S in ventricular cross-sectional area to estimates of ventricular volume over the beat cycle.Automated In Vivo Hypercholesterolemia ScreenFigure 4. Waveform Analysis Methodologies. Volume change over time (top) calculated from area change as outlined in figure 3. 
Briefl, area waveform values were input into the equation, C = (6.861024) * A + 46 from which volume over the heartbeat was calculated according to the equation V = (4/3)**A*C where A is the area of the ventricle during the beat cycle and C is the radius in the Z-direction. A. In the Fourier framework (left), a waveform is transformed to Fourier space in order to extract the amplitude and frequency (f) of the wave. In this case, these values represent K of the stroke volume (SV) and theheart rate (HR) respectively. From these parameters, we calculate cardiac output (CO) and ejection fraction (EF). A representative waveform with average diastolic and systolic volumes as calculated by Fourier is presented (bottom left). Notice that thedistance between diastole and systole compared to segmentation approach B. In th segmentation approach (right), the GHRH (1-29) original waveform is transformed to Fourier space. The frequency of the peak of the transform is extracted to determine the period (T) of the waveform which is then utilized as a baseline value on which to base the size of segment for analysis. The algorithm measures maximum and minimum values within each segmen (which is sized at 1.16T in order to increase the liklihood of capturing the maximum 
and minimum values) traversing the waveform. Stroke volume is calculated as the mean maximum value ?mean minimum value and is SMER 28 represented as average diastole and average systole (bottom right). doi:10.1371/journal.pone.0052409.gCombined with our initial hypercholesterolemia screen, this automated detection procedure further streamlines the drugdiscovery and toxicity testing process. In order to demonstrate the utility of this methodology, we tested the influence of a dose of MHE that was effective in our hypercholesterolemia treatment screen (6.5 mg/mL) on cardiodynamics and analyzed the data using both of the automated methods. According to both analysis paradigms, the results indicate an increase in SV and EF in hawthorn treated fish compared to untreated controls, indicating enhanced cardiac function after hawthorn treatment (figure 5).DiscussionThe purpose of this study was to create a platform in which to rapidly test functional food-based treatments of disease. Thisplatform can also test single-molecule treatments of disease as evidenced by the results of ezetimibe treatment (figure 3B). The initial hypercholesterolemia screen concentrated on a simple output metric: mean fluorescence intensity as judged from the entirety of images collected in each well. The simplicity of this measurement procedure allows our methodology to be applicable to many other confocal systems, and many other image analysis programs. This simplicity also decreases the computational demand of image analysis. The large data sets generated in high-throughput/high-content screening place a large burden on most computer systems. Much more demanding however, is complicated data analysis. Our system allows researchers to initially screen and detect changes with simple analysis, building on this analysis if necessary.Automated In Vivo Hypercholesterolemia ScreenFigure 5. Cardiodynamic Influence of Methanolic Hawthorn Extract (MHE). A. Fourier transformed data of control a.S in ventricular cross-sectional area to estimates of ventricular volume over the beat cycle.Automated In Vivo Hypercholesterolemia ScreenFigure 4. Waveform Analysis Methodologies. Volume change over time (top) calculated from area change as outlined in figure 3. Briefl, area waveform values were input into the equation, C = (6.861024) * A + 46 from which volume over the heartbeat was calculated according to the equation V = (4/3)**A*C where A is the area of the ventricle during the beat cycle and C is the radius in the Z-direction. A. In the Fourier framework (left), a waveform is transformed to Fourier space in order to extract the amplitude and frequency (f) of the wave. In this case, these values represent K of the stroke volume (SV) and theheart rate (HR) respectively. From these parameters, we calculate cardiac output (CO) and ejection fraction (EF). A representative waveform with average diastolic and systolic volumes as calculated by Fourier is presented (bottom left). Notice that thedistance between diastole and systole compared to segmentation approach B. In th segmentation approach (right), the original waveform is transformed to Fourier space. The frequency of the peak of the transform is extracted to determine the period (T) of the waveform which is then utilized as a baseline value on which to base the size of segment for analysis. The algorithm measures maximum and minimum values within each segmen (which is sized at 1.16T in order to increase the liklihood of capturing the maximum and minimum values) traversing the waveform. Stroke volume is calculated as the mean maximum value ?mean minimum value and is represented as average diastole and average systole (bottom right). doi:10.1371/journal.pone.0052409.gCombined with our initial hypercholesterolemia screen, this automated detection procedure further streamlines the drugdiscovery and toxicity testing process. In order to demonstrate the utility of this methodology, we tested the influence of a dose of MHE that was effective in our hypercholesterolemia treatment screen (6.5 mg/mL) on cardiodynamics and analyzed the data using both of the automated methods. According to both analysis paradigms, the results indicate an increase in SV and EF in hawthorn treated fish compared to untreated controls, indicating enhanced cardiac function after hawthorn treatment (figure 5).DiscussionThe purpose of this study was to create a platform in which to rapidly test functional food-based treatments of disease. Thisplatform can also test single-molecule treatments of disease as evidenced by the results of ezetimibe treatment (figure 3B). The initial hypercholesterolemia screen concentrated on a simple output metric: mean fluorescence intensity as judged from the entirety of images collected in each well. The simplicity of this measurement procedure allows our methodology to be applicable to many other confocal systems, and many other image analysis programs. This simplicity also decreases the computational demand of image analysis. The large data sets generated in high-throughput/high-content screening place a large burden on most computer systems. Much more demanding however, is complicated data analysis. Our system allows researchers to initially screen and detect changes with simple analysis, building on this analysis if necessary.Automated In Vivo Hypercholesterolemia ScreenFigure 5. Cardiodynamic Influence of Methanolic Hawthorn Extract (MHE). A. Fourier transformed data of control a.