DIAP – Double Intraperitoneal Artificial Pancreas

DIAP – Double Intraperitoneal Artificial Pancreas

An easier life with diabetes

Highlights 2021


During 2021, our research have focused on developing equipment to be used in long-term closed loop glucose control in pigs. One highlight of 2021 has been that we during short terms (<24 hours) have been able to keep glucose in the range 4.0 - 10.0 mmol/liter nearly 100% of the time. For us the added value of being in a transdisiplinary centre is the inspiration we get by seeing al the great ideas that are explored.

Scientific publications 2021: 5 paper and 4 abstracts/posters

Project overview

Project lead: Sven Magnus Carlsen
Institution: NTNU
Partners: Artificial Pancreas Trondheim (APT), St. Olavs Hospital, Prediktor Medical AS
Funding: NOK 24.7 mill.
Duration: 2016–end of 2022


All results in the CRIStin-database

Research group

The balance between sugar and insulin is a perpetual challenge in diabetes. It quickly becomes too much or too little. An artificial pancreas will ensure proper blood glucose levels around the clock.

Insulin from the pancreas ensures that your body always has the right amount of glucose (sugar) in the blood. Patients with type 1 diabetes lack this ability to regulate their sugar levels and are dependent on insulin injections many times a day.

The first artificial pancreas that is about to come on the market, has limited effect. It consists of a sensor under the skin that measures sugar level, and a pump outside the body that infuses insulin under the skin. However, such a device is not able to compensate for rapid changes in blood sugar, such as after a meal or strenuous exercise, leaving the patient with either too low or high glucose level.

Scientists in Trondheim work on a completely new way to control glucose levels. They will measure the glucose level inside the abdominal cavity and deliver insulin to the same place. The insulin pump and insulin are placed in a small device on the outside. From this device, an optical fibre for glucose sensing and a thin tube for insulin delivery goes through a small port in the abdominal wall. The port will have a connection that allows for removal of the device whenever needed, for example if the patient wants to go swimming.

The goal is to create a device that mimics the way a healthy body controls blood sugar level, so that the patient does not need to think about diabetes even when he or she is eating or exercising.

This ambitious project involves researchers in medicine, cybernetics, biosensor technology, electronics, product development and materials technology, as well as industry.

The project is led by the research group Artificial Pancreas Trondheim (APT) at NTNU, with St. Olav's Hospital as a research partner and Prediktor Medical AS as an industrial partner.

More about the project 

Mimics the body's pancreas

In the project Double Intraperitoneal Artificial Pancreas, the researchers will develop a robust artificial pancreas with fully automatic control of insulin delivery for patients with diabetes type 1. It will give them a better and longer life.

In 1922, insulin was administered to humans for the first time. It stands as one of the great breakthroughs in the history of medicine. Patients with type 1 diabetes who had previously faced a certain death, could now live a long life, but not without complications. Today, these patients still have 10 years’ shorter life expectancy than healthy persons.

One reason for this increased mortality is the difficulty to control the level of glucose (sugar) in patients with type 1 diabetes using current methods. In a healthy body the pancreas produces extra insulin as soon as blood sugar levels begin to rise. In patients with type 1 diabetes, the beta cells that produce insulin are destroyed by autoimmune disease.

Without cells to produce insulin, these patients are dependent on external insulin supply. This can be achieved by injecting insulin into the subcutaneous tissue by a syringe or an insulin pump. Yet, it is very difficult to accurately control the glucose levels in the blood by this manner, and the patient may become hypo- or hyperglycaemic, which in the long-term is harmful to the body.

It has long been a dream to create an artificial pancreas, but no such device is currently able to work fully automatically. Now researchers in Trondheim aim at developing technology to control the supply of insulin automatically around the clock.

Current methods are not accurate enough


The artificial pancreas which today is about to come on the market, consists of a glucose sensor in a needle inserted under the skin, and an insulin pump that infuses insulin under the skin.

This technology will never be accurate enough. Firstly, there is a time lag between when a glucose change occurs in blood until the detection of said change. The delay may be 8 to 20 minutes. Secondly, it takes about 45 minutes from the insulin is pumped under the skin until its level in the blood reaches the top. However, full effect on glucose metabolism is not reached until one and a half to two hours after insulin was delivered.

Another disadvantage with the current approach is that the sensor must be replaced after a few days, and that it must be calibrated a few times a day by the patient using small blood samples.

The researchers in Trondheim will develop technology that closer mimics glucose metabolism in healthy subjects. They place both glucose sensors and insulin delivery inside the abdominal cavity; the fluid-filled space between the intestines. Measurements show that the glucose level is more in line with blood glucose levels. The time delay from the change happens to detection by the glucose sensor, thus becomes small.

More importantly for the response time is that insulin is absorbed much faster in the abdomen than in the skin, and that very much insulin enters the portal vein and then goes directly to the liver. According to preliminary simulations performed by these researchers, this method makes it possible to normalise the level of blood glucose in type 1 diabetes patients.

A small hole in the abdominal wall

An important part of the project is to construct the port to be placed in the abdominal wall. This port must be small, and the goal is that the outside diameter shall not exceed 1 cm. The optical fibre of the glucose sensor and the tubing for administration of insulin must be correspondingly small with a diameter below 4-5 mm.

The port will have a quick release so that the patient can disconnect the equipment when swimming or doing other activities where an insulin pump is a hindrance.

Another important part of the project is to develop technology for continuous measurement of glucose in the abdominal cavity. Here the industrial partner Prediktor Medical in Fredrikstad has an important role.

A third important part is to figure out how the insulin pump must be programmed to automatically give the correct dose of insulin, after the patient eats and glucose levels rise rapidly in the blood. Researchers will find data on the relationship between glucose readings and how supplied insulin in the abdomen affects glucose levels in the blood.


The project brings together researchers from NTNU and St. Olav's Hospital within the disciplines of cybernetics, mathematical modelling, biosensor technology, biotechnology, biochemistry, optical spectroscopy, veterinary medicine and the medical specialties of endocrinology, anaesthesia, intensive care medicine and pharmacology.

Engineers may think completely differently than physicians and attack the issues in a more systematic way. However, it is not always that the doctors realise what the engineers are talking about and vice versa. The project is designed in such a way that researchers from the different disciplines meet each other with an open mind.

Professors, doctoral students and postdocs have meetings every fortnight where there is discussion at the intersection of medicine and technology. Industry partner Prediktor Medical joins the meetings through Skype.

Three PhD students, two postdocs, two scientists and an engineer are working on the project, in addition to the permanent staff at NTNU.

Socially responsible research and innovation


Two of the researchers in the project have diabetes themselves. They have a good understanding of how the treatment of diabetes is at present.

The project will follow national regulations for research on animals and humans. All experiments on animals are approved by the Mattilsynet (the Norwegian animal research authority). All human studies are reviewed by the regional ethics committee before they can be implemented.


If researchers succeed in this project, it will be very important for the treatment of patients with diabetes type 1. Around 26,000 people in Norway have this disease. Globally, there are millions. The market for the new solution is large.

NTNU and St. Olav's Hospital have therefore secured the intellectual property rights of the new technologies being developed, but they will not commercialize innovations themselves. This may be done by industrial partners at the forefront of medical technology, such as Prediktor Medical. The project management of Artificial Pancreas Trondheim (APT) has consciously chosen to work with a Norwegian company rather than major international partners in order to keep the rights and commercialisation in Norway.