Histogram based generation of cardiac state detection thresholds

A method for tracking the cardiovascular condition of a human patient, including the steps of:

• (a) sensing an electrical signal from the patient's heart, the electrical signal comprising a multiplicity of beats;
• (b) computing from the electrical signal a heart signal parameter value associated with at least one of the multiplicity of beats;
• (c) computing a heart rate value based on the at least one of the multiplicity of beats;
• (d) selecting one of a plurality of histograms stored in a histogram memory to update according to the heart rate value, wherein each histogram includes a plurality of histogram bins;
• (e) during a data collection time period, incrementing one of the plurality of histogram bins, wherein the selection of the histogram bin to be incremented is dependent on the computed value of the heart signal parameter;
• (f) after the end of the data collection time period, calculating a statistical value of the heart signal parameter from the selected histogram;
• (g) storing the statistical value in memory;
• and
• (h) calculating a detection threshold for abnormal values of the heart signal parameter based on the statistical value.