Listening to the patients

• Perez, Laura Lema; Erntsen, Maja Grauff; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2026). Early sound-based meal onset detection using Hidden Markov Models (HMMs). IEEE Journal of Translational Engineering in Health and Medicine. doi: 10.1109/jtehm.2026.3732514. Full text in Research Archive Show summary

  Objective: Meal detection is essential for automated insulin delivery systems in patients with Type 1 Diabetes (T1D), as meals are the primary drivers of rapid postprandial glucose excursions. Current closed-loop systems depend on manual meal announcements, which are often inaccurate or omitted, reducing safety and autonomy. To address this, we developed an automatic meal detector leveraging physiological sound sequences from the body. Methods and procedures: The dataset comprised 50 meal recordings from ten participants (aged 24–30 years) using a throat-mounted microphone. Twenty recordings were collected in quiet conditions, while 30 included deliberate, realistic disturbances. Audio signals were preprocessed, segmented, and transformed into 46 features. Linear Discriminant Analysis (LDA) was implemented as a benchmark, while continuous Hidden Markov Model (HMM)s were applied to capture temporal dependencies and estimate meal onset probability. Results: Performance was evaluated using both an 80:20 train-test split and Leave-One-Subject-Out Cross-Validation (LOSO CV). Explainable Artificial Intelligence (XAI) techniques were employed to interpret model behavior and identify informative features of meal events. LDA achieved 90% accuracy, 78% sensitivity, and a detection delay of 12.3 s, while HMM reached 91% for both accuracy and sensitivity, with an average detection time of 11.7 s. Both models performed well under quiet conditions but were sensitive to background disturbances. Conclusion: These findings demonstrate the feasibility of detecting meal onset from physiological sounds. XAI analysis revealed that HMM effectively captures sequential transitions associated with chewing and swallowing, offering interpretable insights beyond non-temporal methods. Thus, HMM shows potential as a rapid, context-aware meal classifier for closed-loop insulin therapy, thanks to its effective modeling of temporal structure. Clinical impact: This pilot study uses an innovative throat-mounted microphone to detect early meal intake by capturing chewing and swallowing sounds, offering an alternative to traditional measures such as continuous glucose monitoring (CGM).

• Setti, Sunilkumar Telagam; Holten, Andrea; Patil, Pallavi; Perez, Laura Lema; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2024). Feasibility of Meal Onset Detection Using Electrocardiograms. In Costin, Hariton; Petroiu, Gladiola Gabriela & Magjarevic, Ratko (Ed.), Advances in Digital Health and Medical Bioengineering: Proceedings of the 11th International Conference on E-Health and Bioengineering, EHB-2023, November 9–10, 2023, Bucharest, Romania. Volume 1: Medical Devices, Measurements, and Artificial Intelligence Applications. Springer Nature. ISSN 9783031625015. p. 755–762. doi: 10.1007/978-3-031-62502-2_85. Full text in Research Archive
• Teigen, Ingrid Anna; Åm, Marte Kierulf; Riaz, Misbah; Christiansen, Sverre Christian & Carlsen, Sven Magnus (2024). Effects of Low-Dose Glucagon on Subcutaneous Insulin Absorption in Pigs. Current Therapeutic Research: Clinical and Experimental. ISSN 0011-393X. 100. doi: 10.1016/j.curtheres.2024.100736. Full text in Research Archive Show summary

  Background Slow insulin absorption prevents the development of a fully automated artificial pancreas with subcutaneous insulin delivery. Objective We have hypothesized that glucagon could be used as a vasodilator to accelerate insulin absorption in a bihormonal subcutaneous artificial pancreas. The present proof-of-concept study is the first study to investigate the pharmacokinetics of insulin after subcutaneous administration of a low dose of glucagon at the site of subcutaneous insulin injection. Methods Twelve anesthetized pigs were randomized to receive a subcutaneous injection of 10 IU insulin aspart with either 100 µg glucagon or the equivalent volume of placebo (0.9% saline solution) injected at the same site. Arterial samples were collected for 180 minutes to determine insulin, glucagon, and glucose concentrations. Results Glucagon did not influence the insulin concentration Tmax in plasma. The plasma insulin AUC0–∞ was significantly larger after glucagon administration (P < 0.01). The glucagon group had significantly higher glucose concentrations in the first 30 minutes after insulin administration (P < 0.05). Conclusions This proof-of-concept study indicates that glucagon may increase the total absorption of a single dose of subcutaneously injected insulin. This is a novel observation. However, we did not observe any reduction in insulin concentration Tmax, as we had hypothesized. Further, glucagon induced a significant, undesirable increase in early blood glucose concentrations.

• Aguirre-Zapata, Estefania; Alvarez, Hernan; Perez, Laura Lema; Sciascio, Fernando Di & Amicarelli, Adriana (2024). Understanding Interpretability in Modeling Biotechnological Processes: A Critical Review, 2023 XX Workshop on Information Processing and Control (RPIC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350329414. doi: 10.1109/RPIC59053.2023.10530792. Full text in Research Archive
• Perez, Laura Lema; Aguirre-Zapata, Estefania; Fougner, Anders Lyngvi; Amicarelli, Adriana & Alvarez, Hernan (2024). Connecting sensitivity, identifiability and interpretability of a glucose minimal model, 2023 XX Workshop on Information Processing and Control (RPIC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350329414. doi: 10.1109/RPIC59053.2023.10530857. Full text in Research Archive Show summary

  Mathematical models have increased their applications in physiology, control and systems science, and biomedical engineering because they offer the opportunity to examine the structure and behavior of complex physiological systems. In addition, they provide a concise description of complex dynamic processes, indicating ways to improve experimental design and allowing the testing of hypotheses related to physiological structure. When building a mathematical model, it is important to assess the impact of different factors on the overall behavior of the system being modeled and to determine which variables have the most significant influence on the system. Properties such as identifiability, sensitivity, and interpretability in mathematical models are crucial for representing real-world phenomena. In this paper, a sensitivity, structural identifiability, and qualitative parameter interpretability analysis is carried out in a new version of the minimal model of the glucose-insulin system proposed by Bergman. The aim was to evaluate the structure of this new mathematical model and the existing relationship among sensitivity, identifiability, and interpretability of the model parameters. The findings demonstrated the connection between the properties of sensitivity, structural identifiability, and parameter interpretability, which provide important details about the model structure.

• Åm, Marte Kierulf; Teigen, Ingrid Anna; Riaz, Misbah; Fougner, Anders Lyngvi; Christiansen, Sverre Christian & Carlsen, Sven Magnus (2023). The artificial pancreas: two alternative approaches to achieve a fully closed-loop system with optimal glucose control. Journal of Endocrinological Investigation. ISSN 0391-4097. 46. doi: 10.1007/s40618-023-02193-2. Full text in Research Archive Show summary

  Introduction Diabetes mellitus type 1 is a chronic disease that implies mandatory external insulin delivery. The patients must monitor their blood glucose levels and administer appropriate insulin boluses to keep their blood glucose within the desired range. It requires a lot of time and endeavour, and many patients struggle with suboptimal glucose control despite all their efforts. Materials and methods This narrative review combines existing knowledge with new discoveries from animal experiments. Discussion In the last decade, artificial pancreas (AP) devices have been developed to improve glucose control and relieve patients of the constant burden of managing their disease. However, a feasible and fully automated AP is yet to be developed. The main challenges preventing the development of a true, subcutaneous (SC) AP system are the slow dynamics of SC glucose sensing and particularly the delay in effect on glucose levels after SC insulin infusions. We have previously published studies on using the intraperitoneal space for an AP; however, we further propose a novel and potentially disruptive way to utilize the vasodilative properties of glucagon in SC AP systems. Conclusion This narrative review presents two lesser-explored viable solutions for AP systems and discusses the potential for improvement toward a fully automated system: A) using the intraperitoneal approach for more rapid insulin absorption, and B) besides using glucagon to treat and prevent hypoglycemia, also administering micro-boluses of glucagon to increase the local SC blood flow, thereby accelerating SC insulin absorption and SC glucose sensor site dynamics.

• Cheema, Muhammad Asaad; Patil, Pallavi; Siddiqui, Salman Ijaz; Rossi, Pierluigi Salvo; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2023). Data-Driven Classifiers for Early Meal Detection Using ECG. IEEE Sensors Letters. 7(9), p. 1–4. doi: 10.1109/LSENS.2023.3307106. Full text in Research Archive Show summary

  This study investigates the potential of the electrocardiogram (ECG) to perform early meal detection, which is critical for developing a fully-functional automatic artificial pancreas. The study was conducted in a group of healthy subjects with different ages and genders. Two classifiers were trained: one based on neural networks (NNs) and working on features extracted from the signals; one based on convolutional NNs (CNNs) and working directly on raw data. During the test phase, both classifiers correctly detected all the meals, with the CNN outperforming the NN in terms of misdetected meals (MMs) and detection time (DT). Reliable meal onset detection with short detection time has significant practical implications: it reduces the risk of postprandial hyperglycemia and hypoglycemia, it reduces the mental burden of meal documentation for patients and related stress.

• Teigen, Ingrid Anna; Riaz, Misbah; Åm, Marte Kierulf; Christiansen, Sverre Christian & Carlsen, Sven Magnus (2022). Vasodilatory effects of glucagon: A possible new approach to enhanced subcutaneous insulin absorption in artificial pancreas devices. Frontiers in Bioengineering and Biotechnology. 10. doi: 10.3389/fbioe.2022.986858. Full text in Research Archive Show summary

  Patients with diabetes mellitus type 1 depend on exogenous insulin to keep their blood glucose concentrations within the desired range. Subcutaneous bihormonal artificial pancreas devices that can measure glucose concentrations continuously and autonomously calculate and deliver insulin and glucagon infusions is a promising new treatment option for these patients. The slow absorption rate of insulin from subcutaneous tissue is perhaps the most important factor preventing the development of a fully automated artificial pancreas using subcutaneous insulin delivery. Subcutaneous insulin absorption is influenced by several factors, among which local subcutaneous blood flow is one of the most prominent. We have discovered that micro-doses of glucagon may cause a substantial increase in local subcutaneous blood flow. This paper discusses how the local vasodilative effects of micro-doses of glucagon might be utilised to improve the performance of subcutaneous bihormonal artificial pancreas devices. We map out the early stages of our hypothesis as a disruptive novel approach, where we propose to use glucagon as a vasodilator to accelerate the absorption of meal boluses of insulin, besides using it conventionally to treat hypoglycaemia.

• Cheema, Muhammad Asaad; Siddiqui, Salman Ijaz & Rossi, Pierluigi Salvo (2022). Comparison of Different Classifiers for Early Meal Detection Using Abdominal Sounds. Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop. ISSN 1551-2282. doi: 10.1109/SAM53842.2022.9827841. Full text in Research Archive
• Teigen, Ingrid Anna; Åm, Marte Kierulf; Carlsen, Sven Magnus & Christiansen, Sverre Christian (2022). Pharmacokinetics of glucagon after intravenous, intraperitoneal and subcutaneous administration in a pig model. Basic & Clinical Pharmacology & Toxicology. ISSN 1742-7835. 130(6), p. 623–631. doi: 10.1111/bcpt.13731. Full text in Research Archive
• Benam, Karim Davari; Khoshamadi, Hasti; Åm, Marte Kierulf; Stavdahl, Øyvind; Gros, Sebastien & Fougner, Anders Lyngvi (2022). Identifiable Prediction Animal Model for the Bi-Hormonal Intraperitoneal Artificial Pancreas. Journal of Process Control. ISSN 0959-1524. 121, p. 13–29. doi: 10.1016/j.jprocont.2022.11.008. Full text in Research Archive Show summary

  To achieve a fully automatic artificial pancreas (AP), i.e., an AP without the need for meal announcements, the intraperitoneal (IP) route is explored. This route has faster dynamics than the typical subcutaneous (SC) route. Model predictive control (MPC) is the most promising control algorithm, but it requires a predictive and identifiable model. This paper presents the design of such a model for MPC-based dual hormone IP APs. This model is trained and tested on recorded data from anesthetized pigs. Animal experiments show that the saturation of the hepatic first-pass effect is essential in how IP insulin and IP glucagon affect glucose levels. These physiological phenomena must be modeled to estimate the system behavior for various conditions. This, in turn, increases the number of parameters and complicates system identification. The availability of rich experimental data from 26 animal trials motivated the design of a technique to exploit this prior information to ensure the identifiability of our model. Through this technique, most parameters were either modeled as body weight functions or common among animals. The correlation between parameter values and body weight is discovered utilizing prior data from various animal experiments, such as blood glucose, plasma insulin, and glucagon levels, in which hormones were administered intraperitoneally or intravenously. This method simplifies the system identification for every new subject while keeping the model’s essential details that improve the prediction capability relative to comparable models. The model can be exploited in MPC or any other model-based controller of a bi-hormonal IP AP. It can also be used as a simulator to develop control approaches for single and bi-hormonal IP APs.

• Teigen, Ingrid Anna; Åm, Marte Kierulf; Carlsen, Sven Magnus & Christiansen, Sverre Christian (2021). Pharmacokinetics of Intraperitoneally Delivered Glucagon in Pigs: A Hypothesis of First Pass Metabolism. European journal of drug metabolism and pharmacokinetics. ISSN 0378-7966. doi: 10.1007/s13318-021-00692-2. Full text in Research Archive Show summary

  Background and Objective Artificial pancreases administering low-dose glucagon in addition to insulin have the scope to improve glucose control in patients with diabetes mellitus type 1. If such a device were to deliver both hormones intraperitoneally, it would mimic normal physiology, which may be beneficial. However, the pharmacokinetic properties of glucagon after intraperitoneal administration are not well known. Hence, the current study aims to evaluate the relationship between the amount of intraperitoneally delivered glucagon and pharmacokinetic variables in a pig model. Methods Pharmacokinetic data was retrieved from experiments on 19 anaesthetised pigs and analysed post hoc. The animals received a single intraperitoneal bolus of glucagon ranging from 0.30 to 4.46 µg/kg. Plasma glucagon was measured every 2–10 min for 50 min. Results Peak plasma concentration and area under the time–plasma concentration curve of glucagon correlated positively with the administered dose, and larger boluses provided a relatively greater increase. The mean (standard deviation) time to maximum glucagon concentration in plasma was 11 (5) min, and the mean elimination half-life of glucagon in plasma was 19 (7) min. Conclusions Maximum plasma concentration and area under the time–plasma concentration curve of glucagon increase nonlinearly in relation to the intraperitoneally administered glucagon dose. We hypothesise that the results are compatible with a satiable first-pass metabolism in the liver. Time to maximum glucagon concentration in plasma and the elimination half-life of glucagon in plasma seem independent of the drug dose.

• Setti, Sunilkumar Telagam; Søiland, Elise; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2020). Pilot study of Early Meal Onset Detection from Abdominal Sounds. In Costin, Hariton (Eds.), IEEE International Conference on e-Health and Bioengineering (EHB 2019), Proceedings of the. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728126036. doi: 10.1109/EHB47216.2019.8969901. Full text in Research Archive Show summary

  A typical artificial pancreas depends only on the continuous glucose monitoring (CGM) value for insulin dosing. However, both the insulin infusion and the glucose sensing are subject to time delays and slow dynamics. An automated and reliable meal onset information could enhance the control outcome of artificial pancreas by making it possible to infuse insulin earlier and thereby avoid large postprandial glucose excursions. In this study we employ abdominal sounds recorded in two healthy volunteers with a condenser microphone and propose an automated approach for meal onset detection from abdominal sounds. We use the Mel-frequency cepstral coefficients (MFCCs) and wavelet entropy extracted from the abdominal sounds as features. These features are fed to a simple feed forward neural network for discriminating meal from no-meal abdominal sounds. This approach detects meal onset with an average delay of 4.3 minutes in our limited number of subjects. More importantly, it provides lesser response delay than the state-of-the-art CGM based approach, which achieved a response delay ranging from 30- 40 minutes. The preliminary results indicate that the proposed abdominal sound-based approach may provide early meal onset information. This can be exploited in an artificial pancreas through allowable earlier meal insulin boluses, resulting in improved glycemic control.

• Kölle, Konstanze; Aftab, Muhammad Faisal; Andersson, Leif Erik; Fougner, Anders Lyngvi & Stavdahl, Øyvind (2019). Data driven filtering of bowel sounds using multivariate empirical mode decomposition. Biomedical engineering online. 18(28), p. 1–20. doi: 10.1186/s12938-019-0646-1. Full text in Research Archive Show summary

  Background The analysis of abdominal sounds can help to diagnose gastro-intestinal diseases. Sounds originating from the stomach and the intestine, the so-called bowel sounds, occur in various forms. They are described as loose successions or clusters of rather sudden bursts. Realistic recordings of abdominal sounds are contaminated with noise and artifacts from which the bowel sounds must be differentiated. Methods The proposed intrinsic mode function-fractal dimension (IMF-FD) filtering utilizes the property of the multivariate empirical mode decomposition (MEMD) to behave as a series of band pass filters. The MEMD decomposes the abdominal signal into its different frequency components. The resulting intrinsic mode functions (IMFs) are modulated in amplitude and frequency where transient sonic events occur. Based on the complexity of the IMFs, measured by their fractal dimension (FD) in sliding windows, the information-carrying IMFs are selected. The filtered signal is formed as the superposition of all selected IMFs. The IMF-FD filter not only enhances the non-linear components of the original signal but also segments them from the rest. Another important aspect of this work is that typical artifacts that occur in the same frequency range as bowel sounds can be subsequently eliminated by heuristic rules. Conclusions The method is tested on a realistic, contaminated data set with promising performance: close to 100% of the manually labeled bowel sounds are identified.

• Kölle, Konstanze; Fougner, Anders Lyngvi; Ellingsen, Reinold; Carlsen, Sven Magnus & Stavdahl, Øyvind (2019). Feasibility of early meal detection based on abdominal sound. IEEE Journal of Translational Engineering in Health and Medicine. 7:3300212, p. 1–12. doi: 10.1109/JTEHM.2019.2940218. Full text in Research Archive Show summary

  Background In classical approaches for an artificial pancreas, continuous glucose monitoring (CGM) is the only measured variable used for insulin dosing and additional control functions. The CGM values are subject to time delays and slow dynamics between blood and the sensing location. These time lags compromise the controller’s performance in maintaining (near to) normal glucose levels. Meal information could enhance the control outcome. However, meal announcement by the user is not reliable, and it takes 30 min to 40 min from meal onset until a meal is detected by methods based on CGM. In this pilot study, the use of bowel sounds for meal detection was investigated. In particular, we focused on whether bowel sounds change qualitatively during or shortly after meal ingestion. Methods After fasting for at least 4 h, 11 healthy volunteers ingested a lunch meal at their usual time. Abdominal sound was recorded by a condenser microphone that was attached to the right upper quadrant of the abdomen by medical tape. Features that describe the power distribution over the frequency spectrum were extracted and used for classification by support vector machines. These classifiers were trained in a leave-one-out cross-validation scheme. Results Meals could be detected on average after less than 10 min with the best parameter choice. Conclusion This shows that abdominal sound monitoring could provide an early meal detection. Further studies should investigate this possibility on a larger population in more general settings.

View all works in NVA

• Patil, Pallavi; Perez, Laura Lema; Siddiqui, Salman Ijaz; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2023). Electrocardiogram as a Predictor for Early and Robust Meal Onset Detection in Artificial Pancreas. Full text in Research Archive
• Carlsen, Sven Magnus (2023). Kunstig pankreas, - en oppdatering. Full text in Research Archive
• Setti, Sunilkumar Telagam; Holten, Andrea; Patil, Pallavi; Perez, Laura Lema; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2023). Feasibility of Meal Onset Detection Using Electrocardiograms. Full text in Research Archive
• Perez, Laura Lema; Aguirre-Zapata, Estefania; Fougner, Anders Lyngvi; Amicarelli, Adriana & Alvarez, Hernan (2023). Connecting sensitivity, identifiability and interpretability of a glucose minimal model. Full text in Research Archive
• Aguirre-Zapata, Estefania; Alvarez, Hernan; Perez, Laura Lema; Sciascio, Fernando Di & Amicarelli, Adriana (2023). Understanding Interpretability in Modeling Biotechnological Processes: A Critical Review. Full text in Research Archive
• Tabella, Gianluca; Siddiqui, Salman Ijaz & Rossi, Pierluigi Salvo (2022). Quickest Meal Detection Using Statistical Machine Learning. Full text in Research Archive
• Siddiqui, Salman Ijaz; Isifan, Ahmed Khalil Omar; Klavins, Davis & Fougner, Anders Lyngvi (2022). Early Meal Detection Based on Different Body Sounds. Full text in Research Archive
• Teigen, Ingrid Anna; Åm, Marte Kierulf; Carlsen, Sven Magnus & Christiansen, Sverre Christian (2022). Glucagon Pharmacokinetics in a Pig Model. Full text in Research Archive
• Åm, Marte Kierulf; Munkerud, Erlend Yttersian; Berge, Mathilde Hallem; Christiansen, Sverre Christian & Carlsen, Sven Magnus (2022). Glucagon and local subcutaneous blood flow; a proof-of-concept study. Full text in Research Archive
• Setti, Sunilkumar Telagam; Søiland, Elise; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2020). Pilot Study of Exploiting Abdominal Sound for Early Meal Onset Detection. Full text in Research Archive
• Setti, Sunilkumar Telagam; Søiland, Elise; Stavdahl, Øyvind & Fougner, Anders Lyngvi (2019). Pilot study of Early Meal Onset Detection from Abdominal Sounds. Full text in Research Archive
• Teigen, Ingrid Anna; Carlsen, Sven Magnus & Christiansen, Sverre Christian (2023). The Bihormonal Artificial Pancreas: New Perspectives on the Pharmacokinetics and Pharmacodynamics of Glucagon. Norges teknisk-naturvitenskapelige universitet. ISSN 9788232668779. Full text in Research Archive Show summary

  Diabetes mellitus type 1 is an autoimmune disease that causes selective destruction of the insulin-producing β-cells in the pancreatic islet of Langerhans. Insulin induces glucose storage and reduces blood glucose concentrations. As it is essential for normal glucose regulation, patients with diabetes mellitus type 1 require lifelong treatment with exogenous insulin. The treatment of diabetes mellitus type 1 has evolved significantly since insulin was discovered in 1921. Modern treatment comprises continuous self-monitoring of glucose concentrations and self-administration of appropriate insulin dosages. Most patients cover their basal and prandial insulin needs through multiple manual injections or infusions via a pump into subcutaneous tissue. Hypoglycaemic episodes are a common side-effect of insulin treatment, and can be reversed with glucagon, which mobilises glucose from the liver and increases blood glucose concentrations. Lately, artificial pancreas technology has emerged as a promising new treatment option for patients with diabetes mellitus type 1. An artificial pancreas is a device that automatically calculates the demand for insulin based on continuous glucose feedback. Both unihormonal devices, delivering only insulin, and bihormonal devices, delivering insulin and glucagon, have been developed. However, only hybrid (partially automated), unihormonal devices that require meal announcements from the user are commercially available. Developing a fully automated subcutaneous artificial pancreas that can achieve excellent glucose control without meal announcements is challenging. One major barrier is the slow absorption rate of insulin from the subcutaneous tissue, which makes timing and adjustments of meal boluses of insulin difficult. As such, measures that can promote insulin absorption are needed. Alternatively, a more efficient route of insulin administration could be chosen. Intraperitoneal insulin administration mimics normal physiology and has a favourable pharmacokinetic profile compared to subcutaneous insulin administration. The first aim of this thesis was to investigate if the absorption, elimination, and utilisation of glucagon may also be improved with intraperitoneal drug delivery, as this could support developing a bihormonal intraperitoneal artificial pancreas, that would deliver both insulin and glucagon into the intraperitoneal space. The second aim was to investigate if subcutaneous insulin absorption may be accelerated by simultaneous subcutaneous glucagon administration, as glucagon is a potent vasodilator. This thesis demonstrates that intraperitoneal delivery of glucagon results in lower systemic glucagon concentrations than subcutaneous administration, probably because of a large first-pass metabolism of glucagon in the liver. However, the effects on glucose metabolism are equal after intraperitoneal and subcutaneous glucagon administration. Because of the increased risk of complications with intraperitoneal drug delivery, we conclude that the subcutaneous route for glucagon delivery is preferable for most patients. Simultaneous administration of glucagon did not accelerate subcutaneous insulin absorption. However, the total amount of insulin absorbed after a single bolus was significantly increased over an observation period of three hours. Clinical trials are needed to investigate if micro-doses of glucagon can be used to enhance insulin absorption and improve postprandial glucose control in subjects with diabetes mellitus type 1.

• Kölle, Konstanze; Stavdahl, Øyvind; Fougner, Anders Lyngvi & Carlsen, Sven Magnus (2018). Towards a Safe Artificial Pancreas: Meal Detection and the Intraperitoneal Route. Norges teknisk-naturvitenskapelige universitet. ISSN 9788232635597. Full text in Research Archive Show summary

  Patients with diabetes mellitus type 1 have insufficient insulin production. If untreated, the insulin deficiency causes diabetic ketoacidosis and death. Long-term elevated blood glucose levels lead to a number of complications. On the other hand, insulin must be carefully dosed to avoid life-threatening hypoglycemia. Food, exercise, as well as daily and seasonal hormonal variations act as disturbances on the glucose homoeostasis and render the treatment a challenge. The aim of this thesis was to develop safety functions that may enable fully automated closed-loop glucose control systems in diabetes mellitus type 1. Safety is defined as the absence of risks. Risks in the context of glucose regulation arise from hyperglycemia and hypoglycemia. A risk analysis of a conceptual artificial pancreas by means of failure modes and effects analysis elucidated that a large variety of system faults may occur. The risk analysis revealed among others that: (1) the glucose homoeostasis is most frequently disturbed by meals, (2) the subcutaneous tissue as the typical route for glucose sensing and insulin infusion is associated with slow dynamics and several other challenges. The achievable postprandial regulation by insulin bolusing upon meal detection was investigated in simulations. Published methods were used that trigger a meal detection when the glucose concentration or its first two derivatives exceed thresholds. Even with the surprisingly fast dynamics that had been reported for glucose sensing in the peritoneal cavity, the postprandial regulation was insufficient. Furthermore, the simulation study recommended to individualise thresholds which is hard to realise in outpatient settings. A non-individualised meal detection based on CGM data was proposed. The method exploits the patterns in horizons of CGM measurements by classifying them either directly or after processing by moving horizon estimation. The classification by linear discriminant analysis was tested in simulations and validated on a clinical data set. Higher sensitivity and fewer false alarms were demonstrated compared to the threshold-based methods. Abdominal sounds were studied as a redundant or alternative information source of food intake. The pilot study investigated whether abdominal sounds change fast enough for an early meal detection in the context of an artificial pancreas. The study showed that the distribution of power in frequency intervals increases within a few minutes after meal start. This enables meal detection by pattern recognition in the early postprandial period. Moreover, a data-driven filtering method was proposed that segments bowel sounds from background noise. The challenges associated with subcutaneous glucose sensing and insulin infusion hamper the development of an artificial pancreas. The peritoneal cavity was explored in pre-clinical trials as alternative route for glucose sensing. Glucose measurements were fitted to a differential model to quantify the dynamics by means of time delay and time constant. Intraperitoneal glucose sensing was less rapid than previously reported but still faster than subcutaneous glucose sensing. This thesis has contributed towards the goal of a safe artificial pancreas by assessing and developing methods for meal detection and insights into intraperitoneal glucose sensing.

View all works in NVA