Further Work and Conclusion Essay Example

Further Work and Conclusion Paper Additional directions for this work include refining and extending our cardiac study with a view to clinical implementation. Furthermore, we suggest that rhythmic signals arising from other biological systems may have application for the techniques described in this paper. An investigation of the optimal windowing parameter set would be instructive since our findings suggest the existence of physiological thresholds in the spectral entropy level and variance that are applicable to a variety of patients. As noted at the end of Section 5.4.3, one challenge would be to investigate and improve the utility of the measure (alone or combining methods) when applied to patients that demonstrate a mix of different pathologies and arrhythmias. Adjusting the spectral entropy window to covary with instantaneous heart rate so that ÃŽ ± always contains ten beats exactly would further reduce issues related to variations in the heart rate. Extending the algorithm to include other dimensions in the disorder map (e.g., heart rate) will likely improve the accuracy of results and may increase the number of arrhythmias the spectral entropy can distinguish between. We will write a custom essay sample on Further Work and Conclusion specifically for you for only $16.38 $13.9/page Order now We will write a custom essay sample on Further Work and Conclusion specifically for you FOR ONLY $16.38 $13.9/page Hire Writer We will write a custom essay sample on Further Work and Conclusion specifically for you FOR ONLY $16.38 $13.9/page Hire Writer An accurate automatic detector of atrial fibrillation would be clinically useful in monitoring for relapse of fibrillation in patients and in assessing the efficacy of antiarrhythmic drugs (Israel 2004). An implementation integrated with an ambulatory ECG or heart rate monitor would be useful in improving the understanding of arrhythmias on time scales longer than that available using conventional ECG analysis techniques alone. Measures of disorder in the frequency domain have practical significance in a range of biological signals. For example, the regularity of the background electroencephalography (EEG is the measurement of electrical activity produced by the brain as recorded from electrodes placed on the scalp) alters with developmental and psychophysiological factors: some mental or motor tasks cause localized desynchronization; in addition, the background becomes more irregular in some neurological and psychiatric disorders (see Inouye et al. 1991; Rosso 2007 and references therein). The spectral entropy method and the concept of the disorder map described in this paper are not cardiac specific: it would be instructive to adapt these ideas to other rhythmic signals where a rapid detection of arrhythmia would be informative. 5.6 Conclusion In this paper we have presented an automatic arrhythmia detection algorithm that is able to rapidly detect the presence of atrial fibrillation using only the time series of patients’ beats. The algorithm employs a general technique for quickly quantifying disorder in high-frequency event series: the spectral entropy is a measure of disorder applied to the power spectrum of periods of time series data. The physiologically motivated use of the spectral entropy is shown to distinguish atrial fibrillation and flutter from other rhythms. For a given set of parameters, we are able to determine from a disorder map two threshold conditions (based on the level and variance of spectral entropy values) that enable the detection of fibrillation in a variety of patients. We apply the algorithm to the MIT-BIH atrial fibrillation database of 25 patients. When the algorithm is set to identify abnormal rhythms within 6 s it agrees with 85.7% of the annotations of professional rhythm assessors; for a response time of 30 s this becomes 89.5%, and with 60 s it is 90.3%. The algorithm provides a rapid way to detect fibrillation, demonstrating usable response times as low as 6 s and may complement other detection techniques. There also exists the potential for our spectral entropy and disorder map implementations to be adapted for the rapid identification of disorder in other rhythmic signals. 5.7 Appendix This appendix contains images of the electrocardiograms referred to in Sections 5.4.1 and 5.4.2. They represent examples where we believe the annotations provided as part of the MIT-BIH atrial fibrillation database to be incorrect, and where rhythms other than atrial fibrillation and atrial flutter are present in patient electrocardiograms. The following figures were obtained using the Chart-O-Matic facility on the physionet website (Goldberger et al. 2000) for patients comprising the MIT-BIH atrial fibrillation database (afdb). We give selected example electrocardiograms (ECGs) to illustrate the point under consideration and stress that there are additional times that could have been used for demonstrative purposes. The rhythm assessments to which we are comparing are provided as annotations included as part of the afdb. For other examples of ECGs corresponding to the rhythms given here, see Bennett 2002. 5.7.1 Disagreements with annotations Rhythm assessments have been questioned before (Tateno Glass 2000, 2001); here we give explicit examples from the afdb where we believe the ECG suggests a rhythm different from that given by the annotation. The figures and ideas in this section pertain to Section 5.4.1 of the main text. Instances where atrial fibrillation has been missed in annotations We observe in Patients 08219 (Figure 5.4) and 08434 (Figure 5.5) periods of atrial fibrillation that we believe to have been missed in the annotations but are correctly identified by our detection algorithm. Cases such as these serve to negatively impact the results of the algorithm unfairly; however, we note that such instances comprise a small proportion of the afdb. Instances where atrial flutter has been missed in annotations Atrial flutter may have been misannotated in Patients 04936 (Figure 5.6) and 08219 (Figure 5.7). This unreliability of rhythm assessment, compounded with the limited number of periods of atrial flutter in the database, prevents us from drawing meaningful quantitative conclusions regarding the success of the detection algorithm in identifying flutter. Despite this, we believe that the spectral entropy is in principle still capable of identifying flutter. 5.7.2 Other rhythms The unreliability of parts of the annotations still does not account for all false predictions produced by the detection algorithm. We suggest the presence of other rhythms within the afdb to be an additional factor that needs to be considered. The figures and ideas in this section pertain to Section 5.4.2 of the main text. Instances of fib-flutter Fib-flutter denotes periods where the rhythm transitions in quick succession between atrial fibrillation and flutter (Horvath et al. 2000). Such behavior naturally causes the variance to increase (thereby exceeding the standard deviation threshold in the algorithm for classification as atrial fibrillation) and one might question whether it is still appropriate to classify those periods as standard atrial fibrillation. We identify in the ECG of Patient 04936 periods of fib-flutter which likely accounts for the high proportion of false negative results (Figures 5.8 and 5.9). Instances of sinus arrest Sinus arrest occurs when the sinoatrial node fails to fire, resulting in increased irregularity of the heart rhythm, whilst still retaining QRS complexes indicative of normal sinus rhythm; this condition (along with sinus arrhythmia) is likely responsible for the high proportion of false positives seen in Patient 05091 (Figure 5.10). Conclusion The natural world makes no distinction between scientific disciplines. Increasingly, answers to scientific questions lie at the intersection of traditional disciplines. This thesis has applied techniques developed in physics and mathematics to problems in ecology and medicine. I have shown how simple methods of time series analysis can enable rapid detection of cardiac arrhythmia; how ecosystems may respond and adapt to the loss of species; how species can modify their feeding interactions in manmodified environments; and how spatial landscape can affect the spread of fluctuations of venture capital firm populations. Moving forward, my current research is motivated by one fundamental question: What does a food web represent? Practically, we must ask: (i) What ecological mechanisms underlie foodweb structure? and (ii) How does food-web structure change through time? What ecological mechanisms underlie food-web structure? How does individual-level species behaviour lead to observed food-web structure (Stouffer 2010)? How does behaviour underlie differences (or similarities) among distinct food-web types (Th ´ebault Fontaine 2010)? Can we combine these distinct food-web types to understand complete patterns of interactions within ecological communities (Lafferty et al. 2008)? How do environmental factors affect species interactions (Lalibert ´e Tylianakis 2010)? Exciting work has begun to address these questions. Answers to these questions will lead to new questions. Progress relies on the exchange of ideas, many of which will originate in fields other than ecology. How does food-web structure change through time? What assembly mechanisms lead to observed food-web structure (Piechnik et al. 2008)? Are current, static, models consistent with empirical data on food-web assembly (Albrecht et al. 2010)? What role do invasive species play in food-web dynamics (Lopezaraiza-Mikel et al. 2007)? Experiment and theory are advancing. Ecological data are improving and more sophisticated methods of analysis are developing. We are increasingly achieving that fundamental goal of ecology—a central tenet shared by the physical sciences—prediction. Previous Page   Algorithm and Discussion