Method and system for breathing analysis using a personal digital assistant (PDA)

A system comprising:

• a memory storing instructions;
• one or more communication interfaces;
• and
• one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
• receive an acceleration data of a subject, using a PDA, at a pre-defined sampling rate, wherein the PDA comprises an accelerometer, and wherein the PDA is configured to be in contact with center of torso of the subject, and with a upper half of the PDA lying over the chest and a lower half of the PDA lying over the belly of the subject with a top of the PDA directed towards the chin of the subject, wherein the acceleration data is associated with breathing of the subject and the acceleration data comprises of a plurality of data samples (i) with a time stamp;
• represent the acceleration data as an acceleration vector (A), based on a vectorization technique, wherein the acceleration vector (A) is represented as a three-dimensional space (X-axis, Y-axis and Z-axis) along with the time stamp and is represented as:
• A
• i
• =
• a
• x
• ⁡
• (
• i
• )
• ⁢
• e
• 1
• +
• a
• y
• ⁡
• (
• i
• )
• ⁢
• e
• 2
• +
• a
• z
• ⁡
• (
• i
• )
• ⁢
• e
• 3
• wherein,
• i represents one or more acceleration data samples,
• ax, ay, and az represent acceleration in three-axes of the accelerometer at time stamp (t), and
• e1, e2 and e3 represent unit vectors pointing in the X-axis, Y-axis and Z-axis respectively;
• estimate a plurality of parameters from the acceleration vector using a parameter estimation technique, wherein the plurality of parameters for an acceleration data sample (i) comprises of a reference vector (R), a theta angle (θi), a phi angle (φi), a normal vector (Li) and a psi angle (Ψi), wherein the reference vector (R) is defined in the Y-axis and the Z-axis rotational plane of the acceleration vector (A) travels, the theta angle (θi) is an angle between the reference vector (R) and the acceleration vector (A), the phi angle (φi) is an angle between the consecutive acceleration vectors, the normal vector (Li) is a vector perpendicular to the plane in which the acceleration vector (A) travels and a psi angle (Ψi) is an angle between the normal vector (Li) and a positive X-axis;
• label each data sample from the plurality of data samples based on the plurality of parameters, via the one or more hardware processors, wherein the labelling comprises one of an inhale phase (in) and an exhale phase (ex), wherein the labelling is identified based on the phi angle (φi) of each of the data samples, where each of the data samples is labelled as the inhale phase (in) when the phi angle (φi) is positive and each of the data samples is labelled as the exhale phase (ex) when the phi angle (φi) is negative;
• identify a plurality of breathing cycles from the plurality of data samples, wherein a breathing cycle from the plurality of breathing cycles is identified from a start of the inhale phase (in) to an end of the exhale phase (ex), wherein the plurality of breathing cycles are represented as a set of continuous breathing cycles Ci represented as a set one inhale and its corresponding exhale phase given by:
• C
• i
• =
• {
• S
• a
• j
• ,
• S
• a
• +
• 1
• j
• ,
• S
• a
• +
• 2
• j
• ,
• …
• ,
• S
• b
• j
• ,
• S
• b
• +
• 1
• j
• ,
• S
• b
• +
• 2
• j
• ,
• …
• ,
• S
• b
• +
• B
• j
• }
• ⁢
• where
• ⁢
• for
• ⁢
• all
• ⁢
• S
• i
• j
• :
• Sii:=in for as i<b, which constitute samples for the inhale phase (in), and
• :=in for bs i<b+B, which constitute samples for the exhale phase (ex),
• wherein breathing session represented as a set of continuous breathing cycles Cj:
• SF={C0,C1,C2, . . . }
• where, SF is a set of breathing cycles with jth breathing cycle represented as Cj,
• a<=i<b,
• wherein a, a+1, a+2 . . . constitute samples for inhale phase,
• b, b+1, b+2 . . . b+B constitute samples for exhale phase;
• and
• analyze in real time the plurality of breathing cycles of the subject, via the one or more hardware processors, by estimating an exhale period and an inhale period, estimating a breathing rate, and determining a type of breathing of the subject in a non-invasive way in real time for identifying and controlling one or more health concerns of the subject and/or early detection of one or more diseases of the subject, wherein the inhale and the exhale period is defined with respect to every breathing cycle Ci, the inhale period and the exhale period is expressed as:
• Inhale
• ⁢
• period
• ⁢
• for
• ⁢
• C
• i
• ,
• I
• i
• =
• t
• b
• -
• 1
• j
• -
• t
• a
• j
• ,
• Exhale
• ⁢
• period
• ⁢
• for
• ⁢
• C
• i
• ,
• E
• i
• =
• t
• b
• -
• B
• j
• -
• t
• b
• j
• ,
• wherein for the breathing cycle Ci, a time stamp for the start of inhale phase is
• t
• a
• j
• ,
• and the time stamp for the start of exhale period is
• t
• b
• j
• ,
• the analyzing process, wherein for the analyzing process, the one or more hardware processors are configured by the instructions to:
• estimate the exhale period and the inhale period for each breathing cycle among the plurality of breathing cycles is based on the time stamp associated with the plurality of data samples identified for the start of the inhale phase to the end of the exhale phase;
• estimate the breathing rate based on the plurality of breathing cycles, wherein the breathing rate is the total number of breathing cycles per minute, wherein a real time rate of breathing is defined with respect to the breathing cycle Ci given by:
• R
• i
• =
• 60
• /
• (
• t
• b
• -
• B
• j
• -
• t
• a
• j
• )
• wherein
• t
• a
• j
• is time stamp for the start of inhale phase,
• t
• b
• j
• is time stamp for the start of exhale phase;
• and
• determine the type of breathing of the subject based on the plurality of parameters, wherein the type of breathing comprises one of a deep/diaphragmatic breathing and a shallow/thoracic breathing, wherein the type of breathing is determined based on proximity of data samples identified along the X-axis based on the normal vector (Li) and the psi angle (Ψi), wherein in inhalation of a deep/diaphragmatic breathing causes the lower part of the torso advance forward more with respect to the upper part leading to the PDA rotated along the positive X-axis and the direction of the normal vector (Li) to be same as the positive X-axis, in exhalation of a deep/diaphragmatic breathing, the direction reverses and the normal vector (Li) becomes parallel to a negative X-axis, in inhalation of a shallow/thoracic breathing, the upper part of the torso advances forward with respect to the lower part of the torso, leading to the PDA rotated along the negative X-axis and the direction of the normal vector (Li) is same as the negative X-axis and in exhalation of shallow/thoracic breathing, the normal vector (Li) is parallel to the positive X-axis.

One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:

• receive an acceleration data of a subject, using the PDA, at a pre-defined sampling rate, wherein the PDA comprises an accelerometer, and wherein the PDA is configured to be in contact with center of torso of the subject, and with a upper half of the PDA lying over the chest and a lower half of the PDA lying over the belly of the subject with a top of the PDA directed towards the chin of the subject, wherein the acceleration data is associated with breathing of the subject and the acceleration data comprises of a plurality of data samples (i) with a time stamp;
• represent the acceleration data as an acceleration vector (A), based on a vectorization technique, wherein the acceleration vector (A) is represented as a three-dimensional space (X-axis, Y-axis and Z-axis) along with the time stamp and is represented as:
• A
• i
• =
• a
• x
• ⁡
• (
• i
• )
• ⁢
• e
• 1
• +
• a
• y
• ⁡
• (
• i
• )
• ⁢
• e
• 2
• +
• a
• z
• ⁡
• (
• i
• )
• ⁢
• e
• 3
• wherein,
• i represents one or more acceleration data samples,
• ax, ay, and az represent acceleration in three-axes of the accelerometer at time stamp (ti), and
• e1, e2 and e3 represent unit vectors pointing in the X-axis, Y-axis and Z-axis respectively;
• estimate a plurality of parameters from the acceleration vector using a parameter estimation technique, wherein the plurality of parameters for an acceleration data sample (i) comprises of a reference vector (R), a theta angle (θi), a phi angle (φi), a normal vector (Li) and a psi angle (Ψi), wherein the reference vector (R) is defined in the Y-axis and the Z-axis rotational plane of the acceleration vector (A) travels, the theta angle (θi) is an angle between the reference vector (R) and the acceleration vector (A), the phi angle (φi) is an angle between the consecutive acceleration vectors, the normal vector (Li) is a vector perpendicular to the plane in which the acceleration vector (A) travels and a psi angle (Ψi) is an angle between the normal vector (Li) and a positive X-axis;
• label each data sample from the plurality of data samples based on the plurality of parameters, via the one or more hardware processors, wherein the labelling comprises one of an inhale phase (in) and an exhale phase (ex), wherein the labelling is identified based on the phi angle (φi) of each of the data samples, where each of the data samples is labelled as the inhale phase (in) when the phi angle (φi) is positive and each of the data samples is labelled as the exhale phase (ex) when the phi angle (φi) is negative;
• identify a plurality of breathing cycles from the plurality of data samples, wherein a breathing cycle from the plurality of breathing cycles is identified from a start of the inhale phase (in) to an end of the exhale phase (ex), wherein the plurality of breathing cycles are represented as a set of continuous breathing cycles Ci represented as a set one inhale and its corresponding exhale phase given by:
• C
• i
• =
• {
• S
• a
• j
• ,
• S
• a
• +
• 1
• j
• ,
• S
• a
• +
• 2
• j
• ,
• …
• ,
• S
• b
• j
• ,
• S
• b
• +
• 1
• j
• ,
• S
• b
• +
• 2
• j
• ,
• …
• ,
• S
• b
• +
• B
• j
• }
• ⁢
• where
• ⁢
• for
• ⁢
• all
• ⁢
• S
• i
• j
• :
• Sji:=in for a≤i<b, which constitute samples for the inhale phase (in), and
• :=in for bs i<b+B, which constitute samples for the exhale phase (ex),
• wherein breathing session represented as a set of continuous breathing cycles Cj:
• SF={C0,C1,C2, . . . }
• where, SF is a set of breathing cycles with jth breathing cycle represented as Cj,
• a<=i<b,
• wherein a, a+1, a+2 . . . constitute samples for inhale phase,
• b, b+1, b+2 . . . b+B constitute samples for exhale phase;
• and
• analyze in real time the plurality of breathing cycles of the subject, via the one or more hardware processors, by estimating an exhale period and an inhale period, estimating a breathing rate, and determining a type of breathing of the subject in a non-invasive way in real time for identifying and controlling one or more health concerns of the subject and/or early detection of one or more diseases of the subject, wherein the inhale and the exhale period is defined with respect to every breathing cycle Ci, the inhale period and the exhale period is expressed as:
• Inhale
• ⁢
• period
• ⁢
• for
• ⁢
• C
• i
• ,
• I
• i
• =
• t
• b
• -
• 1
• j
• -
• t
• a
• j
• ,
• Exhale
• ⁢
• period
• ⁢
• for
• ⁢
• C
• i
• ,
• E
• i
• =
• t
• b
• -
• B
• j
• -
• t
• b
• j
• ,
• wherein for the breathing cycle Ci, a time stamp for the start of inhale phase is
• t
• a
• j
• ,
• and the time stamp for the start of exhale period is
• t
• b
• j
• ,
• wherein for the analyzing process, the one or more hardware processors are configured by the instructions to:
• estimate the exhale period and the inhale period for each breathing cycle among the plurality of breathing cycles is based on the time stamp associated with the plurality of data samples identified for the start of the inhale phase to the end of the exhale phase;
• estimate the breathing rate based on the plurality of breathing cycles, wherein the breathing rate is the total number of breathing cycles per minute, wherein a real time rate of breathing is defined with respect to the breathing cycle Ci given by:
• R
• i
• =
• 60
• /
• (
• t
• b
• -
• B
• j
• -
• t
• a
• j
• )
• wherein
• t
• a
• j
• is time stamp for the start of inhale phase,
• t
• b
• j
• this time stamp for the start of exhale phase;
• and
• determine the type of breathing of the subject based on the plurality of parameters, wherein the type of breathing comprises one of a deep/diaphragmatic breathing and a shallow/thoracic breathing, wherein the type of breathing is determined based on proximity of data samples identified along the X-axis based on the normal vector (Li) and the psi angle (Ψi), wherein in inhalation of a deep/diaphragmatic breathing causes the lower part of the torso advance forward more with respect to the upper part leading to the PDA rotated along the positive X-axis and the direction of the normal vector (Li) to be same as the positive X-axis, in exhalation of a deep/diaphragmatic breathing, the direction reverses and the normal vector (Li) becomes parallel to a negative X-axis, in inhalation of a shallow/thoracic breathing, the upper part of the torso advances forward with respect to the lower part of the torso, leading to the PDA rotated along the negative X-axis and the direction of the normal vector (Li) is same as the negative X-axis and in exhalation of shallow/thoracic breathing, the normal vector (Li) is parallel to the positive X-axis.