DIACONS — A Consultation System to Assist in the Management of Diabetes

Schneider, Julia Stocker; Piwernetz, K.; Engelbrecht, Rolf; Renner, R

doi:10.1007/978-3-642-48706-4_5

Cited by 11 publications

(7 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During that time period, times and dosages of insulin injections (basal insulin u t,1 and normal insulin u t,2 ), the times and amounts of food intake (fast u t,3 , intermediate u t,4 and slow u t,5 carbohydrates), the times and durations of exercise (regular u t, 6 or intense u t,7 ) and the blood glucose level x t (measured a few times a day) were recorded with 15 minute time resolution. The u t, j , j = 1, .…”

Section: Modelling the Blood Glucosementioning

confidence: 99%

See 2 more Smart Citations

A Nonlinear State Space Model for the Blood Glucose Metabolism of a Diabetic (Ein nichtlineares Zustandsraummodell für den Blutglukosemetabolismus eines Diabetikers)

Briegel¹,

Tresp²

2002

Auto

View full text Add to dashboard Cite

The blood glucose metabolism of a diabetic is a complex nonlinear process closely linked to a number of internal factors which are not easily accessible to measurement. Based on accessible information -such as occasional blood glucose measurements and information about food intake and physical exercise -the system appears highly stochastic and the quantity of main interest, the blood glucose concentration, is very difficult to model and to predict. In this paper we describe a stochastic nonlinear state space model for modeling the blood glucose concentration of a diabetic patient. The model structure is based on physiological prior knowledge and the main nonlinearities are modeled using artificial neural networks. Offline training of the model is performed using a newly developed Monte-Carlo generalized EM (expectation maximization) algorithm. Online prediction is performed using particle filters. Our experimental results show that our approach provides better prediction results than a number of competing approaches.Der Blutglukosemetabolismus eines Diabetikers ist ein komplexer nichtlinearer Prozess, der sehr stark von einer Reihe interner Einflussfaktoren abhängt, welche Messungen nicht leicht zugänglich sind. Basierend auf zugänglicher Information -wie gelegentlicher Blutglukosemessungen und Information über Nahrungsaufnahme und physikalische Betätigung -erscheint das System hochgradig stochastisch und ist daher sehr schwierig zu modellieren. In dieser Veröffentlichung beschreiben wir ein stochastisches nichtlineares Zustandsraummodell zur Modellierung des Blutglukosemetabolismus eines Diabetikers. Die Modellstruktur basiert auf physiologischem Vorwissen, wobei die Hauptnichtlinearitäten durch ein künstliches Neuronales Netz modelliert werden. Offline Training wird mit Hilfe eines neu entwickelten Monte-Carlo EM-Algorithmus (expectation maximization) durchgeführt. Zur Online Vorhersage verwenden wir ein Partikel-Filter. Unsere experimentellen Resultate zeigen, dass unser Ansatz bessere Ergebnisse liefert als eine Vielzahl konkurrierender Verfahren.

show abstract

Section: Modelling the Blood Glucosementioning

confidence: 99%

“…For missing measurement we set σ 2 v → ∞. 6 Equations (12) and (13) can easily be transformed into the state space formulation (1).…”

Section: Trainingmentioning

confidence: 99%

See 1 more Smart Citation

A Nonlinear State Space Model for the Blood Glucose Metabolism of a Diabetic (Ein nichtlineares Zustandsraummodell für den Blutglukosemetabolismus eines Diabetikers)

Briegel¹,

Tresp²

2002

Auto

View full text Add to dashboard Cite

show abstract

“…The basis functions are implemented as Gaussians of the form (9) We derived the widths and centers of the basis functions from the time windows used in the previous section (Section III-C.1). The overlapping basis functions "compete" for the input leading to the normalized weighting function (10) The input to the TSCNN at time is obtained by weighting the amplitudes of the past (raw) input time series at time by over a certain time window (11) where is the beginning, is the end 4 of the th soft time window for grouped input Fig.…”

Section: B Soft Competitive Fixed-time Windows (Tscnn1-sc)mentioning

confidence: 99%

Neural-network models for the blood glucose metabolism of a diabetic

Tresp

Briegel²,

Moody

1999

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

We study the application of neural networks to modeling the blood glucose metabolism of a diabetic. In particular we consider recurrent neural networks and time series convolution neural networks which we compare to linear models and to nonlinear compartment models. We include a linear error model to take into account the uncertainty in the system and for handling missing blood glucose observations. Our results indicate that best performance can be achieved by the combination of the recurrent neural network and the linear error model.

show abstract

“…Rodbard's initial program has evolved into three closely related programs (SUGAR3, M A R T K l and INSULIN) [44, 1041. These provide various kinds of data analyses and displays including mean or median blood glucose levels, lowest and highest values, 25th and 75th percentiles, average number of measurements per day, number of hypoglycaemic episodes, and the variability of these findings (via histograms and standard deviations for the blood glucose readings). Another program, the DIACONS system, aims to assist medical professionals and patients alike [105]. It includes a medical documentation module called DlADOC which performs watchdog functions (for example, recognizing trends or warning of too high or too low blood glucose levels).…”

Section: Computers As Data Viewers and Interpretersmentioning

confidence: 99%

Application of computers in diabetes carea review. I. Computers for data collection and interpretation

Lehmann

Deutsch

1995

Medical Informatics

View full text Add to dashboard Cite

The paper reviews the contribution of information technology (IT) to diabetes care. An appraisal of this topic with respect to insulin-dependent (type 1) diabetic patients is carried out in view of the landmark findings of the Diabetes Control and Complications Trial (DCCT) which has demonstrated that maintaining tight blood glucose control can delay the onset and slow the progression of the later life complications of diabetes. The review starts with the clinical background and the main features of the control schema in which diabetic patients receive insulin therapy. An overview is then provided of recent IT initiatives in diabetes care, and the application of IT techniques to assist in the diagnosis and characterization of patients with diabetes mellitus is considered. The role of IT approaches for short-term glycaemic control is discussed and the utilization of computers for collecting, viewing and interpreting home monitoring blood glucose data is reviewed; both quantitative and qualitative techniques being considered. In the second paper the role of decision support tools for planning insulin therapy using clinical algorithms, hand-held devices, knowledge-based approaches, telemedicine techniques and interactive simulations is reviewed, and the validation and clinical evaluation of these tools is discussed. The likely impact of the routine clinical application of implantable/non-invasive blood glucose monitoring devices is also considered. Finally, the application of computers as teaching tools is reviewed and the ways in which such educational approaches might be applied for disseminating the benefits of the DCCT trial more widely are discussed.

show abstract

DIACONS — A Consultation System to Assist in the Management of Diabetes

Cited by 11 publications

References 1 publication

A Nonlinear State Space Model for the Blood Glucose Metabolism of a Diabetic (Ein nichtlineares Zustandsraummodell für den Blutglukosemetabolismus eines Diabetikers)

A Nonlinear State Space Model for the Blood Glucose Metabolism of a Diabetic (Ein nichtlineares Zustandsraummodell für den Blutglukosemetabolismus eines Diabetikers)

Neural-network models for the blood glucose metabolism of a diabetic

Application of computers in diabetes carea review. I. Computers for data collection and interpretation

Contact Info

Product

Resources

About