On the Design of an Efficient Cardiac Health Monitoring System Through Combined Analysis of ECG and SCG Signals

. 2018 Jan 28;18(2):379. doi: 10.3390/s18020379

Algorithm 3: Abnormal ECG morphology detection

Input:

V_{e c g} :

Vector of ECG data points with amplitude,

S_{r} :

ECG data sampling rate,

H_{r} :

Heart rate,

R F V :

Set of referenced feature values,

k :

User defined constant to group k-cycles.
Output: Detection of abnormal morphologies
Notations:

θ X_{w v}

: Measured amplitude of X wave,

ψ X_{w v}

: Measured duration of X wave,

c X_{w v}

: Counter for X wave abnormal morphologies, where

X \in {P, Q R S, T}

c R R_{i n v}

: Counter for

R R

interval abnormal morphologies,

ψ R R_{i n v}

: Measured duration of

R R

interval.
1 Initialize k;
2 Initialize

c P_{w v} = c Q R S_{w v} = c T_{w v} = c R R_{i n v} = 0

;
3 Estimate Cardiac cycle Length:

C L = \frac{S_{r}}{H_{r}}

;
4 Calculate # of cardiac cycles:

C C s = \frac{| V_{e c g} |}{C L}

;
5 for $i = 1$ to $C C s - k + 1$ do
6 j=i;
7 while j≤ $(i + k)$ do
8 if

(θ P_{w v} > Ω P_{w v}) \lor (θ P_{w v} < ω P_{w v}) \lor (ψ P_{w v} > Δ P_{w v}) \lor (ψ P_{w v} < δ P_{w v})

then
9

c P_{w v} = c P_{w v} + 1

; // P wave abnormal morphology detection
10 end
11 ;
12 if

(θ Q R S_{w v} > Ω Q R S_{w v}) \lor (θ Q R S_{w v} < ω Q R S_{w v}) \lor (ψ Q R S_{w v} > Δ Q R S_{w v}) \lor (ψ Q R S_{w v} < δ P_{w v})

then
13

c Q R S_{w v} = c Q R S_{w v} + 1

; //

Q R S

wave abnormal morphology detection;
14 end
15 if

(θ T_{w v} > Ω T_{w v}) \lor (θ T_{w v} < ω T_{w v}) \lor (ψ T_{w v} > Δ T_{w v}) \lor (ψ T_{w v} < δ T_{w v})

then
16

c T_{w v} = c T_{w v} + 1

; // T wave abnormal morphology detection;
17 end
18 if

(ψ R R_{i n v} > Δ R R_{i n v}) \lor (ψ R R_{i n v} < δ R R_{i n v})

then
19

c R R_{i n v} = c R R_{i n v} + 1

; //

R R_{i n v}

interval abnormal morphology detection;
20 end
21 end
22 endfor