Method and Apparatus for Prediction and Warning of Hypoglycaemic Attack

• US 20090062678A1
• Filed: 11/21/2005
• Published: 03/05/2009
• Est. Priority Date: 12/20/2004
• Status: Active Grant

• Current Assignee: T&W Engineering A/S (WS Audiology A/S)
• Original Assignee: Hypo-Safe A/S
• Inventors: Beck-Nielsen, Henning
• Granted Patent: US 8,118,741 B2
• Time in Patent Office: Days
• US Class Current: 600/544
• CPC Class Codes: A61B 5/0031 Implanted circuitry
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/374 Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00

Abstract

The present invention includes a method and device for warning of hypoglycaemic attacks, which is a big problem for especially diabetics, who are in insulin treatment. It is estimated that about 1 million people have a decreased quality of life as a result of hypoglycaemic attacks. The invention collects EEG signals from people, who must be supervised and analyses these in order to detect characteristic changes in the EEG signals, which occurs in advance of a hypoglycaemic attack. In a preferred embodiment the EEG signals are collected with subcutaneously placed electrodes and the signals are lead via wires drawn under the person'"'"'s skin to a similar subcutaneously placed signal processing unit. The signal processing analyses the EEG signals and when the pre-seizure characteristics signal changes are detected, a warning signal is given to the person from an alarm giver build into the signal processing unit, e.g. in form of a vibrator. When a person detects a warning signal, the person can prevent the hypoglycaemic attack by e.g. consuming a sugary food.

Claims

1. Method for warning of imminent hypoglycaemic attacks for people such as diebetics characterized in that the warning is based on analysis of one or more EEG signals from the person without the use of ECG signals.

2. Method for warning of hypoglycaemic attacks according to claim 1 characterized in that the warning is given after signal analytic identification of pre-seizure characteristic EEG changes such as decrease in frequency and increase in amplitude.

3. Method for warning of hypoglycaemic attacks according to claim 1 characterized in that the EEG signals are registered from two or more electrodes placed intracranially or extracranially on the skin surface or subcutaneously.

4. Method for warning of hypoglycaemic attacks according to claim 3 characterized in that the EEG signals from the electrodes are transmitted via subcutaneously drawn wires to a subcutaneous placed signal interface unit or signaling processing—

and alarm unit.

5. Method for warning of hypoglycaemic attacks according to claim 4 characterized in that the EEG signals from the signal interface unit communicates with a highly portable signal processing device from which the warning signal can be given.

6. Method for warning of hypoglycaemic attacks according to claim 1 characterized in that the EEG signals in advance of the signal analysis are filtered through e.g. band-pass filtering.

7. Method for warning of hypoglycaemic attacks according to claim 1 characterized in that the signal analysis of the EEG signal is carried out in the time—

or frequency domain.

8. Method for warning of hypoglycaemic attacks according to claim 1 characterized in that the pre-seizure characteristic EEG signal changes are detected by application of signal analysis based on algorithms such as Baye'"'"'s methods including bayesian classifier, logistic regression or neutral network.

9. Device for warning of imminent hypoglycaemic attacks for people such as diabetics characterized in that the device collects the EEG signal from two or more electrodes and by signal analysis identifies pre-seizure characteristic changes in the EEG signal such as a decrease in frequency content and an increase in amplitude and gives a warning signal after identification of the characteristic EEG changes and without the use of ECG signals.

10. Device for warning of hypoglycaemic attacks for people such as diabetics in accordance to claim 9 characterized in that the device is portable and fully or partly is placed subcutaneously.