Limitations and Future Directions
- Incorporate qualitative methods including Hilbert Transform phase mapping and dominant frequency analysis, and explore additional algorithms to have a more comprehensive model with better quantification accuracy.
- Increase number of training data used for learning the model to avoid overfitting and decrease training and testing errors.
- Consult trained physicians for manual inspection of training data.
- Model should be tested by different labs to ensure consistency and accuracy in the evaluation of cardiac arrhythmias.
Intellectual Property
- The project is built on the previous work of members in Efimov Lab at Washington University in St. Louis. The code of the original Graphical User Interface is open source at https://code.google.com/p/rhythm-analysis-software/ and does not have a traditional intellectual property. Although modification of the software requires the consent from Dr. Igor Efimov, this uncopyrighted, freely reproducible software can be used by any lab as long as the review paper describing proper analysis of optical mapping data is referenced.
- Algorithms are adopted from publications from Lammers W.J. et al., Robert M. Haralick, and Rose C.J. et al. The methodologies described in these papers are not copyrighted and therefore the implementation of the ideas does not carry any intellectual property.
- No marketing involved. Software for lab and research use only.
Ethical Considerations
- Optical Maps are obtained from the client. All experiments were approved by the Institutional Animal Care and Use Committee of Washington University in St. Louis.