New Respironics AUTO-BiPap A/H detection explained in patent

Yes, there is a difference in the method Respironics uses to detect apneas and hypopneas - compared to other manufacturers.

The patent for this intellectual invention, filed July 21, 2004 and pub. January 27, 2005, explains how Respironics patent differs from "conventional A/H detection techniques" (their words) and explains how their method works.

Following is an extraction of the general explanation from the patent: Begin Quote

Paragraph 0012: "...convention A/H detection techniques and auto-titration pressure support systems use a myriad of different techniques to detect apneas and hypopneas. One such techniques requires measuring the airflow from the patient and monitoring this airflow to look for reductions during the inspiratory phase indicative of an apnea or hypopnea. This often requires detecting the airflow accurately, which can be difficult in some conditions, for example, if the airflow is being measured via a cannula and the patient is experiencing mouth breathing.

"Conventional A/H detection techniques also typically require distinguishing between the inspiratory and the expiratory states of the patient in order to focus on the changes in patient flow or pressure occurring during the inspiratory state, which is where the apneas/hypopneas occur. Although techniques exist for distinguishing between the inspiratory and the expiratory states of a patient, this remains a complicated task and is subject to errors.

SUMMARY OF THE INVENTIONS

Portions of paragraphs 0013 thru 0016: "Accordingly, it is an object of the present invention to provide an apnea/hypopnea detection system that overcomes the shortcomings of conventional A/H detection techniques. This object is achieved according to one embodiment of the present invention by providing an apnea/hypopnea detection system that includes a flow sensor adapted to detect patient respiratory flow and a processor .

"The processor calculates the following parameters: (a) a long term RMS [root-mean-square] energy based on the flow monitored by the flow sensor, (b) a long term threshold determined based on the long term RMS energy, and (c) a short term RMS energy also based on the output of the flow sensor. The processor determines whether a patient is experiencing an apnea/hypopnea event by comparing the short term RMS energy with the long term threshold.

"By using the RMS energy of the patient's respiratory flow, the present invention avoids the need to distinguish between the inspiratory and expiratory phases of the respiratory cycles, and does not require a highly accurate measurement of the patient's flow in order to provide a relatively accurate A/H detection.

"It is yet another object of the present invention to provide an apnea/hypopnea detection method that odes not suffer from the disadvantages associated with conventional A/H detection techniques....

"It is a further object of the present invention to employ the A/H detection system...in an auto-titration support system...."

End Quote {italics and bold type added by me][/i][/b]


Obviously, there is a lot more to learn from the patent, but I thought a summary would give a general idea of the major differnece between the new Respironics auto-titration method and the methods used by other machine manufacturers.

Someone posted a link to the Respironics patent site a week or so ago. I downloaded from there, but after I printed it, I lost the link. I would post it here for others to look at but now I can't find it.


Andy

Thanks for the info, Andy. Since their algorithm uses an rms measurement it would not be able to distinguish between an actual apnea and very shallow breathing. Either way though, the auto pressure setting would increase, thereby forcing more air into the lungs and achieving the desired result.

Kinda makes me feel better already about my new purchase.

Regards,
Bill

Here is a link to the Pdf's of Respironics Patents (thanks to Derek).

Respironics APAP Patent Application (49 pages/3 Mbytes)

Respironics Apnea/Hypopnea Detection Patent Application (14 pages/1Mbyte)

Not sure how new the basic concept really is (trying to determine patient specific precursors); but they are correct in that I doubt anyone else has patented using "RMS" analysis before.

It looks like to me that the intent is to build a patient specific profile based on analysis of a patients flow patterns - and then determine what the "trigger" is for the patient from that analysis.

It might even work for some. But, I wonder about people who have events that change based on the degree of sleep level they are in. How long will it take the machine to adapt?

This is unlikely to appear in a machine you are currently buying - but may well appear a year or so down the line.

What I find interesting about the whole thing is how much Respironics depends on patenting their "idea" for detections & responses; yet most other companies manage to build great APAPs without a lot of patents on their detection & response methods. Somehow the fact that Respironics patented "how to detect snore" did not stop anyone else from building great snore detectors.

Perry

Perry wrote:

This is unlikely to appear in a machine you are currently buying - but may well appear a year or so down the line.

I wonder if the *new* Respironics Auto BiPap uses this detection method. The patent was filed in July 2004.

Andy

You might find this thread interesting. It eventually gets into the nitty gritty details:

Respironics Bipap Auto First Night Impression