Signal analysis method
Improvement in the reliability of segmentation of a signal, such as an ECG signal, is achieved through the use of duration constraints. The signal is analysed using a hidden Markov model. The duration constraints specify minimum allowed durations for specific states of the model. The duration constraints can be incorporated either in the model itself or in a Viterbi algorithm used to compute the most probable state sequence given a conventional model. The derivation of a confidence measure from the model can be used to assess the quality and robustness of the segmentation and to identify any signals for which the segmentation is unreliable, for example due to the presence of noise or abnormality in the signal.
1. A computer-implemented method for analysing a signal which has been segmented using a probabilistic segmentation model, the method comprising:
calculating a confidence measure for each of a plurality of segmented signal features;
plotting each calculated confidence measure against a length of its respective signal feature;
applying density modelling techniques to the plot to determine a suitable region associated with high confidence measures;
determining whether a confidence measure for a specific signal feature falls outside this region; and
wherein the signal is a bio-medical signal selected from the group consisting of an electrocardiogram signal, an electroencephalogram signal, an electromyogram signal, a cardiac electrophysiology signal, and an endocardial electrogram.
2. A method according to claim 1 , wherein each of the plurality of segmented signal features comprise a portion of a longer signal.
3. A method according to claim 1 , wherein the confidence measure is one selected from the group consisting of: the log likelihood of the signal feature given the model; and the joint log likelihood of the signal feature and the most probably state sequence given the model.
4. A method according to claim 1 , further comprising displaying the plot of each calculated confidence measure against the length of its respective signal feature.
5. A method according to claim 1 , wherein the signal is an electrocardiogram signal and information on a QT interval is obtained from the probabilistic segmentation model.
6. A non-transitory computer-readable medium storing a computer program capable of being executed on a computer system and capable when so executed of causing the computer system to perform a method according to claim 1 .