AU2015204360B2 - Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea - Google Patents

Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea Download PDF

Info

Publication number
AU2015204360B2
AU2015204360B2 AU2015204360A AU2015204360A AU2015204360B2 AU 2015204360 B2 AU2015204360 B2 AU 2015204360B2 AU 2015204360 A AU2015204360 A AU 2015204360A AU 2015204360 A AU2015204360 A AU 2015204360A AU 2015204360 B2 AU2015204360 B2 AU 2015204360B2
Authority
AU
Australia
Prior art keywords
flow
wave
inspiratory
time
apparatus
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
AU2015204360A
Other versions
AU2015204360A1 (en
Inventor
Jeffrey Peter Armitstead
Chinmayee Somaiya
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Resmed Pty Ltd
Original Assignee
ResMed Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US60/779,577 priority Critical
Priority to AU2007222882A priority patent/AU2007222882B2/en
Priority to AU2012258376A priority patent/AU2012258376B2/en
Application filed by ResMed Ltd filed Critical ResMed Ltd
Priority to AU2015204360A priority patent/AU2015204360B2/en
Publication of AU2015204360A1 publication Critical patent/AU2015204360A1/en
Application granted granted Critical
Publication of AU2015204360B2 publication Critical patent/AU2015204360B2/en
Assigned to ResMed Pty Ltd reassignment ResMed Pty Ltd Request to Amend Deed and Register Assignors: RESMED LIMITED
Application status is Active legal-status Critical
Anticipated expiration legal-status Critical

Links

Abstract

Abstract In a respiratory apparatus for treatment of sleep apnea and other disorders associated with an obstruction of a patient's airway and which uses an airflow 5 signal, an obstruction index is generated which detects the flattening of the inspiratory portion of the airflow. The flattening index serves as an obstruction index used to differentiate normal and obstructed breathing. The obstruction index is based upon comparison of values of airflow in different sectors of the inspiratory peak of the wave function and is particularly suitable for distinguishing 10 M shaped or square shaped respiratory patters indicative of partially obstructed airways. w -0.5 002 0.4 0.6 118 Fig. 7 1A-- --- _ I 0.002 0.4 0.6 0.8 1 Fig. 8

Description

Method and Apparatus for Improved Flow Limitation Detection Of 2015204360 16M2015

Obstructive Sleep Apnea

FIELD OF THE INVENTION 5 This invention relates to an improved method and apparatus for the diagnosis and treatment of respiratory conditions including sleep apnea or hypopnea. In particular it relates to detecting either partial or complete obstruction of the airway of a patient, based upon an analysis of the time development of an inspiratory portion of a respiratory airflow curve. It also relates to apparatus that 10 adjust treatment pressure to be applied by a CPAP ventilator.

BACKGROUND OF THE INVENTION

Obstructive Sleep Apnea (OSA), a syndrome that includes apnea, hypopnea and heavy snoring, causes sleep disruption that brings about serious health problems possibly including types of heart disease. OSA is caused by the collapse of 15 portions of a person’s airway passage. The treatment of choice for OSA is the administration of Continuous Positive Airway Pressure (CPAP) to keep the patient’s airway open. The air, usually in the pressure range 4-20 cm H20, is supplied by a motor driven blower whose output passes via an air delivery device to sealingly engage a patient's airway. A mask, tracheotomy tube, endotracheal 20 tube, nasal pillows or other appropriate device may be used. An exhaust port is provided in a delivery tube proximate to the air delivery device. Some CPAP devices, termed bi-level CPAP, sense the breathing cycle of inspiration and expiration and provide different positive pressure levels during inhaling and exhaling. Some self-titrating CPAP devices are designed to determine 25 appropriate pressure levels for the individual patient by selecting the least pressure that resolves the OSA. In such devices patterns of respiratory parameters are monitored to determine when OSA is present, and CPAP pressure is raised on the detection of appropriate patterns to provide increased 1 airway pressure to, ideally, subvert the occurrence of obstructive episodes and the other forms of breathing disorder. Such devices are described in U.S. patents 5,148,802 and 5,245,995. 2015204360 16 Jul 2015

Typically a person suffering from OSA is diagnosed and treated in a sleep 5 laboratory where the presence of the ailment is confirmed during a first night sleeping session and the appropriate treatment pressure is determined during a second night sleeping session. One problem that arises is that the appropriate pressure varies during the night as the person goes through different stages of sleep. Therefore there has been a long felt need for simplified apparatus for use 10 in a patient’s own home that could determine the presence of OSA and modify the pressure level to an optimum pressure. There have been several attempts, with varying success, to determine the presence of OSA from the analysis of the shape of airflow curves as a function of time.

The monitoring of upper airway pressure-flow relationships in obstructive sleep 15 apnea has been described in Smith et al., 1988, J. Appl Physiol. 64: 789-795. FIG. 1 of that article shows polygraphic sleep recordings at varying levels of increasing nasal pressure. It was noted that inspiratory volumetric flow plateaued in certain breaths suggesting the presence of airflow limitation. Pressure-flow curves were constructed by plotting mid-inspiratory airflow against 20 either mask pressure or endoesophageal pressure. The pressure-flow plots of nasal pressure against mean midinspiratory flow were then fit by least-squares linear regression to calculate resistance upstream to the collapsible site.

The effect of positive nasal pressure on upper airway pressure-flow relationships has been described in Schwartz et al., 1989, J. Appl Physiol. 66:1626-1634. 25 FIG. 4 of the article shows that pressure-flow tracings plateau at a low pressure level. It was further shown when the pressure was increased, flow did not plateau. U.S. Pat. No. 5,335,654 (Rapoport) shows the effect of CPAP on the airflow versus time curve for a patient suffering from OSA. Figures 1-5 in Rapoport 2 show that as the pressure is reduced from 10 cm H2O to 2 cm H2O in steps of 2 cm H20 the curve of airflow versus time changes from an almost smooth sinusoidal pattern to one with a flattening of the inspiration portion of the curve with initial and terminal flow spikes. At 2 cm H2O the curve has developed a so-5 called M shape (i.e. with a ripple in the middle) and has also developed 2015204360 16 Μ 2015 overshoots (i.e. peaks) at each end of the flattened central region. In an attempt to characterize the flow shapes that indicate obstruction, Rapoport lists several indices said to be indications of flow limitation and/or partial obstruction patterns including: (1) The derivative of the flow signal equals zero; (2) The second 10 derivative between peaks of the flow signal is zero for a prolonged interval; (3) The ratio of early inspirational flow to mid-inspirational flow is less than or equal to 1. The patent further lists events said to be indications of obstructions: (1) Reduced slope of the line connecting the peak inspiratory flow to the peak expiratory flow; (2) Steep upward or downward stroke (dV/dt) of the flow signal; 15 and (3) Ratio of inspiratory flow to expiratory flow over 0.5.

With regard to the control of CPAP treatment, various techniques are known for sensing and detecting abnormal breathing patterns indicative of obstruction. For example, U.S. Pat. No. 5,245,995 (Sullivan et al.) describes how snoring and abnormal breathing patterns can be detected by inspiration and expiration 20 pressure measurements while sleeping, thereby leading to early indication of preobstructive episodes or other forms of breathing disorder. Particularly, patterns of respiratory parameters are monitored, and CPAP pressure is raised on the detection of pre-defined patterns to provide increased airway pressure to ideally prevent the occurrence of the obstructive episodes and the other forms of 25 breathing disorder. U.S. Pat. No. 5,645,053 (Remmers) describes calculating a flatness index, wherein flatness is defined to be the relative deviation of the observed airflow from the mean airflow. In Remmers, individual values of airflow are obtained between 40% and 80% of the inspiratory period. The mean value is calculated 30 and subtracted from individual values of inspiratory flow. The individual 3 differences are squared and divided by the total number of observations minus one. The square root of this result is used to determine a relative variation. The relative variation is divided by the mean inspiratory airflow to give a relative deviation or a coefficient of variation for that breath. 2015204360 16 Μ 2015 5 U.S. Pat. No. 5,704,345 (Berthon-Jones) disclosed a method for detecting partial obstruction of a patient's airway by calculating two obstruction index values that parameterize a flattening of an inspiratory portion of a patient's monitored respiratory airflow. Either obstruction index is then compared to a threshold value. The first shape factor involves a ratio of the mean of a mid-portion of the 10 inspiratory airflow of the breathing cycle and the mean of the inspiratory airflow. l 48 ~ Σ fi(t) shapeX = —-

M where fs(t) is a sample of the patient inspiratory airflow and M is the mean of 1 65 inspiratory airflow given by M= — Σ M0 · 65 ,=1

The second shape factor involves a ratio of the Root Mean Square deviation of a 15 mid-portion of inspiratory airflow and the mean inspiratory airflow according to the formula __ ιϊίΣο&ίΟ-Μ)1 shape 2 = -—-.

M U.S. patent 6,814,073 (Wickham) discloses a method and apparatus for detecting some forms of obstruction based upon the inspiratory airflow. In this 20 method, inspiratory airflow samples corresponding to mid-inspiration are identified. In one embodiment, weighting factors are applied based on whether the inspiratory airflow samples are less than or greater than a threshold level, such as the mean airflow. In another embodiment, different weighting factors are 4 applied to samples based on their time positions in a breath. Samples taken prior to a certain event during inspiration, for example samples preceding the half way point of inspiration, are assigned lower weighting factors than samples succeeding the event. An obstruction index is then calculated using these 5 samples with their corresponding weighting factors. 2015204360 16 Μ 2015

All these disclosed techniques fail to detect flow limitation in certain types of flow patterns, particularly the M-wave pattern. More generally, flow limitation in the inspiratory patterns that present with leading or lagging overshoot, is not appropriately detected. The detectability depends on two factors. 1) the extent to 10 which the overshoot traverses the mid-portion of the inspiratory airflow, and 2) the size of the overshoot. The solution presented by Wickham, works satisfactorily if the size of the overshoot is relatively small, does not span a large portion of mid-inspiration, and is not present in the latter half of the inspiratory airflow. However, flattening indices from the Wickham method are sometimes 15 not as accurate as the indices from the method disclosed by Berthon-Jones.

One aspect of the present invention is to simplify the algorithm mentioned in US patent 5,704,345 (Berthon-Jones), the disclosure of which is incorporated by reference. Thus, an objective of the current invention is to present a method by which flow-limitation can be detected in the presence of the mentioned 20 limitations, particularly the M wave pattern with overshoots.

The flow limitation detection/estimation technique described by the prior art is also expensive in terms of digital processing power, and the accuracy of flow detection required. Therefore its utility in low cost electronic/software platforms is 25 limited. Therefore another objective of the present invention is to simplify the algorithm to render it more amenable to low-end electronic and software platforms. 5

BRIEF DESCRIPTION OF THE INVENTION 2015204360 16 Μ 2015

The present invention involves an improved method and apparatus for detecting some forms of respiration obstruction based upon the flattening of the inspiratory airflow. 5 One aspect of the invention is a method for detecting partial obstruction of the airway of a patient, the method comprising the steps of Measuring respiratory air flow from the patient; Detecting the inspiratory part of that airflow; Trimming the inspiratory part to remove any expiratory pause; Normalizing the effect of offshoot; Scaling the inspiratory flow to a reference value; Calculating the mean 10 deviation of a central portion of the inspiratory air flow; Calculating a flattening index from the mean deviation; and Filtering the flattening index with a moving average filter

In connection with the method for detecting partial obstructions, a normalizing correction is made for overshoots that would otherwise exaggerate the 15 roundness estimate of the inspiratory airflow.

In an embodiment of the invention the normalization of the overshoot is accomplished by truncating airflow values outside a central portion of the inspiratory airflow that exceed the average value of a central portion of the inspiratory airflow. 20 In another embodiment of the invention, the overshoot is evaluated by dividing the inspiratory airflow wave into multiple parts, calculating mean amplitudes for each part, analyzing the mean amplitudes to determine the presence of overshoot, and, if overshot is indicated, treating flattening as low.

In a further embodiment of the invention, a hunt for maximal flatness is carried 25 out by calculating a set of flattening indices for each possible sample in a middle part of the inspiratory wave form and taking the flattening index for the inspiratory waveform to be the minimum of the calculated indices. 6

Hence it is an aspect of the invention to estimate the extent of flow limitation in patient breath by the extent to which the position of peak flow deviates from the center of the inspiration flow curve. 2015204360 16 Μ 2015 5 It is a further aspect of the invention to estimate the flow limitation by the extent to which the relative mean amplitude of different parts of the breath deviate from the pattern exhibited by normal breaths.

It is a still further aspect of the invention to provide a method to estimate the 10 extent of flow limitation by hunting for the part of the breath that exhibits maximal flatness.

It is an objective of the present invention to provide an apparatus in which obstruction, either partial or complete, of the patient's airway is detected by analyzing respiratory airflow by a method for detecting partial obstruction of the 15 airway of a patient, the method comprising the steps of Measuring respiratory air flow from the patient; Detecting the inspiratory part of that airflow; Trimming the inspiratory part to remove any expiratory pause; Normalizing the effect of offshoot; Scaling the inspiratory flow to a reference value; Calculating the mean deviation of a central portion of the inspiratory airflow; Calculating a flattening 20 index from the mean deviation; and Filtering the flattening index with a moving average filter. A further objective is to provide an apparatus in which an improved algorithm for detecting airway obstruction is implemented without using additional components or making substantial changes to the structure of existing respiratory apparatus, 25 Accordingly, a respiratory apparatus is provided in which the respiratory airflow of a patient is continuously monitored and the part of respiratory airflow associated with inspiration is identified and sampled. 7

In one aspect, the subject invention pertains to a respiratory apparatus which includes an air source adapted to selectively provide pressurized breathable air to a patient, a flow detecting device to sense the respiratory airflow from the patient and to generate an airflow signal indicative of airflow, an obstruction 5 detector coupled to said flow sensor which implements a method for detecting partial obstruction of the airway of a patient, the method comprising the steps of Measuring respiratory air flow from the patient; Detecting the inspiratory part of that airflow; Trimming the inspiratory part to remove any expiratory pause; Normalizing the effect of offshoot; Scaling the inspiratory flow to a reference 10 value; Calculating the mean deviation of a central portion of the inspiratory air flow; Calculating a flattening index from the mean deviation; and Filtering the flattening index with a moving average filter, and a pressure controller connected to a pressure sensor equivalent - either physically or logically arranged to control the operation of the air source, receive the obstruction signal and alter the 15 operation of the air source in response to the obstruction signal. The term flow detecting device is intended to be a general term that includes flow sensors or alternative flow detecting devices or algorithms, such as, for example, those determining flow from a motor current. 2015204360 16 Μ 2015

Another aspect of the invention concerns an apparatus for monitoring and/or 20 treating a patient having a sleep disorder, the apparatus including a flow detecting device that senses patient respiration and generates a corresponding flow signal; and an obstruction detector coupled to the flow detecting device and adapted to implement a method for detecting partial obstruction of the airway of a patient, the method comprising the steps of Measuring respiratory air flow from 25 the patient; Detecting the inspiratory part of that airflow; Trimming the inspiratory part to remove any expiratory pause; Normalizing the effect of offshoot; Scaling the inspiratory flow to a reference value; Calculating the mean deviation of a central portion of the inspiratory air flow; Calculating a flattening index from the mean deviation; and Filtering the flattening index with a moving average filter, 30 wherein the obstruction detector including a signal generator that generates a signal indicative of an airway obstruction based on the flattening index. 8 A further aspect of the invention concerns an apparatus for treating a patient having a sleep disorder, the apparatus comprising a mask, a gas source selectively supplying pressurized breathable air to the patient through the mask, a flow detecting device that senses airflow and generates a flow signal indicative 5 of respiration, an obstruction detector coupled to the flow detecting device and adapted to implement a method for detecting partial obstruction of the airway of a patient, the method comprising the steps of Measuring respiratory air flow from the patient; Detecting the inspiratory part of that airflow; Trimming the inspiratory part to remove any expiratory pause, Normalizing the effect of offshoot, Scaling 10 the inspiratory flow to a reference value, Calculating the mean deviation of a central portion of the inspiratory airflow, Calculating a flattening index from the mean deviation, and Filtering the flattening index with a moving average filter, and a controller receiving the obstruction signal and generating in response a command for activating the gas source. 2015204360 16 Μ 2015 15 Another aspect of the invention concerns a method for detecting obstruction in the airways of a patient, including measuring an airflow of the patient, detecting a predetermined section of said air flow, implementing a method for detecting partial obstruction of the airway of a patient, the method comprising the steps of Measuring respiratory air flow from the patient; Detecting the inspiratory part of 20 that airflow; Trimming the inspiratory part to remove any expiratory pause; Normalizing the effect of offshoot; Scaling the inspiratory flow to a reference value; Calculating the mean deviation of a central portion of the inspiratory air flow; Calculating a flattening index from the mean deviation; and Filtering the flattening index with a moving average filter.

25 BRIEF DESCRIPTION OF DRAWINGS FIG. 1 shows a block diagram of a respiratory apparatus constructed in accordance with this invention; FIG. 2 shows a flow chart illustrating the operation of the apparatus of FIG. 1; 9

FiG. 3 shows the inspiration phases of typical respiration signals for a healthy person and a person with a partial airway obstruction; 2015204360 16 Μ 2015

Fig. 4 shows a normal respiratory cycle.

Fig. 5 shows normal and flow limited flow vs. time curves. The left hand panel 5 shows a normal curve. The right hand panels show flow limited curves.

Fig. 6 shows inspiratory flow vs. time for two possible breaths.

Fig. 7 is a graph of basis vectors for use in calculating an M shape index.

Fig. 8 is a graph of a typical M shaped breath.

DETAILED DESCRIPTION OF THE INVENTION 10 Apparatus and Methodology FIG. 1 shows an example respiratory apparatus 10 constructed in accordance with the invention. The respiratory apparatus 10 includes a mask 12 connected to a blower 14 by a flexible tube 16. The mask 12 is fitted to the patient and may be either a nose mask or a facemask. The blower 14 has an air outlet 22 is 15 driven by a motor 18 in accordance with control signals from a servocontroller 20. This arrangement allows the respiratory apparatus 10 to deliver pressurized air (or air enriched with oxygen from a source, not shown). The pressurized air is delivered by tube 16 to the mask 12. The tube 16 is provided with a narrow exhaust port 26 through which air exhaled by the patient is expelled. 20 A control circuit 24 is used to control the operation of servocontroller 20 and motor 18 using certain predetermined criteria, thereby defining modes of operation for the apparatus 10. Preferably, in accordance with this invention, the control circuit 24 is adapted to operate the apparatus 10 to provide CPAP to the patient. 10

Control circuit 24 includes a flow restrictive element 28. Tubes 30 and 31 lead from different sides of the restrictive element 28 to a mask pressure transducer 32 and a differential pressure transducer 34 respectively. The mask pressure transducer 32 is also connected through another tube 33 to the differential 5 pressure transducer 34. 2015204360 16M2015

The mask pressure transducer 32 generates a first electrical signal, which is amplified by an amplifier 36 to generate an output P(t) proportional to the air pressure within the mask 12. This output is fed directly to the servocontroller 20.

The differential pressure transducer 34 senses the differential pressure across 10 the flow restrictive element 28, which differential pressure is related to the airflow rate through the flow restrictive element 28 and tube 16. Differential pressure transducer 34 generates a second electrical signal that is amplified by an amplifier 38. This amplified signal F(t) is termed an airflow signal since it represents the airflow through the tube 16. Alternatively, the flow may be derived 15 from the motor current.

The airflow signal F(t) is fed to a filter 40 that filters the signal within a preset range. The outputs of the filter 40 and amplifier 36 are fed to an ADC (analog-to-digital) converter 42, which generates corresponding signals to a microprocessor 44. The microprocessor 44 generates analog control signals that are converted 20 into corresponding digital control signals by DAC 46 and used as a reference signal Pset (t) for the servo 20.

One method for the operation of a respiratory apparatus 10 is shown in the flow chart of FIG. 2. Individuals skilled in the art will recognize other methodologies for utilizing the improved flow flattening index that is disclosed herein. The 25 embodiment of the methodology of FIG. 2 is also detailed in U.S. Pat. No. 5,704,345 (the '345 patent). The first step 100 is the measurement of respiratory flow over time. This information is processed in step 102 to generate Index values to be used as qualitative measures for subsequent processing. Thus,

Step 102 includes the generation of obstruction index values based upon the 11 averaging method as disclosed herein. Step 104 detects whether an apnea is occurring by comparison of the breathing Index with a threshold value. 2015204360 16 Μ 2015

If the answer in step 104 is "Yes", an apnea is in progress and there then follows a determination of patency in step 110 by methods disclosed in the mentioned 5 Wickham or Berthon-Jones patents. If there is patency of the airway, a central apnea with an open airway is occurring, and, if desired, the event is logged in step 112. If the result of step 110 is that the airway is not patent, then a total obstructive apnea or a central apnea with closed airway is occurring, which results in the commencement or increase in CPAP treatment pressure in step 10 108. If desired, step 108 may include the optional logging of the detected abnormality. Alternatively, flattening is simply rejected when apnea occurs since the breath pattern will not satisfy the criteria of either the number of samples or the required tidal volume - that is required for a valid breath for flattening - when apnea is present. 15 If the answer in step 104 is "No", one or more obstruction indices, such as the improved flow flattening indices, are compared with threshold values in step 106, by which the determination of obstruction of the airway is obtained. If the answer is "Yes" in step 106, then there is a partial obstruction, and if "No", there is no obstruction (normalcy). 20 In the case of a complete or partial obstruction of the airway Step 108 applies and a consequential increase in CPAP treatment pressure occurs. In the instance of normal breathing with no obstruction, the CPAP treatment pressure is reduced, in accordance with usual methodologies that seek to set the minimal pressure required to obviate, or at least reduce, the occurrence of apneas. The 25 amount of reduction in step 107 may, if desired, be zero. Similarly, in the event of a central apnea with patent airway (step 110,112) treatment pressure is not increased. Such increases in pressure reflexively inhibit breathing, further aggravating the breathing disorder. 12

Improved Flow Flattening Indices 2015204360 16 Μ 2015 FIG. 3 depicts an airflow signal with respect to the inspiratory portion of a typical breathing cycle. During the inspiratory portion of the breathing cycle of a healthy person (solid line), the airflow rises smoothly with inspiration, reaches a peak and 5 falls smoothly to zero. However, a patient with a partially obstructed airway (dashed line) exhibits a breathing pattern characterized by a significant flat zone during inspiration. Theoretically, for an obstructed flow, as the degree of partial obstruction increases, the airflow signal for inspiration would tend to a square wave. 10 Detection of Inspiratory and Expiratory half cycles in breathing

As shown in Fig. 4, a normal respiratory cycle consists of inspiration and expiration and a brief end-expiratory pause. During the end-expiratory pause small oscillations in airflow due to heartbeat occur. In order to correctly detect the start of inspiration, it is necessary to ignore the small pulses of air due to 15 heartbeat. Similarly it is necessary to ignore brief cessation due to swallowing.

The start of inspiration may be taken provisionally as the time that inspiratory airflow exceeds a threshold of about 0.05 L/sec. If the flow then falls below zero again before the total volume reaches about 0.1 liters, it is assumed that expiration is continuing. 20 The start of expiration is taken provisionally as the time that expiratory airflow falls below zero. If the airflow then goes above about 0.05 L/sec in the inspiratory direction before the total volume expired reaches about 0.1 liters, it is assumed that inspiration is continuing.

The purpose of the above algorithm is to detect the start and end of each 25 respiratory half-cycle. This information is required for the shape detector. It is immune to events such as swallows and cardiogenic airflow (which do not achieve the 0.1 litre volume requirement). A normal breath has a peak flow of about 0.3 L/sec, and a tidal volume of 0.5 liters. Hence there is a 3 to 5-fold 13 reserve ability to detect normal breaths. Since the timing information is used for shape algorithms, where very small breaths are irrelevant, it is not necessary to respond to extremely small breaths. Because the threshold crossing point is detected without further low pass filtering, the algorithm does not introduce phase 5 delay. 2015204360 16 Μ 2015

Calculation of Curvature or Flattening index A normal inspiratory flow-time curve is rounded, or quasi-sinusoidal in shape. Diaphragm muscular effort available for producing airflow is low at the beginning and end of the breath, and high in the middle of the breath. With adequate CPAP 10 pressure the airway is acting essentially as a rigid tube, and flow is a function of effort: increasing effort produces increasing flow.

If the CPAP pressure is adequate to prevent apnoeas and snoring, but not adequate to achieve full patency, the airway now behaves like a floppy elastic tube, and increasing effort causes increasingly subatmospheric pressure, leading to 15 progressive narrowing of the airway. Above a certain threshold effort, the airway narrows: the increased narrowing exactly compensates for the increasing effort, so flow becomes constant, independent of effort. Therefore the flow-vs.-time curve will approximate a square wave as shown in the middle panel of Fig. 5.

In practice, if effort is only above the threshold flattening behavior during the middle 20 part of the breath where effort is maximal, the flattening will be most noticeable over the middle half of the inspiration. Conversely, if effort is very high, inertial effects can lead to a brief period of high flow early in the breath, causing the initial overshoot shown in the right hand frame of Fig. 5. Hence, the present algorithm looks specifically at the middle half of the inspiration, not at the entire breath. 25 In Fig. 5, the inspiratory flow-vs.-time curve is scaled to unit duration and area. RMS deviation from unit flow is then calculated over the middle 50% of inspiratory time. The left hand panel shows a normal curve, where the RMS deviation is large, and the right hand panels show flow limited curves, with a small RMS deviation. 14 A curvature or flattening index is derived, in order to quantify the degree of mid inspiratory flattening of the flow-time curve, using the following steps: 2015204360 16 Μ 2015 1. Referring to Fig. 6, any end-expiratory pause that has been included in the inspiratory half cycle is trimmed off as follows: Find where flow first reaches 75% of 5 peak inspiratory flow “a", search backwards to the point where the flow last reached 25% of the peak “b”, and extrapolate backwards to locate the time at which ray Mab” reaches the x-axis, i.e. to find the time where flow should be zero “c”. This is taken as the earliest time at which inspiration could have started. In addition if “c” is to the left of origin “d”, use origin “d”. 10 2. The approximate effect of a 0.1 Hz low pass filter on the flow-time curve may be calculated and reversed as follows: (a) Pass the wave through an identical 0.1 Hz low pass filter a second time, (b) Subtract the doubly filtered wave from the singly filtered wave to give a first-order approximation of the effect of the filter, (c) Add this difference back to the original wave, Alternatively, this compensation for leak filter 15 may be expected not to alter the results significantly and may therefore be removed for computation efficiency.

For breaths of interest, inspiratory duration is about 2 seconds. The 0.1 Hz filter produces some distortion of the wave shape. For example, the trailing edge of a 2 second square wave sags by 20% of its amplitude, and the area under the curve is 20 reduced by 10%. After correction, the trailing edge sags by only 2% of its amplitude, and the area under the curve is correct to 0.7%. The filter reduces the area under a sine wave by 10%, and after correction, the area is correct to 0.6%. This is more than adequate for our current purposes. 3. The effect of overshoot is normalized as will be described below. 25 4. The inspiratory flow-time curve is scaled to unit length and unit mean height. The scaling is because we are interested in the shape of the flow-time curve (rounded vs. flattened), not its amplitude or duration. Even with correct CPAP pressure, amplitude and duration are very variable, (particularly in REM sleep but also in stage 15 1 sleep and awake with sighs etc). Therefore, length and amplitude convey no important Information about the adequacy of the CPAP pressure. 2015204360 16 Μ 2015 5. Breaths with grossly abnormal shapes are detected by comparing with a template. Breaths that differ from the template by more than a threshold value at any point 5 over the middle half are rejected. The default threshold is 1,0 unit, 6. The mean deviation of the inspiratory wave from the reference amplitude (M), for a middle portion of the wave is calculated. For example if the middle portion is one quarter of the inhalation wave and is centered then the mean deviation is: 40 £O(0-m)

MeanDeviation = —- 16 10 7. The Flattening Index (FI) is calculated as __ MeanDeviation FI =-.

M

Typical values of the Flattening index for actual patients gives ~0.21 for normal curves; 0.15 is a threshold below which the device should increase the CPAP pressure. 15 8. If the breath is not rejected, a multiple breath pointwise moving average flow-time curve is updated, for example a 5 breath average.

The purpose of step 8 (averaging the waveform over 5 breaths) is to reduce the effect of cardiogenic airflow, which could disguise an otherwise flattened flow-time curve. Swallows, coughs, talking, etc can produce very abnormally shaped breaths. 20 The purpose of step 5 is to prevent such breaths being included in the 5 breath average, which would delay the recognition of genuine flow limitation.

Step 8 may be unnecessary, because during flow limitation, cardiogenic pressure oscillations may not cause important changes in flow rate, due to effort 25 independence. If this is the case, step 5 is also unnecessary, because coughs, 16 swallows, etc will not be flat over the middle half of inspiration, and will therefore be correctly classified as not flow-limited. 2015204360 16 Μ 2015

Step 3, the normalization of overshoot, may be accomplished in several embodiments. This aspect of the invention assumes that 1)peak flow occurs 5 close to middle portion of the wave, 2)the first 3rd of the wave should have a generally positive gradient 3) the last third a generally negative gradient. 4)Any deviation from these conditions would indicate an overshoot.

In one embodiment the method comprises of 1) estimating peak flow in a middle portion of the wave, for example the middle third of the wave 2) any point outside 10 that portion, for example in the first and/or last third of the said wave that exceeds the said peak flow is reset to the peak flow. This method therefore reduces the impact of the overshoot.

In a second embodiment the method comprises 1) estimating peak flow in the middle portion (e.g. a third) of the wave 2) any point outside that portion (e.g. in 15 the first and/or last third of the said wave) that exceeds the peak flow is reset to the average of the sample before and after. 3) The process continues until no sample in the outside portions is greater than the peak flow. This method again reduces the impact of the overshoot - while also maintaining the shape of the waveform. 20 in a third (preferred) embodiment the method consists of 1) Dividing the wave into n parts and calculate mean amplitudes for each part. 2) Apply logic to determine what sort of wave the pattern indicates. 3) If overshoot is indicated, mark flattening as low.

In a fourth embodiment the method consists of 1) Calculating estimate of 25 flattening for every n samples in the middle half of the wave (e.g. calculate flattening for sample 16..24,17..25, and so on). 2) Marking the flattening index as the value that corresponds to the lowest value obtained for all sample permutations. 17

Alternative Statistical Calculation for Μ-Shape Detection 2015204360 16 Μ 2015

An alternative embodiment of the invention uses a different measure to determine whether an inspiratory wave pattern has an M shape. The technique is to decompose the inspiratory wave into a vector of features (floating point 5 numbers) and a matrix pair of basis vectors that are orthogonal. Normalizing the time interval so that inspiration begins at zero and ends at B radians, the basis vectors may be chosen to be sin (Bt) and sin(3Bt). The inspiratory waveform is thus fit to the curve a sin (Bt) + b sin(3Bt), where a and b are determined by fitting the basis vectors to the inspiratory waveform in a least-squares sense 10 using singular value decomposition. This is equivalent to finding the pseudoinverse of the basis matrix and then multiplying the inspiratory waveform by this pseudo-inverse to find the factors. The pseudo-inverse can be calculated offline, i.e. not by the CPU contained in the ventilator.

To reduce the amount of calculation involved in this statistical calculation it is 15 possible to map the [a,b] space for many patients and note the clusters into which the points representing the individual patients fall. The M-shaped breaths will be found clustered away from the non-M-shaped breaths. Thus whether a breath is M shaped can be determined by the Euclidean measure of distance from the M cluster center. 20 Example of M Shape Index Calculation

Each inspiration is interpolated over a grid of N points, preferably N = 65. Two basis functions as shown in Fig. 7are calculated for t = i/(N-1) where i goes from 0 to N-1, as B1 = sin( π t) 25 B2 = sin( 3 π t)

These basis functions can then be stored for use with all subsequent calculations of the M-shaped feature. 18

Each inspiration is then extracted and interpolated over a grid of N points. Two factors are then calculated as: 2015204360 16 Μ 2015

Fi = sum( B1 · fs ) F2 = sum( B2 · fs ) 5 where fs represents the interpolated inspiration points and · the dot-product operator.

The final shape value is obtained by normalising as: shape index = 10 This shape factor is then limited to vary between zero (purely sinusoidal) to one (very M-shaped),

For a typical M-shaped breath, as shown in Fig. 8, the above calculations yield F1 = 4.6082 F2 = 2.6538 15 and shape index = 0.50. This may be contrasted with a typical non-flowlimited breath that has an M-shape index of only about 0.2

Although the invention has been described with reference to a particular embodiment, it is to be understood that this embodiment is merely illustrative of 20 the application the principles of the invention. Thus, it is to be understood that numerous modifications may be made in the illustrative embodiment of the invention and other arrangements may be devised without departing from the spirit and scope of the invention. For example, while the preferred embodiment of the invention divides the waveform into specific segments for analysis and 25 concentrates on the central values, other divisions and off center regions might equally apply. 19

Claims (31)

  1. CLAIMS:
    1. An apparatus for detection of inspiratory and expiratory half cycles in breathing comprising: an air source adapted to selectively provide pressurized breathable air to a patient, a flow detecting device configured to sense respiratory airflow from the patient and to generate an airflow signal indicative of airflow of a breathing cycle, and a microprocessor configured to: provisionally take a start of inspiration as a time that inspiratory airflow exceeds a predetermined threshold, reject the start of inspiration if the flow then falls below zero again before a total volume reaches a predetermined value, provisionally take a start of expiration as a time that expiratory airflow falls below zero, and reject the start of expiration if the airflow then goes above a predetermined value in an inspiratory direction before a total volume expired reaches a predetermined value.
  2. 2. The apparatus of claim 1, wherein the microprocessor is further configured to trim any end-expiratory pause that has been included in the inspiratory half cycle.
  3. 3. The apparatus of any one of claims 1 or 2, wherein the microprocessor is further configured to find an earliest time at which inspiration could have started by: locating the time where flow first reaches a predetermined percentage of peak inspiratory flow, searching backwards to the time where the flow last reached a second predetermined percentage of the peak, and extrapolating backwards to find the time where flow should be zero.
  4. 4. The apparatus of any one of claims 1 to 3, wherein the microprocessor is further configured to reverse an approximate effect of a low pass filter on a flow-time curve of the airflow signal by: (a) passing a wave of the flow-time curve through an identical low pass filter a second time, (b) subtracting the doubly filtered wave from the singly filtered wave to give a first-order approximation of an effect of the filter, and (c) adding this difference back to the original wave.
  5. 5. The apparatus of any one of claims 1 to 4, wherein the microprocessor is further configured to normalize an effect of overshoot by scaling an inspiratory flowtime curve to unit length and unit mean height.
  6. 6. The apparatus of any one of claims 1 to 5, wherein the microprocessor is further configured to calculate a mean deviation of an inspiratory wave from a reference amplitude (M), for a middle portion of the wave.
  7. 7. The apparatus of claim 6, where the middle portion is one quarter of an inhalation wave and is centered then the mean deviation is determined as:
    where fs(t) represents a sample of the inspiratory wave.
  8. 8. The apparatus of claim 7, wherein the microprocessor is further configured to determine a flattening Index (FI) as
  9. 9. The apparatus of any one of claims 1 to 8, wherein the microprocessor is further configured to update a multiple breath pointwise moving average flow-time curve.
  10. 10. The apparatus of any one of claims 1 to 9, wherein the microprocessor is further configured to reduce an impact of an overshoot by: estimating peak flow in a middle portion of a wave, and resetting to the peak flow each point outside a portion that exceeds the said peak flow, wherein the wave is of a flow-time curve of the airflow signal.
  11. 11. The apparatus of claim 10, wherein the microprocessor is further configured to reduce the impact of the overshoot by continuing until a magnitude of no sample in the outside portions is greater than the peak flow.
  12. 12. The apparatus of claim 1, wherein the microprocessor is further configured to: divide a wave into n parts and calculate mean amplitudes for each part; determine what sort of pattern the wave indicates; and if overshoot is indicated, mark flattening as low, wherein the wave is of a flow-time curve of the airflow signal.
  13. 13. The apparatus of claim 1, wherein the microprocessor is further configured to: calculate an estimate of flattening for every n samples in a middle half of a wave, and mark a flattening index as a value that corresponds to a lowest value obtained for all sample permutations, wherein the wave is of a flow-time curve of the airflow signal.
  14. 14. The apparatus of claim 1, wherein the microprocessor is further configured to determine whether an inspiratory wave pattern has an M shape by: decomposing the inspiratory wave into a vector of features and a matrix pair of basis vectors that are orthogonal, normalizing so that inspiration begins at zero and ends at B radians, finding a pseudo-inverse of the basis matrix, and multiplying the inspiratory wave by this pseudo-inverse to find factors.
  15. 15. The apparatus of claim 14, wherein the microprocessor is further configured to determine whether an inspiratory wave pattern has an M shape by determining a Euclidean measure of distance from an M cluster center of a mapped linear space.
  16. 16. A method for detecting inspiratory and expiratory half cycles in a breathing signal comprising, with a processor: obtaining a flow signal representative of respiratory flow of a breathing cycle; analyzing the flow signal to provisionally take a start of inspiration as a time that inspiratory flow exceeds a predetermined threshold; confirming that inspiration has started when the flow remains above zero before a total volume reaches a predetermined value or else rejecting the start of inspiration; analyzing the flow signal to provisionally take a start of expiration as a time that expiratory flow falls below zero; and confirming that expiration is occurring when the flow remains below the predetermined threshold before a total volume expired reaches a predetermined value or else rejecting the start of expiration, wherein the inspiratory half cycle is detected from a confirmed start of inspiration until a confirmed start of expiration and the expiratory half cycle is detected from a confirmed start of expiration until a confirmed start of inspiration.
  17. 17. The method of claim 16, further comprising, with a processor, analyzing the detected inspiratory half cycle to detect a presence of an end-expiratory pause and trimming the end-expiratory pause from the detected inspiratory half cycle.
  18. 18. The method of claim 16 or 17, further comprising, with a processor, finding an earliest time at which inspiration could have started by: analyzing the flow signal to locate a first time where flow first reaches a predetermined percentage of peak inspiratory flow, searching backwards to a second time where the flow last reached a second predetermined percentage of the peak, and extrapolating backwards to find the time where the flow should be zero.
  19. 19. The method of any one of claims 16 to 18, further comprising, with a processor, reversing an approximate effect of a low pass filter on a flow-time curve of the flow signal by: (a) passing a wave of the flow-time curve through an identical law pass filter a second time, (b) subtracting the doubly filtered wave from the singly filtered wave to give a first-order approximate of an effect of the filter, and (c) adding this difference back to the original wave.
  20. 20. The method of any one of claims 16 to 19, further comprising, with a processor, normalizing an effect of overshoot by scaling an inspiratory flow-time curve to unit length and unit mean height.
  21. 21. The method of any one of claims 16 to 20, further comprising, with a processor, calculating a mean deviation of an inspiratory wave from a reference amplitude (M), for a middle portion of the wave.
  22. 22. The method of claim 21, wherein the middle portion is one quarter of the inspiratory wave and is centered.
  23. 23. The method of claim 22, further comprising, with a processor, determining the mean deviation as:
    where fs(t) represents a sample of the inspiratory wave.
  24. 24. The method of claim 23, further comprising, with a processor, determining a flattening index (FI) as
  25. 25. The method of claims 16 to 24, further comprising, with a processor, updating a multiple breath pointwise moving average flow-time curve.
  26. 26. The method of claims 16 to 25, further comprising, with a processor, normalizing an effect of overshoot by: determining a peak flow of a central portion of an inspiratory part of the flow, and truncating measured flow values outside the central portion of the inspiratory part that exceed a peak flow value of the central portion.
  27. 27. The method of claim 26, wherein the truncation is to the peak flow value of the central portion.
  28. 28. The method of claim 16, further comprising, with a processor: dividing a wave into n parts and calculate mean amplitudes for each part; determining what sort of pattern the wave indicates; and marking flattening as low if overshoot is indicated, wherein the wave is of a flow-time curve of the flow signal.
  29. 29. The method of claim 16, further comprising, with a processor: calculating an estimate of flattening for every n samples in a middle half of a wave, and marking a flattening index as a value that corresponds to a lowest value obtained for all sample permutations, wherein the wave is of a flow-time curve of the flow signal.
  30. 30. The method of claim 16 further comprising determining whether an inspiratory wave pattern has an M shape by: decomposing the inspiratory wave into a vector of features and a matrix pair of basis vectors that are orthogonal, normalizing so that inspiration begins at zero and ends at B radians, finding a pseudo-inverse of the basis matrix, and multiplying the inspiratory wave by this pseudo-inverse to find factors.
  31. 31. The method of claim 30 further comprising determining whether a breath is M shaped with a Euclidean measure of distance from an M cluster center of a mapped linear space.
AU2015204360A 2006-03-06 2015-07-16 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea Active AU2015204360B2 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US60/779,577 2006-03-06
AU2007222882A AU2007222882B2 (en) 2006-03-06 2007-03-06 Method and apparatus for improved flow limitation detection of obstructive sleep apnea
AU2012258376A AU2012258376B2 (en) 2006-03-06 2012-11-23 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea
AU2015204360A AU2015204360B2 (en) 2006-03-06 2015-07-16 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
AU2015204360A AU2015204360B2 (en) 2006-03-06 2015-07-16 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
AU2012258376A Division AU2012258376B2 (en) 2006-03-06 2012-11-23 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea

Publications (2)

Publication Number Publication Date
AU2015204360A1 AU2015204360A1 (en) 2015-08-06
AU2015204360B2 true AU2015204360B2 (en) 2017-09-14

Family

ID=53761675

Family Applications (1)

Application Number Title Priority Date Filing Date
AU2015204360A Active AU2015204360B2 (en) 2006-03-06 2015-07-16 Method and Apparatus for Improved Flow Limitation Detection of Obstructive Sleep Apnea

Country Status (1)

Country Link
AU (1) AU2015204360B2 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5632269A (en) * 1989-09-22 1997-05-27 Respironics Inc. Breathing gas delivery method and apparatus
US5704345A (en) * 1993-11-05 1998-01-06 Resmed Limited Detection of apnea and obstruction of the airway in the respiratory system
US6814073B2 (en) * 2000-08-29 2004-11-09 Resmed Limited Respiratory apparatus with improved flow-flattening detection

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5632269A (en) * 1989-09-22 1997-05-27 Respironics Inc. Breathing gas delivery method and apparatus
US5704345A (en) * 1993-11-05 1998-01-06 Resmed Limited Detection of apnea and obstruction of the airway in the respiratory system
US6814073B2 (en) * 2000-08-29 2004-11-09 Resmed Limited Respiratory apparatus with improved flow-flattening detection

Also Published As

Publication number Publication date
AU2015204360A1 (en) 2015-08-06

Similar Documents

Publication Publication Date Title
US7992557B2 (en) Gas supply device for sleep apnea
EP1900324B1 (en) Apparatus for determining patency of the airway
US6752766B2 (en) Method and device for sleep monitoring
US6810876B2 (en) Assisted ventilation to match patient respiratory need
EP0656216B1 (en) Estimation of flow and detection of breathing in CPAP treatment
US6988498B2 (en) Administration of CPAP treatment pressure in presence of apnea
AU2004308537B2 (en) Mechanical ventilation in the presence of sleep disordered breathing
US6409676B2 (en) Process for determining respiratory phases of the sleep of a user
US7730886B2 (en) Determination of patency of the airway
EP0956066B1 (en) Apparatus for optimizing the continuous positive airway pressure for treating obstructive sleep apnea
US6588422B1 (en) Method and apparatus to counterbalance intrinsic positive end expiratory pressure
JP3635097B2 (en) Determining the leakage and respiratory airflow
US7089936B2 (en) Device for detecting the respiratory activity of a person
JP5657380B2 (en) Automatic control for the detection of flow limitation
JP5546544B2 (en) Model predictions online identification of the patient's respiratory effort dynamics in a medical ventilator
JP4429599B2 (en) Apparatus for supplying a variable positive airway pressure
EP1654024B1 (en) Positive airway pressure system for treatment of sleeping disorder in patient
US6299581B1 (en) Method and apparatus for optimizing the continuous positive airway pressure for treating obstructive sleep apnea
US6257234B1 (en) Apparatus and method for determining respiratory mechanics of a patient and for controlling a ventilator based thereon
US6761168B1 (en) Gas supply device for sleep apnea
US6761167B1 (en) Gas supply device for sleep apnea
US20030213488A1 (en) Auto CPAP system profile information
JP3701670B2 (en) Sleep disordered breathing control method and apparatus
US20130025596A1 (en) Methods and systems for model-based transformed proportional assist ventilation
CN101043913B (en) For detecting invalid improve patient inspiratory effort and ventilator - interaction means and methods

Legal Events

Date Code Title Description
FGA Letters patent sealed or granted (standard patent)
HB Alteration of name in register

Owner name: RESMED PTY LTD

Free format text: FORMER NAME(S): RESMED LIMITED