Introduction: Beat detection is a key step in the analysis of photoplethysmogram (PPG) signals. The MSPTD algorithm was recently identified as one of the most accurate beat detection algorithms, but its current open-source implementation is substantially more computationally expensive than other leading algorithms such as qppgfast. The aim of this work was to develop a more efficient, open-source implementation of the MSPTD algorithm.
Methods: Five potential improvements were identified to increase the efficiency of MSPTD. Each potential improvement was evaluated in turn, and an optimal algorithm configuration named MSPTDfast was developed, incorporating all of the improvements found to reduce algorithm execution time whilst not substantially reducing the accuracy of beat detection. The performance of MSPTDfast was assessed on the PPG-DaLiA dataset, which contains wrist PPG signals alongside reference ECG signals from 15 subjects.
Results: MSPTDfast was found to be substantially more efficient than MSPTD (a reduction in execution time of 64.4%), with minimal difference in beat detection accuracy (F1-score 87.8% vs. 87.7%). In addition, the performance of MSPTDfast was much closer to that of the state-of-the-art qppgfast algorithm than the MSPTD algorithm, with a comparable F1-score (87.4% vs. 87.7%), and an execution time which was only 30.0% longer than that of qppgfast (vs. 257.4% longer for MSPTD).
Conclusion: MSPTDfast is an efficient and accurate open-source PPG beat detection algorithm with a substantially faster execution time than MSPTD. It is available under the permissive MIT licence.