Department of Signal and Communications, Institut Télécom; Télécom Bretagne, 29238 Brest, France

Medical Intensive Care Unit, CHRU de Brest/INSERM U1101 Latim Ubo, 29200 Brest, France

Abstract

Background

Dynamic hyperinflation, hereafter called AutoPEEP (auto-positive end expiratory pressure) with some slight language abuse, is a frequent deleterious phenomenon in patients undergoing mechanical ventilation. Although not readily quantifiable, AutoPEEP can be recognized on the expiratory portion of the flow waveform. If expiratory flow does not return to zero before the next inspiration, AutoPEEP is present. This simple detection however requires the eye of an expert clinician at the patient’s bedside. An automatic detection of AutoPEEP should be helpful to optimize care.

Methods

In this paper, a platform for automatic detection of AutoPEEP based on the flow signal available on most of recent mechanical ventilators is introduced. The detection algorithms are developed on the basis of robust non-parametric hypothesis testings that require no prior information on the signal distribution. In particular, two detectors are proposed: one is based on SNT (Signal Norm Testing) and the other is an extension of SNT in the sequential framework. The performance assessment was carried out on a respiratory system analog and

Results

The experiment results have shown that the proposed algorithm provides relevant AutoPEEP detection on both simulated and real data. The analysis of clinical data has shown that the proposed detectors can be used to automatically detect AutoPEEP with an accuracy of 93% and a recall (sensitivity) of 90%.

Conclusions

The proposed platform provides an automatic early detection of AutoPEEP. Such functionality can be integrated in the currently used mechanical ventilator for continuous monitoring of the patient-ventilator interface and, therefore, alleviate the clinician task.

Introduction

Mechanical ventilation is routinely used in the clinical ward and/or in nursing/rehabilitation institutions. Unfortunately, imperfect interaction between patient and ventilator is frequently exhibited in intubated patients

It has been demonstrated that the graphical curves (flow, airway pressure and air volume) available on most recent mechanical ventilators provide much information to analyze the patient-ventilator interface

This paper addresses automatic detection of AutoPEEP, a common ventilatory abnormality that usually occurs in patients with acute severe asthma or chronic obstructive pulmonary disease. The presence of AutoPEEP basically indicates an insufficient expiratory time. The amount of time given over to expiration therefore needs to be lengthened, either by reducing the respiration rate or by decreasing the inspiratory time, or both. AutoPEEP can be measured at the patient’s bedside by using the pressure transducer of the ventilator. However, this quantification requires intervention from the therapist, who must perform an expiratory pause, in order to monitor tele-expiratory pressure

Methods

Automatic detection of AutoPEEP and System overview

AutoPEEP can be visually observed and detected through flow signal. Figure
_{
t
} be the clean flow signal. AutoPEEP can be regarded as the non-return of the flow signal at the end of each expiratory phase to the null value. In practice, during the observation of the air flow, various factors might get involved, including the mechanical vibration of the air tube, the patient movement, the electro-magnetic interference, etc. Therefore, the flow signal at the end of the expiratory phase will never be exactly zero, even in absence of noise. Testing directly the hypothesis
_{
k
} is the end-expiration instant of the considered breath, might thus not be realistic. A tolerance

An example of flow signal

**An example of flow signal.** This signal was recorded during the assisted mechanical ventilation on a patient. The (blue) curve shows a typical waveform of flow signal with squared inspiratory phase. The arrows point to some end-expiration instants where the markers for AutoPEEP detection are present.

With respect to the discussion above, a platform for automatic detection of AutoPEEP based on a noisy observation of the flow signal can be developed. Figure

Automatic AutoPEEP Detection Platform - System overview

**Automatic AutoPEEP Detection Platform - System overview.** The platform functions on the basis of respiratory flow signal. For each end-expiration _{k} it detects, the Phase change detector triggers the data acquisition/conversion process. Based on observations **Y**_{k} provided by the Data acquisition/conversion module and parameters

Data acquisition and Serial/Parallel conversion

This very-first module acquires the discrete flow signal _{
n
} provided by the ventilator or by an independent flow sensor installed inside the air-tube during the mechanical ventilation. Although every flow datum is acquired, only end-expiration flow data of each breath is useful for the detection of AutoPEEP. When the end-expiration instant _{
k
} of the **Y**
_{
k
} is finally injected into the AutoPEEP detector module.

Respiration phase change detection

The main role of this module is to detect the end-expiration of each breath and provide this instant to trigger the data logging process and the Serial/Parallel conversion described above. This can also be regarded as a breath detector, which separates the continuous flow signal into different breaths.

Estimator

This module consists of two estimators, which estimate necessary parameters for the AutoPEEP detection algorithms. These parameters are the so-called waveform vector (**p**
_{
k
}for the

AutoPEEP detector

The AutoPEEP detector is the main core of the whole platform. Given a specified tolerance **Y**
_{
k
} and estimated parameters

AutoPEEP detectors

Given tolerance _{
n
} of the noisy flow signal, the AutoPEEP detection is the testing of the null hypothesis

Single-breath detector

Signal Norm Testing

To begin with, let us consider the signal model:

where _{0}:|_{1}:|

In the sequel, a _{0}, regardless of which hypothesis actually holds, i.e.

The _{0}:|

and its

in which _{
γ
}(

Single-breath SNT-based AutoPEEP detector

Although the definition of AutoPEEP is based solely on the final sample of the expiratory phase of each breath, it is expected that taking multiple samples into account will improve the detection performance. By introducing the waveform vector, namely **p**
_{
k
}, with dimension **Y**
_{
k
} be the observation vector containing the last **Y**
_{
k
} is modeled as:

where
**f**
_{
k
} can be factorized as:

where
**p**
_{
k
} corresponds to the local form of the flow signal near the end of the expiratory phase. It is also worth mentioning that this local waveform vector **p**
_{
k
} depends mainly on the configuration of the interface, including the patient condition and the ventilator settings. As long as the interface stays unchanged, the waveform vector remains almost the same regardless whether or not an AutoPEEP might occur. In practice, either **p**
_{
k
} is known prior to the detection or it can be estimated from the observation using one of the methods proposed in Section Waveform regression to compute **p**
_{
k
}.

To aggregate **Y**
_{
k
} is projected onto the direction generated by **p**
_{
k
}. We thus have:

where
_{2} norm of waveform vector **p**
_{
k
}. By such proceeding, noise _{
k
} follows normal distribution with zero mean and variance
_{
k
} tends to a normally distributed random variable, as long as _{
w
}≤**X**
_{
k
} is required. The two hypotheses are unchanged:
_{
γ
}(.) is calculated as in (3). Otherwise, the considered breath is labeled with Non-AutoPEEP.

It should be noted that ∥**p**
_{
k
}∥ increases with respect to the number _{
w
} will thus decreases when more samples are taken into account. By reducing the noise standard deviation, the detection probability is improved while the false-alarm rate is still limited to the specified level

Sequential detector

SNT extension in sequential decision framework

By using SNT, one can restrict the false-alarm rate to some value

and the from-below one:

into account. On the one hand, put

has size

We also have:

For

for testing the hypothesis _{0}:|_{1}:|_{FA}) is limited to the specified value _{FA} <_{D}) is guaranteed to be higher than 1−_{D} > 1 −

In a sequential framework, the test

Sequential SNT-based AutoPEEP detector

Let us consider

and

where:

It is worth mentioning that

Assuming that the true hypothesis (AutoPEEP/NON-AutoPEEP) remains the same for

for testing

for testing

and

where

Summarizing, in a sequential decision framework, the AutoPEEP detection is carried out as follows. Firstly, the detector tries to make a decision based solely on the observation of the first breath, using the test:

If the decision cannot be made yet (i.e.
_{2}) is obtained and the test is performed based on _{1:2} using

As shown in Figure

The value

Thresholds convergence

**Thresholds convergence.** This figure illustrates the convergence of the two thresholds in Sequential SNT framework. This convergence suggests that, in sequential SNT framework, the decision will probably be made after a finite number of samples are acquired.

Phase change detection

Since the detection is performed on the basis of the flow samples at the end of the expiratory phase of each breath, it is required that the end-expirations are precisely retrieved. As aforementioned, the main role of the Phase change detection/segmentation block is to provide a detection of the end-expiration for each breath. This can be achieved by detecting the change in flow signal _{
n
} from the expiratory phase of the current breath (negative values) to the inspiratory phase of the next breath (positive values) (c.f. Figure

where 2

Since the wavelet transform is a powerful processing tool to retrieve irregularities in a signal, it can be used to carry out the detection of change from the expiratory phase of a breath to the inspiratory phase of the next one. The wavelet transform is applied on either the flow signal _{
n
} on its smoothed version

Wavelet decomposition of the flow signal

**Wavelet decomposition of the flow signal.** The peaks in detail bands correspond to changes from inspiratory phase to respiratory phase and vice versa.

The end-expiration detection is performed by thresholding these peaks in the detail bands of the wavelet transform coefficients. Let us consider the level-2 detail band for instance. This detail band signal is composed of noise and peaks, which represent the irregularities in the original flow signal. Since the flow signal is supposed to be in independent additive gaussian noise and since the wavelet transform is linear, the noise in detail band is also gaussian. Therefore, each coefficient in the detail band can be modeled as _{
D
}=_{
D
} + _{
D
}, where _{
D
} is a signal coefficient and _{
D
} is gaussian noise. Let _{
D
} be the standard deviation of this noise and let _{
u
}(_{
u
}(

with threshold height
_{
γ
}(^{−4},10^{−5},10^{−7}, etc. Since a peak is only one point, the results of the thresholding test should be post-processed in such a way that consecutive 1s are removed. In particular, in case of consecutive decisions equal to 1, only the first one will be kept. End-expirations are negative peaks.

As long as noise standard deviation _{
D
} is concerned, it can be estimated using the same methods as those described in Section Estimation of the noise standard deviation. Figure

End-Expiration Detection using Wavelet transform

**End-Expiration Detection using Wavelet transform.** This figure illustrates the detection of end-expirations based on respiratory flow signal: (top) respiratory flow curve obtained from a patient, (middle) signal in the level-2 detail band of the wavelet transform coefficients and the calculated detection threshold, (bottom) detection result, where 1s (peaks) represent end-expirations.

Estimation

As aforementioned, some estimations have to be made prior to the AutoPEEP detection, including: the waveform vector (**p**
_{
k
}) and the standard deviation of the unknown noise. In the following, these two estimations will be addressed.

Waveform regression to compute **p**
_{
k
}

With regard to Section Single-breath detector, the waveform vector **p**
_{
k
} is the key which makes it possible to aggregate multiple end-expiration flow samples into one decision. This vector **p**
_{
k
} can be calculated from the regression of the flow signal at the end of the expiration. Indeed, during the expiratory phase of a breath, the mechanical ventilation system works based solely on the passive response of the patient lung. Due to the resistance of the airways and the elasticity of human lung, the flow signal during the expiratory phase of a breath can be modeled by:

with _{
i,}
_{
i
})),_{
i
}) is the observation at instant _{
i
}, the non-linear robust regression aims at solving the least square problem:

where the introduction of weight vector [_{1},_{2},_{
N
}] makes it possible to reduce the influence of outliers onto the final result. The MATLAB routine

Fitness of the model function

**Fitness of the model function.** An example of the flow signal at the end of an expiratory phase with its regression curve using the model function in (12). The result firmly shows the relevance of the considered model function to the regression task.

Even though only **p**
_{
k
}, more samples should be used to achieve a better regression curve. Let _{ext}(_{ext} ≥_{ext} is only limited by the length, namely _{
e
} (in samples), of the expiratory phase, i.e. _{ext} ≤_{
e
}. Regarding the transition between different respiratory phases, samples at the beginning of the expiration are very sensitive to transition and may bias the regression. Therefore, only a proportion of the _{
e
} samples of the expiratory phase should be taken into account:

where 0 <

Detection results on clinical data

**Detection results on clinical data.**

Given the regression at the end of expiration, namely

According to Section Single-breath detector, waveform vector **p**
_{
k
} concerns the current (**p**
_{
k
}. In this respect, the following strategies can be considered to compute the waveform vector estimate to be used in AutoPEEP detectors:

"**Static waveform vector:** The waveform vector is computed based on the first _{ref} breaths right after a verification/tuning session of the clinician. These _{ref} breaths are used as reference after validation by the clinician.

This waveform vector will be updated each time the machine is tuned or after a verification session by the clinician. One may also want to update the estimation on a regular time basis.

**Dynamic waveform vector:** The waveform vector to be used is the one estimated from the current breath:

**Adaptive waveform vector:** In this strategy, the waveform vector is updated every time a new breath is observed. Previous estimates are taken into account with a forgetting factor

Estimation of the noise standard deviation

Noise is unknown in practice. As long as the noise standard deviation in concerned, it must be estimated from the observation. In this work, we propose two solutions: one based directly on the result obtained by waveform regression, whereas the other is based on an estimation from the wavelet coefficients of the flow signal.

Estimate from regression

By using the regression, the residue can be considered as noise. Therefore, the noise standard deviation can be estimated directly from this residue. For the

To aggregate

Estimation from wavelet coefficients

Studies on nonparametric estimation based on Wavelet Shrinkage have shown that most of the wavelet coefficients obtained from the first level wavelet decomposition of a piecewise smooth signal are of very small amplitude. Only a small number of these wavelet coefficients, which correspond to signal, are of higher amplitude
_{1},_{2},…_{
N
} be the wavelet coefficients obtained from the first level discrete wavelet decomposition of an

where

knowing that med_{
i
}
_{
i
}= 0.

In
_{(1)}
_{(2)},…,_{(N)} be sequence of wavelet coefficients _{1},_{2},…_{
N
} sorted by increasing magnitude. Put
_{min}≤

If such an integer _{min}. The estimate

It has been shown in

Results and discussion

Simulations

To illustrate the detection performance of the proposed algorithms, the flow signal was first synthesized on computer. For each breath, _{1}|≥|_{2}|≥…≥|_{
L
}|=1 and, as a result,
_{
i
}=1 for all _{
w
}, by setting
_{FA} is always restricted to the specified value _{D} versus different values of

Detection curves yielded by the two proposed AutoPEEP detectors with different noise levels

**Detection curves yielded by the two proposed AutoPEEP detectors with different noise levels.** The simulations were carried out with

Emulations with a respiratory system analog

The proposed AutoPEEP detectors were also tested in a more realistic setting in which the interface between a ventilator and a lung model was established. In these experiments, the respiratory system analog was constituted by a G5 ventilator (Hamilton Medical, Bonaduz, Switzerland) connected to the ASL5000 computerized lung model (Ingmar Medical Ltd., Pittsburgh, PA, USA), making it possible to modify respiratory mechanics. Thirteen sets of parameters (cf. Table

**Parameters**

**True**

**N. of**

**Det. by SNT**
^{c}

**Det. by Sequential SNT**
^{c}

**Id**

**Ventilator**
^{a}

**Lung model**
^{b}

**Label**

**breaths**

**P**

**N**

**Label**

**P**

**N**

**Label**

^{a}Ventilator parameters include: Positive Expiratory Pressure PEP [cm_{2}

^{b}Lung model parameters include: compliance C [ml/cm_{2}
_{2}

^{c}For each of the experiments, the AutoPEEP detection provides: the number of breaths detected as AutoPEEP (denoted as P for Positive), the number of breaths detected as NON-AutoPEEP (denoted as N for Negative) and the overall label for the considered setting.

1

PEP=0, Vt=500, f=15, P=0, I:E=1:2

C=80, R=5

N

21

0

21

N

0

21

N

2

PEP=0, Vt=500, f=15, P=0, I:E=1:2

C=30, R=5

N

20

0

20

N

0

20

N

3

PEP=0, Vt=500, f=25, P=0, I:E=1:2

C=80, R=5

P

33

33

0

P

33

0

P

4

PEP=0, Vt=500, f=25, P=0, I:E=1:1

C=80, R=5

P

34

34

0

P

34

0

P

5

PEP=0, Vt=300, f=20, P=0, I:E=1:2

C=80, R=5

N

27

0

27

N

0

27

N

6

PEP=0, Vt=500, f=12, P=0, I:E=1:2

C=80, R=5

N

16

0

16

N

0

16

N

7

PEP=0, Vt=500, f=20, P=15, I:E=1:3

C=80, R=5

N

27

0

27

N

0

27

N

8

PEP=5, Vt=500, f=20, P=0, I:E=1:3

C=80, R=5

N

27

0

27

N

0

27

N

9

PEP=5, Vt=500, f=20, P=0, I:E=1:2

C=120, R=10

P

27

27

0

P

27

0

P

10

PEP=0, Vt=700, f=20, P=0, I:E=1:2

C=120, R=10

P

27

27

0

P

27

0

P

11

PEP=0, Vt=700, f=20, P=0, I:E=1:6

C=120, R=10

P

24

24

0

P

24

0

P

12

PEP=0, Vt=700, f=20, P=0, I:E=1:1

C=120, R=10

P

27

27

0

P

27

0

P

13

PEP=0, Vt=700, f=20, P=0, I:E=1:2

C=140, R=25

P

13

13

0

P

13

0

P

Analysis of clinical data

For further evaluation, the AutoPEEP detectors were tested

The analysis was performed both manually by a set of experts and automatically by the proposed methods. On the one hand, each breath was carefully screened by two experts of the domain. They performed a dual analysis, separately, before confronting their points of view and delivering a final assessment of the data. For each breath of the dataset, their decision was then regarded as the ground-truth label (AutoPEEP/NON-AutoPEEP). On the other hand, the proposed detectors were used to predict the label of every breath of the dataset. The two analyses were carried out independently and anonymously. The results were then compared together to evaluate the detection performance of the proposed methods.

In these experiments, the tolerance was set to

To quantitatively assess the detection performance of the proposed methods, we considered four usual evaluation measures: Accuracy, Precision, Recall (Sensitivity) and Specificity. These measures are defined as follows:

where:

**Measure**

**Single-breath SNT-based detector**

**Sequential SNT-based detector**

The experiments were carried out with

Accuracy

**93.09%**

**93.09%**

Precision

99.44%

99.37%

Recall

**90.53%**

**90.60%**

Specificity

98.86%

98.70%

Conclusion

To the best of our knowledge, this is the first work on the automatic detection of AutoPEEP for continuous monitoring of the patient-ventilator interface during controlled mechanical ventilation. With the introduction of the waveform vector to aggregate multiple samples into a unique decision, the SNT has been successfully applied to provide a good AutoPEEP detector. Finally, we have extended SNT in a sequential framework, namely

Although the algorithm is proposed for the detection of AutoPEEP during controlled mechanical ventilation, it could be extended to assisted mechanical ventilation and pressure support ventilation since the algorithm investigates the expiratory part of the flow curve, which mainly depends on characteristics of the patient rather than on the ventilatory settings and mode of ventilation. The platform may also be extended to the detection of other types of ventilatory abnormalities that are deviations of the observed signal from some reference. In this respect, other signals such as pressure and volume curves could also be taken into account.

For the present work, by using the retrospective data files with a double-blinded and dual expert analysis, we were able to assess whether the system automatic analysis was concordant with that of the experts. In the next validation step, continuous and prospective recordings of the curves will be carried out to get better insight into cases where any disagreement between the proposed system and the therapist might occur. Furthermore, it is also worth performing a semi-closed-loop analysis, in which the therapist supervises, validates the decisions yielded by the proposed platform and adjusts the ventilatory parameters to correct any possible abnormality.

The deviation detection approach proposed in this paper is very general and could be used in many other applications, including fault detection and structural health monitoring. A theoretical general approach in Sequential SNT should also be investigated.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

QTN and DP carried out the system design, algorithm development, implementation, and wrote the paper. DP proposed SNT and QTN worked on the adaptation of SNT to the problem and extended SNT in the sequential framework. ELH stated the medical issue, contributed to suggestions on the topic, provided data, clinical analysis, discussion and manuscript writing. All authors read and approved the final manuscript.

Acknowledgements

The authors are very grateful to François Lellouche, Département de médecine - Institut universitaire de cardiologie et de pneumologie de Québec - Université Laval, Québec, Canada for his help while collecting data and for his valuable remarks without which this work might have never been done.