Diabetes, Diabetes Mellitus, Type 1
Conditions
Keywords
Automotive Technology, Hypoglycemia, Hypoglycaemia, Driving, Car
Brief summary
To analyse driving behavior of individuals with type 1 diabetes in eu- and mild hypoglycaemia while driving in a real car. Based on the in-vehicle variables, the investigators aim at establishing algorithms capable of discriminating eu- and hypoglycaemic driving patterns using machine learning classifiers.
Detailed description
Hypoglycaemia is among the most relevant acute complications of diabetes mellitus. During hypoglycaemia physical, psychomotor, executive and cognitive function significantly deteriorate. These are important prerequisites for safe driving. Accordingly, hypoglycaemia has consistently been shown to be associated with an increased risk of driving accidents and is, therefore, regarded as one of the relevant factors in traffic safety. Therefore, this study aims at evaluating a machine-learning based approach using in-vehicle data to detect hypoglycaemia during driving. During controlled eu- and hypoglycaemia, participants with type 1 diabetes mellitus drive in a driving school car on a closed test-track while in-vehicle data is recorded. Based on this data, the investigators aim at building machine learning classifiers to detect hypoglycemia during driving.
Interventions
Participants will drive on a designated circuit with a real car on a test track accompanied by a driving instructor. Initially, a euglycaemic state (5.0 - 8.0 mmol/L) is established and blood glucose is then declined to hypoglycaemia (3.0 - 3.5 mmol/L) by administering insulin. Thereafter, blood glucose is raised again to euglycaemia (5.0 - 8.0mmol/L). During the procedure, driving data is recorded. Additionally, eye movement, head pose, facial expression, heart rate, skin conductance, and CGM values are recorded throughout the glycemic trajectory. Participants are blinded to the blood glucose values during the procedure.
Sponsors
Swiss Federal Institute of Technology
University of St.Gallen
Insel Gruppe AG, University Hospital Bern
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
OTHER
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
21 Years to 60 Years
Healthy volunteers
No
Inclusion criteria
* Informed consent as documented by signature * Type 1 Diabetes mellitus as defined by WHO for at least 1 year or confirmed C-peptide negative (\<100pmol/l with concomitant blood glucose \>4 mmol/l) * Age between 21-60 years * HbA1c ≤ 9.0 % * Functional insulin treatment with good knowledge of insulin self-management * Passed driver's examination at least 3 years before study inclusion. Possession of a valid, definitive Swiss driver's license. * Active driving in the last 6 months.
Exclusion criteria
* Contraindications to the drug used to induce hypoglycaemia (insulin aspart), known hypersensitivity or allergy to the adhesive patch used to attach the glucose sensor. * Pregnancy or intention to become pregnant during the course of the study, lactating women or lack of safe contraception * Other clinically significant concomitant disease states as judged by the investigator * Physical or psychological disease likely to interfere with the normal conduct of the study and interpretation of the study results as judged by the investigator * Renal failure * Hepatic dysfunction * Coronary heart disease * Other cardiovascular disease * Epilepsy * Drug or alcohol abuse * Inability to follow the procedures of the study, e.g. due to language problems, psychological disorders, dementia, etc. of the participant * Participation in another study with an investigational drug within the 30 days preceding and during the present study * Total daily insulin dose \>2 IU/kg/day * Specific concomitant therapy washout requirements prior to and/or during study participation * Current treatment with drugs known to interfere with metabolism or driving performance
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Diagnostic accuracy of the hypoglycaemia warning system using in-vehicle data to detect hypoglycaemia quantified as the area under the receiver operating characteristics curve (AUROC). | 240 minutes | The machine learning model is developed and evaluated based on in-vehicle data generated in eu- and hypoglycaemia. Detection performance of hypoglycaemia is quantified as AUROC. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Diagnostic accuracy of the hypoglycaemia warning system using in-vehicle data and recordings of the continous glucose monitoring (CGM) system to detect hypoglycaemia quantified as sensitivity and specificity. | 240 minutes | The CGM device is in use during controlled eu- and hypoglycaemia. Detection performance of hypoglycaemia is quantified as sensitivity and specificity. |
| Diagnostic accuracy of the hypoglycaemia warning system using wearable data and recordings of the CGM system to detect hypoglycaemia quantified as sensitivity and specificity. | 240 minutes | The CGM device is in use during controlled eu- and hypoglycaemia. Detection performance of hypoglycaemia is quantified as sensitivity and specificity. |
| Change in driving features over the glycaemic trajectory. | 240 minutes | Driving signals are recorded using a driving simulator. |
| Change of gaze coordinates over the glycaemic trajectory. | 240 minutes | Gaze coordinates are recorded using an eye-tracker device. |
| Change of head pose over the glycaemic trajectory. | 240 minutes | Head pose (position/rotation) is recorded using an eye-tracker device. |
| Change of heart rate over the glycaemic trajectory | 240 minutes | Heart rate is recorded using a holter-ECG device and a wearable. |
| Change of heart rate variability over the glycaemic trajectory | 240 minutes | Heart rate variability is recorded using a holter-ECG device and a wearable. |
| Change of electrodermal activity over the glycaemic trajectory | 240 minutes | Electrodermal activity is recorded using a wearable. |
| Hypoglycaemic symptoms over the glycaemic trajectory. | 240 minutes | Hypoglycemic symptoms are rated using a validated questionnaire (minimum score = 0, maximum score = 6, a higher score means more symptoms) |
| Diagnostic accuracy of the hypoglycaemia warning system using wearable data to detect hypoglycaemia quantified as the area under the receiver operating characteristics curve (AUROC). | 240 minutes | The machine learning model is developed and evaluated based on wearable data recorded in eu- and hypoglycaemia. Detection performance of hypoglycemia is quantified as AUROC. |
| Time course of the hormonal response over the glycaemic trajectory | 240 minutes | Epinephrine, norepinephrine, glucagon, cortisol and growth hormone will be measured at pre-defined time points. |
| Self assessment of driving performance over the glycaemic trajectory. | 240 minutes | Participants rate their driving performance on a 7-point Likert Scale (lower value means poorer driving performance). |
| Number of driving mishaps over the glycaemic trajectory. | 240 minutes | Any driving mishaps, accidents and interventions by the driving instructor will be documented. |
| CGM accuracy over the glycaemic trajectory | 240 minutes | CGM values will be recorded using a CGM sensor. Venous blood glucose is considered as the reference. Accuracy will be quantified using mean absolute relative difference (MARD) from the gold-standard and using the Clarke error grid. |
| Accuracy of our protocol to induce hypoglycaemia in achieving the intended hypoglycaemic range. | 240 minutes | Accuracy will be quantified using mean absolute relative difference from the intended hypoglycaemic range. |
| Number of Adverse Events (AEs) | 2 weeks, from screening to close out visit in each participant | Adverse Events will be recorded at each study visit. |
| Number of Serious Adverse Events (SAEs) | 2 weeks, from screening to close out visit in each participant | Serious Adverse Events will be recorded at each study visit. |
| Emotional response to the hypoglycaemia warning system | 240 minutes | Physiological response will be measured using an electro-dermal activity sensor (skin conductance) and eye tracker (eye blinks). Self-reported emotional response will be assessed with scales (e.g., valence, arousal, annoyance, sense of urgency). |
| Technology acceptance of the hypoglycaemia warning system | 240 minutes | Technology acceptance will be measures with user experience questionnaires, such as the Unified Technology Acceptance and Use of Technology Questionnaire and free words associations. |
| Change of cognitive performance over the glycaemic trajectory. | 240 minutes | Cognitive performance will be assessed using the Trail Making B Test (lower time in seconds means better performance) and using the Digital Symbol Substitution Test (higher score means better performance). |
Countries
Switzerland
Outcome results
None listed