed rapidly in the field of medicine. All sorts of models based on AI have been applied to the screening, early diagnosis, treatment, prognosis prediction of patients with pancreatic cancer.Three-dimentional visualization and augmented reality navigation technologies have also been developed and applied in pancreatic cancer surgery.This paper reviews the status quo of AI application in pancreatic cancer from various aspects,and anticipates its future application prospects.
Humans ; Artificial Intelligence ; Pancreatic Neoplasms/surgery* ; Pancreas ; Machine Learning

Artificial pancreas is a technique developed to automatically control blood glucose in people with diabetes by providing an endocrine function instead of a healthy pancreas. The technique was developed for the replacement of insulin secretion deficiencies among various exocrine and endocrine functions of the pancreas and is mainly used for people with type 1 diabetes or those who need intensive insulin treatment. This review briefly summarizes the working principles, components, recent clinical research, and future perspectives of artificial pancreas.
Blood Glucose ; Diabetes Mellitus ; Extracellular Fluid ; Glucose ; Insulin ; Pancreas ; Pancreas, Artificial*

The goals of this research are to improve the functionality (insulin secretion rate and pattern) and to expand the life-span of immunoprotected pancreatic islets. The low functionality (less than 15% of the insulin release rate of native islets in pancreas) required a large number of islets within the implant, which causes complications in surgery and discomfort for patients. The limited life-span of the islets in a biohybrid artificial pancreas (BAP) may require frequent cell reseeding and cause further supply problems in islet transplantation. Improved islet functionality and prolonged life-span will minimize the volume of the BAP by reducing the number of islets needed for diabetic patients to achieve normoglycaemia and reduce problems associated with islet supply. It is hypothesized in this research that 1) by mimicking facilitated oxygen transport in avascular tissues, the immunoprotected islets release a higher amount of insulin, recover their intrinsic biphasic release pattern, and prolong their life-span, and 2) insulinotropic agents further promote insulin secretion from islets. Based on these hypotheses, a new BAP system will be designed which contains the water-soluble polymeric conjugates of oxygen carriers (or oxygen binding vehicles) and islet stimulants of sulfonylurea compounds and glucagon-like insulinotropic peptide-1 with entrapped islets in the BAP. The research examines their effects on islet viability, the amount of insulin secretion, the insulin release profile, and the life-span of immunoprotected pancreatic islets. Especially, the combined synergy effects of both hypotheses will be emphasized. The successful results in improving functionality and life- span of islets entrapped in an immunoprotected membrane can be applied in the delivery of microencapsulated therapeutic cells and to the miniaturization of a BAP. In addition, the approaches proposed in this research will provide a potential solution to the shortage problem of human cell or tissue sources.
Diabetes Mellitus, Type 1/*surgery ; Diffusion Chambers, Culture ; Humans ; Islets of Langerhans ; *Pancreas, Artificial

Type 1 diabetes is an autoimmune disease with insulin deficiency which causes microvascular complications such as retinopathy, nephropathy and neuropathy. There have been some trials to simulate the pancreatic endocrine function of insulin and glucagon for homeostatic equilibration of blood glucose, developing artificial pancreas. There are three major functional components of the modern artificial pancreas, a continuous glucose-monitoring system, an insulin-infusion pump and a control algorithm. There are commercially available continuous glucose monitoring systems with subcutaneous glucose measuring, however, there have been many attempts to develop more efficient glucose monitoring systems, including noninvasive systems. Thanks to technological advances and the miniaturization of electronics, recent advances in the accuracy and performance of these systems have placed research on the threshold of prototype commercial devices and large-scale outpatient feasibility studies. In addition, smartphone technology has created the opportunity for caregivers to receive push notification alerts and makes it possible to provide patients with advisory or decision-support systems. Even though there are still some remaining challenges to develop a successful artificial pancreas, glucose control in type 1 diabetes will be more efficient with its advent.
Autoimmune Diseases ; Blood Glucose ; Caregivers ; Glucagon ; Glucose ; Humans ; Insulin ; Miniaturization ; Outpatients ; Pancreas, Artificial