A study to Assess the Performance of Do-It-Yourself Artificial Pancreas in Adults with Type 1 Diabetes.

The intervention in this study is 4-week use of a Hybrid Closed loop Android Artificial Pancreas System (AndroidAPS) in patients with type 1 diabetes. Android APS consists of three main components: 1. Dexcom G5® Continuous Glucose Monitoring (CGM): includes a glucose sensor and transmitter attached to the user with ability to measure interstitial fluid glucose continuously. The CGM data are transferred via Bluetooth to Dexcom G5® mobile App (an application on the study android phone). The App can display real time glucose data. 2. Upgraded YpsoPump® insulin pump: An upgraded version of YpsoPump® with bidirectional communication capability will be developed for the purpose of this study to enable communication with AndroidAPS interface.YpsoPump has been approved and is currently being used in Australia. 3. AndroidAPS App: an application on the study android phone which contains the controller algorithm. It receives real time CGM data from Dexcom G5® mobile App and controls the basal insulin delivery by sending commands to YpsoPump®. The basal insulin delivery rate is adjusted every 5 minutes based on varying glucose levels. When it comes to meal bolus, participant needs to enter carbohydrate content of the meal (meal announcement) into the AndroidAPS and the integrated bolus calculator within the App (based on settings determined by the investigators) calculates the required bolus dose which is then sent as a command to YpsoPump® to deliver the dose. Because the system requires meal announcement, it is a hybrid closed-loop system. - Using Ypsopump involves wearing an insulin pump connected to the body via an insulin infusion set. The infusion set is inserted via a serter. - Participants will be admitted to the clinical trial ward for 24 hours for initialization of AndroidAPS system and supervision. The following will be performed: 1. Dexcom CGM will be inserted and participants will be educated about CGM insertion by a diabetes nurse educator. 2. CGM will be linked via transmitter to the Android and Ypsomed pump. 3. Patients will be educated to use the system component including Ypsomed pump, CGM, android phone and associated apps. 4. AndroidAPS will be activated - The day after discharge from the clinical trail ward, participants will be phoned by the study coordinator to assess patient welfare. Data will be reviewed remotely. Clinical investigators will be granted remote access to the participants' real time glucose and insulin delivery data allowing them to follow participants' data remotely throughout this arm of the trial. One week and 3 weeks into intervention, participants will attend clinical trial centre and following will be performed: 1. Participants will be reviewed by Endocrinologist, diabetes educator and research coordinator. 2. Their pump will be uploaded and pump settings will be optimized. - Duration of wash-out period between intervention and control arms is 1 week.