Physiological trend monitor
View Patent ↗A physiological trend monitor has a sensor signal responsive to multiple wavelengths of light transmitted into a tissue site. The transmitted light is detected after attenuation by pulsatile blood flow within the tissue site. A processor has an input responsive to the sensor signal and a physiological parameter output. Features are extracted from the physiological parameter output. Criteria are applied to the features. An alarm output is generated when the criteria are satisfied.
1. An oximeter, comprising:
a plurality of processing modules, each responsive to a noninvasive sensor signal indicative of light attenuated by body tissue, at least one processing module having different characteristics from another processing module, each of said plurality of processing modules outputting a parallel measurement,
an alarm;
a memory storing criteria indicative of alarm conditions, said criteria including at least a duration of at least one of said parallel measurements crossing a threshold;
an alarm indicator receiving said parallel measurements from said plurality of processing modules, extracting features from said parallel measurements, comparing said features to said criteria, and when said features match said criteria, triggering said alarm on said oximeter.
2. The oximeter according to claim 1 wherein said criteria including at least a trend of said parallel measurements and a behavior of a slope of said trend.
3. The oximeter according to claim 1 wherein said criteria including at least a pattern.
4. The oximeter according to claim 1 wherein said criteria including at least a number of times said measurements crossed said threshold.
5. The oximeter according to claim 1 wherein at least one of said parallel measurements comprises a relatively slow measurement and at least one of said parallel measurements comprises a relatively fast measurement.
6. The oximeter according to claim 5 , wherein said alarm indicator extracts threshold features from said slow measurement and at least one waveform feature from said fast measurement and said alarm indicator triggers said alarm based on a combination of said threshold and said waveform features.
7. The oximeter according to claim 5 , wherein said alarm indicator extracts threshold features from said slow measurement and triggers said alarm based on said threshold features, said alarm indicator also extracting pattern features from said fast measurement and triggers said alarm based on said pattern features.
8. A method of determining when to alert a caregiver during oximeter-based patient monitoring, the method comprising:
providing a plurality of electronic processing modules, at least one processing module having different characteristics from another processing module;
providing an alarm;
providing a memory;
providing an alarm indicator;
receiving a noninvasive sensor signal indicative of light attenuated by body tissue;
electronically processing said sensor signal with said plurality of processing modules;
outputting a parallel measurement from each of said plurality of processing modules;
storing criteria indicative of alarm conditions, said criteria including at least a duration of at least one of said parallel measurements crossing a threshold;
extracting features from said parallel measurements;
comparing said features to said criteria; and
when said features match said criteria, triggering said alarm.
9. The method according to claim 8 wherein storing said criteria comprises storing criteria including at least a trend of said parallel measurements and a behavior of a slope of said trend.
10. The method according to claim 8 wherein storing said criteria comprises storing criteria including at least a pattern.
11. The method according to claim 8 wherein storing said criteria comprises storing criteria including at least a number of times said measurements crossed said threshold.
12. The method according to claim 8 wherein outputting said parallel measurement includes outputting a relatively slow measurement and outputting a relatively fast measurement.
13. The method according to claim 12 , wherein extracting features includes extracting threshold features from said slow measurement and extracting at least one waveform feature from said fast measurement, and wherein said triggering said alarm includes triggering said alarm based on a combination of said threshold and said waveform features.
14. The method according to claim 12 , wherein extracting features includes extracting threshold features from said slow measurement and said triggering said alarm includes triggering said alarm based on said threshold features, and wherein extracting features includes extracting pattern features from said fast measurement and said triggering said alarm includes triggering said alarm based on said pattern features.