Could you elaborate on what exactly you're looking for?
Are you trying to develop a custom algorithm for your dataset or are you trying to implement something from the existing literature?
You do have ECGs and want to predict emotion. You are capable to produce HRV but you will find out that it is not much useful type of data distilled from the original ECG. Original biosignal is much richer and this richness must be addressed.
I strongly recommend you to read papers that are
- predicting the depth of anesthesia from ECGs
- classify gait from gait data
- predict epileptic seizures from EEGs
- predict arrhythmias from ECGs
- classify sleep alone from ECGs
All of them are based on the application of entropy measures and many on the application of machine learning techniques (not only neuronal networks).
Your chose study are is very interesting but requires a deep knowledge of entropic measures and machine learning.
****
Those who want to know more: Some of them are cited and the general overview of complex systems in biology and medicine is provided in the review paper
https://www.researchgate.net/publication/330546521_Complex_Systems_and_Their_Use_in_Medicine_Concepts_Methods_and_Bio-Medical_Applications
The changes in ECG related to emotional changes are too non specific to draw any meaningful conclusions as a "stand alone" modality. Basically it is related to the changes in the sympathetic and parasympathetic activity. However, it could perhaps be useful if it is combined with more reliable and predictable investigations.
ECG changes related to emotions are nonspecific and may be attributed to other factors. Emotional disorders causing hypersympathetic activity may present with changes related to the increased adrenergic activity such as tachycardia, which may be true for other causes such as thyrotoxicosis. Medicines given for psychologic problems such as anti-depressants may cause nonspecific ECG changes, which are drug related rather than emotion-related.
I would like to revisit this discussion with emphasizing of the knowledge of complex systems and how they are used in biology and medicine. Exactly for this purpose, I did prepare a review paper that saves you about a decade or even more of reading and practicing of complex systems research.
The knowledge covered there is designed in the way that everyone can easily dive into depths of complexity without being lost in details and technicalities.
Tell me whether you find it useful as an introduction to the subject. There might come a continuation of this paper focused on the prediction of biosignals what is actually my main current research area.
The review paper:
https://www.researchgate.net/publication/330546521_Complex_Systems_and_Their_Use_in_Medicine_Concepts_Methods_and_Bio-Medical_Applications
You will have sensitivity... Specificity will be difficult :-)
- Badges
- Science topic