1. A system 100 for myocardial performance determination, the system 100 comprising:
a memory 126, 208 storing program instructions;
a processor 124, 206 configured to execute the instructions stored in the memory 126, 208; and
a computation engine 122, 204 executed by the processor 124, 206 and configured to:
generate a first dataset representing a set of events associated with a pre-defined parameter of a biomarker extracted from physiological parameters of a subject, wherein the set of events is determined by processing the pre-defined parameter at a first level and a second level of a multi-level artificial neural network architecture recursively for a pre¬defined number of times;
generate a second dataset representing characteristics associated with the set of events by processing the first dataset at a third level and a fourth level of the multi-level artificial neural network architecture;
compute a set of values associated with the set of events by processing the second dataset at a fifth level of the multi-level artificial neural network architecture; and
compute a myocardial performance index based on the set of values, wherein the myocardial

performance index is representative of the myocardial performance of the subject.
2. The system 100 as claimed in claim 1, wherein the computation engine 122, 204 extracts the biomarker from the micro-voltage digital datasets by detecting the pre¬defined parameter as a time-domain parameter; sand converting the detected time-domain parameter to a frequency-domain parameter.
3. The system 100 as claimed in claim 1, wherein the biomarker is representative of a heart rate present in micro-voltage digital datasets.
4. The system 100 as claimed in claim 1, wherein the computation engine 122, 204 comprises a filtering unit 214 executed by the processor 124, 206 and configured to extract the biomarker employing a bandpass frequency in a range of between 5 Hz and 15 Hz.
5. The system 100 as claimed in claim 1, wherein the pre¬defined parameter associated with the biomarker is representative of time-domain heartbeat signals.
6. The system 100 as claimed in claim 2, wherein the computation engine 122, 204 comprises a heartbeat extraction unit 216 executed by the processor 124, 206 and configured to convert the time-domain parameter to the frequency-domain parameter by dividing the time-domain parameter into short time-intervals prior to conversion of the time-domain parameter into an image representation in the frequency-domain, and wherein the time-interval is 5 seconds.
7. The system 100 as claimed in claim 6, wherein the image representation is representative of a spectrogram, and wherein the spectrogram represents the pre-defined parameter in the form of specific color variations.

8. The system 100 as claimed in claim 2, wherein the time-domain parameter is converted to the frequency-domain parameter by the heartbeat extraction unit 216 based on a Fourier Transform technique, and wherein the Fourier Transform technique is a Short Term Fourier Transform (STFT) technique.
9. The system 100 as claimed in claim 2, wherein the computation engine 122, 204 comprises a heartbeat processing unit 218 executed by the processor 124, 206 and configured to determine the set of events by:

a) processing the frequency-domain parameter at the first level of the multi-level artificial neural network architecture to extract features associated with the set of events, wherein the first level is based on a time distributed convolutional 2-dimensional (2D) neural network level;
b) processing the features at the second level of the multi-level artificial neural network architecture, wherein the second level is based on a time distributed Max Pooling 2D neural network level for downsampling the features;
c) repeating the steps a and b recursively at least three times to generate the first dataset in the form of a multi-dimensional data;
d) processing the multi-dimensional first dataset at the third level of the multi-level artificial neural network architecture, wherein the third level is based on a flattening neural network level for converting the multi-dimensional data into a 1D tensor; and

e) processing the 1D tensor at the fourth level of the multi-level artificial neural network architecture, wherein the fourth level is based on a bidirectional Long Short Term Memory (Bi-LSTM) neural network level for generating the second dataset.
10. The system 100 as claimed in claim 1, wherein the fifth level is based on a time distributed dense neural network level.
11. The system 100 as claimed in claim 1, wherein the set of values are computed by consecutively inserting multiple two concentric rectangular bounding boxes around the pre-determined parameter associated with the biomarker.
12. The system 100 as claimed in claim 1, wherein the set of values represents time based features of the pre-defined parameter including isovolumetric contraction time (IVCT) in the range of between 20ms and 70ms, isovolumetric relaxation time (IVRT) in the range of between 50ms and 90ms, left ventricular ejection time (LVET) in the range of between 150ms and 350ms and mitral closing to opening time (MCOT).
13. The system 100 as claimed in claim 1, wherein the computation engine 122, 204 comprises a myocardial performance computation unit 220 executed by the processor 124, 206 and configured to compute the myocardial performance index based on the second set of values.
14. A method for myocardial performance determination, the method comprising:
generating, by a processor 124, 206, a first dataset representing a set of events associated with a pre-defined parameter of a biomarker extracted from

physiological parameters of a subject, wherein the set of events is determined by processing the pre-defined parameter at a first level and a second level of a multi-level artificial neural network architecture recursively for a pre-defined number of times;
generating, by the processor 124, 206, a second dataset representing characteristics associated with the set of events by processing the first dataset at a third level and a fourth level of the multi-level artificial neural network architecture;
computing, by the processor 124, 206, a set of values associated with the set of events by processing the second dataset at a fifth level of the multi-level artificial neural network architecture; and
computing, by the processor 124, 206, a myocardial performance index based on the set of values, wherein the myocardial performance index is representative of the myocardial performance of the subject.
15. The method as claimed in claim 14, wherein the processor 124, 206 extracts the biomarker from micro-voltage digital datasets by detecting the pre-defined parameter as a time-domain parameter; and converting the detected time-domain parameter associated with the biomarker to a frequency-domain parameter.
16. The method as claimed in claim 14, wherein the biomarker is representative of a heart rate present in micro-voltage digital datasets.
17. The method as claimed in claim 14, wherein the biomarker is extracted based on a bandpass frequency in the range of between 5 Hz and 15 Hz.

18. The method as claimed in claim 14, wherein the pre¬defined parameter associated with the biomarker is representative of time-domain heartbeat signals.
19. The method as claimed in claim 15, wherein the time-domain parameter is converted to the frequency-domain parameter by dividing the time-domain parameter into short time-intervals prior to conversion of the time-domain parameter into an image representation in the frequency-domain, and wherein the time-interval is 5 seconds.
20. The method as claimed in claim 19, wherein the image representation is representative of a spectrogram, and wherein the spectrogram is representative of the pre¬defined parameter in the form of specific color variations.
21. The method as claimed in claim 15, wherein the time-domain parameter is converted to the frequency-domain parameter based on a Fourier Transform technique, and wherein the Fourier Transform technique is a Short Term Fourier Transform (STFT) technique.
22. The method as claimed in claim 15, wherein the set of events are determined by:

a) processing the frequency-domain parameter at the first level of the multi-level artificial neural network architecture to extract features associated with the set of events, wherein the first level is based on a time distributed convolutional 2-dimensional (2D) neural network level;
b) processing the features at the second level of the multi-level artificial neural network architecture, wherein the second level is based on a time

distributed Max Pooling 2D neural network level for downsampling the features;
c) repeating the steps a and b recursively at least three times to generate the first dataset in the form of a multi-dimensional data;
d) processing the multi-dimensional first dataset at the third level of the multi-level artificial neural network architecture, wherein the third level is based on a flattening neural network level for converting the multi-dimensional data into a 1D tensor; and
e) processing the 1D tensor at the fourth level of the multi-level artificial neural network architecture, wherein the fourth level is based on a bidirectional Long Short Term Memory (Bi-LSTM) neural network level for generating the second dataset.

23. The method as claimed in claim 14, wherein the fifth level is based on a time distributed dense neural network level.
24. The method as claimed in claim 14, wherein set of values are computed by consecutively inserting multiple two concentric rectangular bounding boxes around the pre-determined parameter associated with the biomarker.
25. The method as claimed in claim 14, wherein the set of values represents time based features of the parameter including isovolumetric contraction time (IVCT) in the range of between 20ms and 70ms, isovolumetric relaxation time (IVRT) in the range of between 50ms and 90ms, left ventricular ejection time (LVET) in the range of between 150ms and 350ms and mitral closing to opening time (MCOT).