Real-Time Detection of Infusion Site Failures in a Closed-Loop Artificial Pancreas

Development of a Real-time Force-based Algorithm for Infusion Failure Detection

BACKGROUND

Continuous subcutaneous insulin infusion (CSII) is a common treatment option for people with diabetes (PWD), but insulin infusion failures pose a significant challenge, leading to hyperglycemia, diabetes burnout, and increased hospitalizations. Current CSII pumps' occlusion alarm systems are limited in detecting infusion failures; therefore, a more effective detection method is needed.

METHODS

We conducted five preclinical animal studies to collect data on infusion failures, utilizing both insulin and non-insulin boluses. Data were captured using in-line pressure and flow rate sensors, with additional force data from CSII pumps' onboard sensors in one study. A novel classifier model was developed using this dataset, aimed at detecting different types of infusion failures through direct utilization of force sensor data. Performance was compared against various occlusion alarm thresholds from commercially available CSII pumps.

RESULTS

The testing dataset included 251 boluses. The Bagging classifier model showed the highest performance metrics among the models tested, exhibiting high accuracy (96%), sensitivity (94%), and specificity (98%), with lower false-positive and false-negative rate compared with traditional occlusion alarm pressure thresholds.

CONCLUSIONS

Our study developed a novel non-threshold classifier that outperforms current occlusion alarm systems in CSII pumps in detecting infusion failures. This advancement has the potential to reduce the risk of hyperglycemia and hospitalizations due to undetected infusion failures, offering a more reliable and effective CSII therapy for PWD. Further studies involving human participants are recommended to validate these findings and assess the classifier's performance in a real-world setting.

Artificial Intelligence and Machine Learning for Improving Glycemic Control in Diabetes: Best Practices, Pitfalls, and Opportunities

OBJECTIVE

Artificial intelligence and machine learning are transforming many fields including medicine. In diabetes, robust biosensing technologies and automated insulin delivery therapies have created a substantial opportunity to improve health. While the number of manuscripts addressing the topic of applying machine learning to diabetes has grown in recent years, there has been a lack of consistency in the methods, metrics, and data used to train and evaluate these algorithms. This manuscript provides consensus guidelines for machine learning practitioners in the field of diabetes, including best practice recommended approaches and warnings about pitfalls to avoid.

METHODS

Algorithmic approaches are reviewed and benefits of different algorithms are discussed including importance of clinical accuracy, explainability, interpretability, and personalization. We review the most common features used in machine learning applications in diabetes glucose control and provide an open-source library of functions for calculating features, as well as a framework for specifying data sets using data sheets. A review of current data sets available for training algorithms is provided as well as an online repository of data sources.

SIGNIFICANCE

These consensus guidelines are designed to improve performance and translatability of new machine learning algorithms developed in the field of diabetes for engineers and data scientists.

Frequency and Detection of Insulin Infusion Site Failure in the Type 1 Diabetes Exchange Online Community

Insulin infusion site (IIS) failures are a weakness in insulin pump therapy. We examined experience with IIS failures among U.S. individuals with diabetes on insulin pump through survey distributed to the T1D Exchange Online Community. Demographic factors, IIS characteristics, and diabetes-related perceptions were assessed by logistic regression to determine odds of higher (≥1 per month) or lower (<1 per month) reported IIS failure frequency. IIS failures were common; 41.4% reported ≥1 per month. IIS failure is usually detected through development of hyperglycemia rather than pump alarm. No assessed demographic factor or IIS characteristic was predictive; however, higher odds of ≥1 failure per month were associated with feelings of burnout (odds ratios [OR] 1.489 [1.024, 2.165]) and considering pump discontinuation (OR 2.233 [1.455, 3.427]). IIS failures are frequent and unpredictable, typically require hyperglycemia for detection, and are associated with negative perceptions. More should be done toward preventing IIS failures and/or detecting them sooner.

Machine Learning-Based Anomaly Detection Algorithms to Alert Patients Using Sensor Augmented Pump of Infusion Site Failures

BACKGROUND

Personal insulin pumps have shown to be effective in improving the quality of therapy for people with type 1 diabetes (T1D). However, the safety of this technology is limited by the possible infusion site failures, which are linked with hyperglycemia and ketoacidosis. Thanks to the large availability of collected data provided by modern therapeutic technologies, machine learning algorithms have the potential to provide new way to identify failures early and avert adverse events.

METHODS

A clinical dataset (N = 20) is used to evaluate a novel method for detecting real-time infusion site failures using unsupervised anomaly detection algorithms, previously proposed and developed on in-silico data. An adapted feature engineering procedure is introduced to make the method able to operate in the absence of a closed-loop (CL) system and meal announcements.

RESULTS

In the optimal configuration, we obtained a performance of 0.75 Sensitivity (15 out of 20 total failures detected) and 0.08 FP/day, outperforming previously proposed literature algorithms. The algorithm was able to anticipate the replacement of the malfunctioning infusion sets by ~2 h on average.

CONCLUSIONS

On the considered dataset, the proposed algorithm showed the potential to improve the safety of patients treated with sensor-augmented pump systems.

Empfehlungen zur Diabetes-Behandlung mit automatischen Insulin-Dosierungssystemen

Das Prinzip der automatischen Insulindosierung, kurz „AID“ genannt, zeigt in Zulassungsstudien und Real-World-Erfahrungen ausgezeichnete Behandlungsergebnisse. Beim AID wird eine Insulinpumpe mit einem System zur kontinuierlichen Glukosemessung zusammengeschaltet, während ein Rechenprogramm, der sogenannte Algorithmus, die Steuerung der Insulingabe nach Bedarf übernimmt. Idealerweise wäre das System ein geschlossener Kreis, bei dem die Menschen mit Diabetes keine Eingabe mehr machen müssten. Jedoch sind bei den heute verfügbaren Systemen verschiedene Grundeinstellungen und Eingaben erforderlich (insbesondere von Kohlenhydratmengen der Mahlzeiten oder körperlicher Aktivität), die sich von den bisherigen Empfehlungen der sensorunterstützten Pumpentherapie in einzelnen Aspekten unterscheiden. So werden die traditionellen Konzepte von „Basal“ und „Bolus“ mit AID weniger nützlich, da der Algorithmus beide Arten der Insulinabgabe verwendet, um die Glukosewerte dem eingestellten Zielwert zu nähern. Daher sollte bei diesen Systemen statt der Erfassung von „Basal“ und „Bolus“, zwischen einer „nutzerinitiierten“ und einer „automatischen“ Insulindosis unterschieden werden. Gemeinsame Therapieprinzipien der verschiedenen AID-Systeme umfassen die passgenaue Einstellung des Kohlenhydratverhältnisses, die Bedeutung des Timings der vom Anwender initiierten Insulinbolusgaben vor der Mahlzeit, den korrekten Umgang mit einem verzögerten oder versäumten Mahlzeitenbolus, neue Prinzipien im Umgang mit Sport oder Alkoholgenuss sowie den rechtzeitigen Umstieg von AID zu manuellem Modus bei Auftreten erhöhter Ketonwerte. Das Team vom Diabetes-Zentrum AUF DER BULT in Hannover hat aus eigenen Studienerfahrungen und der zugrunde liegenden internationalen Literatur praktische Empfehlungen zur Anwendung und Schulung der gegenwärtig und demnächst in Deutschland kommerziell erhältlichen Systeme zusammengestellt. Für den Erfolg der AID-Behandlung scheint das richtige Erwartungsmanagement sowohl beim Behandlungsteam und als auch beim Anwender von großer Bedeutung zu sein.

Human-in-the-Loop Insulin Dosing

The last ten years of efforts in developing automated insulin dosing systems have led to one hybrid closed-loop device in the US market with more in the late stages of development. Much of the focus has been on algorithms, including closed-loop, detection of sensor and pump faults, and safety. There has been less discussion in the open literature about user interface design and related options. This article provides perspectives on automated insulin delivery (AID) system design by analyzing commonly used devices, such as bicycles and car entertainment systems. The recent Boeing 737 Max 8 disasters are used to highlight related challenges with AID systems. The role that system engineers can play in the do it yourself artificial pancreas system movement is also discussed. The human-in-the-loop remains by far the most important "component" of any AID system.

Practical implementation of automated closed-loop insulin delivery: A French position statement

Automated closed-loop (CL) insulin therapy has come of age. This major technological advance is expected to significantly improve the quality of care for adults, adolescents and children with type 1 diabetes. To improve access to this innovation for both patients and healthcare professionals (HCPs), and to promote adherence to its requirements in terms of safety, regulations, ethics and practice, the French Diabetes Society (SFD) brought together a French Working Group of experts to discuss the current practical consensus. The result is the present statement describing the indications for CL therapy with emphasis on the idea that treatment expectations must be clearly defined in advance. Specifications for expert care centres in charge of initiating the treatment were also proposed. Great importance was also attached to the crucial place of high-quality training for patients and healthcare professionals. Long-term follow-up should collect not only metabolic and clinical results, but also indicators related to psychosocial and human factors. Overall, this national consensus statement aims to promote the introduction of marketed CL devices into standard clinical practice.

Realizing a Closed-Loop (Artificial Pancreas) System for the Treatment of Type 1 Diabetes

Recent, rapid changes in the treatment of type 1 diabetes have allowed for commercialization of an "artificial pancreas" that is better described as a closed-loop controller of insulin delivery. This review presents the current state of closed-loop control systems and expected future developments with a discussion of the human factor issues in allowing automation of glucose control. The goal of these systems is to minimize or prevent both short-term and long-term complications from diabetes and to decrease the daily burden of managing diabetes. The closed-loop systems are generally very effective and safe at night, have allowed for improved sleep, and have decreased the burden of diabetes management overnight. However, there are still significant barriers to achieving excellent daytime glucose control while simultaneously decreasing the burden of daytime diabetes management. These systems use a subcutaneous continuous glucose sensor, an algorithm that accounts for the current glucose and rate of change of the glucose, and the amount of insulin that has already been delivered to safely deliver insulin to control hyperglycemia, while minimizing the risk of hypoglycemia. The future challenge will be to allow for full closed-loop control with minimal burden on the patient during the day, alleviating meal announcements, carbohydrate counting, alerts, and maintenance. The human factors involved with interfacing with a closed-loop system and allowing the system to take control of diabetes management are significant. It is important to find a balance between enthusiasm and realistic expectations and experiences with the closed-loop system.

Detection of Insulin Pump Malfunctioning to Improve Safety in Artificial Pancreas Using Unsupervised Algorithms

BACKGROUND

Recent development of automated closed-loop (CL) insulin delivery systems, the so-called artificial pancreas (AP), improved the quality of type 1 diabetes (T1D) therapy. As new technologies emerge, patients put increasing trust in their therapeutic devices; therefore, it becomes increasingly important to detect malfunctioning affecting such devices. In this work, we explore a new paradigm to detect insulin pump faults (IPFs) that use unsupervised anomaly detection.

METHODS

We generated CL data corrupted with IPFs using the latest version of the T1D Padova/UVA simulator. From the data, we extracted several features capable to describe the patient dynamics and making more apparent suspicious data portions. Then, a feature selection is performed to determine the optimal feature set. Finally, the performance of several popular unsupervised anomaly detection algorithms is analyzed and compared on the identified optimal feature set.

RESULTS

Using the identified optimal configuration, the best performance is obtained by the Histogram-Based Outlier Score (HBOS) algorithm, which detected 87% of the IPF with only 0.08 false positives per day on average. Isolation forest is the best algorithm that offers more conservative performances, detection of 85% of the faults but only 0.06 false positives per day on average.

CONCLUSION

Unsupervised anomaly detection algorithms can be used effectively to detect IPFs and improve the safety of the AP. Future studies will be dedicated to test the presented method inside dedicated clinical trials.

Design and Clinical Evaluation of the Interoperable Artificial Pancreas System (iAPS) Smartphone App: Interoperable Components with Modular Design for Progressive Artificial Pancreas Research and Development

BACKGROUND

There is an unmet need for a modular artificial pancreas (AP) system for clinical trials within the existing regulatory framework to further AP research projects from both academia and industry. We designed, developed, and tested the interoperable artificial pancreas system (iAPS) smartphone app that can interface wirelessly with leading continuous glucose monitors (CGM), insulin pump devices, and decision-making algorithms while running on an unlocked smartphone.

METHODS

After algorithm verification, hazard and mitigation analysis, and complete system verification of iAPS, six adults with type 1 diabetes completed 1 week of sensor-augmented pump (SAP) use followed by 48 h of AP use with the iAPS, a Dexcom G5 CGM, and either a Tandem or Insulet insulin pump in an investigational device exemption study. The AP system was challenged by participants performing extensive walking without exercise announcement to the controller, multiple large meals eaten out at restaurants, two overnight periods, and multiple intentional connectivity interruptions.

RESULTS

Even with these intentional challenges, comparison of the SAP phase with the AP study showed a trend toward improved time in target glucose range 70-180 mg/dL (78.8% vs. 83.1%; P = 0.31), and a statistically significant reduction in time below 70 mg/dL (6.1% vs. 2.2%; P = 0.03). The iAPS system performed reliably and showed robust connectivity with the peripheral devices (99.8% time connected to CGM and 94.3% time in closed loop) while requiring limited user intervention.

CONCLUSIONS

The iAPS system was safe and effective in regulating glucose levels under challenging conditions and is suitable for use in unconstrained environments.