초록 ▼

A method and apparatus that provides an expert system for determining respiratory phase during ventilatory support of a subject. Discrete phase states are partitioned and prior probability functions and observed probability functions for each state are defined. The probability functions are based up

A method and apparatus that provides an expert system for determining respiratory phase during ventilatory support of a subject. Discrete phase states are partitioned and prior probability functions and observed probability functions for each state are defined. The probability functions are based upon relative duration of each state as well as the flow characteristics of each state. These functions are combined to determine phase probabilities for each state using Bayes' theorem. The calculated probabilities for the states may then be compared to determine which state the subject is experiencing. A ventilator may then conform respiratory support in accordance with the most probable phase. To provide a learning feature, the probability functions may be adjusted during use to provide a more subject specific response that accounts for changing respiratory characteristics.

대표청구항 ▼

1. An apparatus for delivering ventilatory support to a subject synchronized to the subject's respiratory phase, comprising: a means for providing a controlled supply of breathable gas to a subject's airway;a means for generating an air flow signal representative of the subject's respiration; anda p

1. An apparatus for delivering ventilatory support to a subject synchronized to the subject's respiratory phase, comprising: a means for providing a controlled supply of breathable gas to a subject's airway;a means for generating an air flow signal representative of the subject's respiration; anda processor to determine a phase of the subject's respiratory cycle and to control the supply of breathable gas delivered to the subject's airway;wherein said processor determines a phase in the subject's respiratory cycle by (a) designating a plurality of phase states within said subject's respiratory cycle; (b) calculating a measure of the subject's respiration from said air flow signal; (c) calculating a rate of change of the air flow signal from said measure of the subject's respiration, (d) determining probabilities for said subject being in each of said phase states based on said rate of change; and (e) identifying the phase in the subject's respiratory cycle from the determined probabilities of said subject's respiratory cycle being in each of said phase states; andwherein said processor causes the delivery of a supply of breathable gas synchronized with the determined phase of the subject's respiratory cycle. 2. The apparatus of claim 1 wherein said means for providing a controlled supply of breathable gas includes a ventilator and said means for generating an air flow signal representative of the subject's respiration includes a transducer. 3. The apparatus of claim 1 wherein said probabilities are determined based on the ratio of each phase state's duration to the duration of an entire respiratory cycle. 4. The apparatus of claim 1 wherein said probabilities are determined based on the ratio of each phase state's duration to the duration of an entire respiratory cycle and also as a function of previously determined probabilities. 5. The apparatus of claim 1 wherein said probabilities are determined using a comparing function that relates said plurality of phase states to sub-ranges selected from a range of measures of a respiratory cycle. 6. The apparatus of claim 1 wherein said comparing function changes as a result of changing respiratory cycles of the subject. 7. The apparatus of claim 1 wherein said processor identifies a phase by additionally comparing determined probabilities to identify a determined probability that is more probable than other determined probabilities. 8. The apparatus of claim 1 wherein said designating a plurality of phase states involves partitioning a respiration cycle in relation to a peak flow. 9. The apparatus of claim 1 wherein said designating a plurality of phase states is repeated to adjust said phase states to account for the subject's changing respiratory characteristics as assessed by a measure of respiration over time. 10. The apparatus of claim 9 wherein said calculating a measure of the subject's respiration includes smoothing data and performing a low pass filtering operation on said data from said air flow signal. 11. The apparatus of claim 1 wherein said determining probabilities involves applying Bayes' theorem to an evaluation of prior probabilities and observed probabilities. 12. The apparatus of claim 1 wherein said determining probabilities includes adjusting a probability based upon state succession. 13. The apparatus of claim 1 wherein said designating a phase state involves calculating a continuous phase variable from said probabilities. 14. A method for delivering ventilatory support to a subject at a pressure synchronized with the subject's respiratory phase, comprising: generating with a sensor an air flow signal representing said subject's air flow;measuring with a processor the subject's respiration from said air flow signal;calculating with the processor a rate of change of the subject's respiration based on the subject's measured respiration;designating with the processor a plurality of phase states within the subject's respiratory cycle;determining with the processor probabilities for said subject's respiratory cycle being in each of said phase states based on the rate of change of the subject's respiration;identifying with the processor a phase in the subject's respiratory cycle from the calculated probabilities of said phase states; anddelivering a supply of breathable gas using a ventilator at a pressure synchronized with the determined phase of the subject's respiratory cycle. 15. The method of claim 14 wherein said air flow signal is generated using a transducer. 16. The method of claim 14 wherein said probabilities are determined based on the ratio of each phase state's duration to the duration of an entire respiratory cycle. 17. The method of claim 14 wherein said probabilities are determined based on the ratio of each phase state's duration to the duration of an entire respiratory cycle and also as a function of previously calculated probabilities. 18. The method of claim 14 wherein said probabilities are determined using a comparing function that relates said plurality of phase states to sub-ranges selected from a range of measures of a respiratory cycle. 19. The method of claim 18 wherein said comparing function changes as a result of changing respiratory cycles of the subject. 20. The method of claim 14 wherein a phase is identified by additionally comparing determined probabilities for one that is more probable than other determined probabilities. 21. The method of claim 14 wherein said designating a plurality of phase states involves partitioning a respiration cycle in relation to a peak flow. 22. The method of claim 14 wherein said designating a plurality of phase states is repeated to adjust said phase states to account for the subject's changing respiratory characteristics as assessed by a measure of respiration over time. 23. The method of claim 22 wherein calculating a measure of the subject's respiration includes smoothing data and performing a low pass filtering operation on said data from said respiration. 24. The method of claim 14 wherein said determining probabilities involves using Bayes' theorem in an evaluation of prior probabilities and observed probabilities. 25. The method of claim 14 wherein said step of determining probabilities includes adjusting a probability based upon state succession. 26. The method of claim 14 wherein said step of identifying a phase involves calculating a continuous phase variable from said determined probabilities.