Semi-automated Optical Heartbeat Analysis of Small Hearts

Hi, I'm Karen er. Hi, I'm Anthony Kado and we're part of the Development and Aging program at the Burnham Institute for Medical Research. We're going to demonstrate our semi-automated heartbeat analysis program for understanding and quantifying heartbeat parameters in model systems with small hearts.

This is a high speed movie of a beating fruit fly heart. Information about heart morphology and contractility can be obtained by measuring heart diameters during contraction and relaxation cycles. To do this, the edges of the heart need to be identified and marked during maximum diastole and cyst.

In the pre-process window, select the movie or movies you wish to look at. We will select the movie you just saw of a heart from a young fly. Optical recordings can be advanced at slow speeds and even one frame at a time to allow precise identification of the frames, where maximal contraction and relaxation of the heart occur.

Make one pair of marks to identify these edges. Advance the movie to a frame where the heart is maximally contracted. Make another pair of marks to identify the systolic edges.

Marks can be made in duplicate at different horizontal locations. Along the heart, the heart diameters obtained here will be used to calculate the percent fractional shortening, which provides an estimate of the contractility of the heart. This measure represents the extent to which the heart edges move toward each other during a contraction.

The percent fractional shortening is calculated as the difference between the diastolic and systolic diameters divided by the diastolic diameter, and that fraction is multiplied by 100. For this wild type fly, the percent fractional shortening is 45%Here is an example of a heart from a fly that has a mutation in the muscle protein myosin compared to the wild type fly heart you just saw. This heart is visibly narrower.

In addition, it doesn't contract as much as the wild type heart. It should be clear from the positions of the marked edges that this heart also cannot relax, as well as the heart from the young wild type fly. The percent fractional shortening that is calculated from the diameters measured here is only 15%a dramatically smaller value than we obtained for the wild type fly heart.

The points identified in this module are used only for diameter and fractional shortening measurements and are not used for movement detection. Movement detection is done automatically by the program and uses two different algorithms to analyze movement in each movie. The first algorithm is the average frame darkness algorithm, which uses a frame by frame approach.

The overall darkness intensity is calculated for an entire movie frame. The values are normalized to an interval between zero and one, and values are plotted for every frame in the movie. The second method of movement detection is the pixel by pixel algorithm, which calculates the darkness intensity for every pixel in a movie frame, and then compares those values with those in the next frame.

This is the output for a movie of a heart from a young one week old fly. Although movement detection is performed automatically, the output must be checked for accuracy. The output from each of the two movement detection algorithms is displayed in this module with the average darkness output displayed above the pixel by pixel output window.

In order to be sure that the detection intervals agree with the movement seen in the actual movie, the program displays an edge trace or M mode made from the frames being analyzed directly below the algorithm output windows. The M mode is made by digitally cutting slices one pixel wide through the heart from each frame in the movie, and aligning them horizontally to provide a snapshot of the heart edge movements over time. In the pixel by pixel window, the identified diastolic intervals are indicated by a horizontal green line with the number of frames contributing to that interval.

Directly above systolic intervals are identified as the interval between two successive diastole. The beginning of a systole is indicated in the M mode by a vertical blue line, and the end is indicated by a vertical red line. With this display, it is relatively easy to see if the algorithm output agrees with the actual heart movements.

For this heart, the contraction intervals identified by the program agree with the contraction seen in the M mode trace, so this output is accepted. Here is the output from a movie taken of a heart from a relatively old fly. The output from the pixel by pixel algorithm does not agree with the contraction pattern seen in the M mode.

This is because the heart contractions are more sustained than in younger flies. The algorithm detects both the contraction movement as well as the relaxation movement and interprets these two movements as separate beats. In this case, we can add information from the average frame darkness algorithm by clicking on the used darkness checkbox.

This permits the program to identify the entire contraction interval as a single event using information from the average frame darkness algorithm correctly, identify some but not all of the contraction events. This is because an underlying low frequency fluctuation in darkness places some of the contraction peaks outside the window used to identify peaks. This slow wave can be eliminated by use of a built-in high pass filter.

Once all the peaks are within the window, the program correctly identifies all of the contraction intervals. At this point, you can accept the algorithm output by clicking on data.Okay. If after checking the used darkness box and applying filters, the algorithm output cannot be reconciled with the actual heart function as displayed in the M mode.

Be sure to click on discard data. This prevents erroneous numbers from being included in your data set. Once the output for a given movie has been accepted, the program can be instructed to use that information to automatically calculate a number of parameters.

Statistical output is provided for each movie individually, and this data is located in the FHM statistics folder. In the movie file. Subdirectory statistics for the entire data are also generated, and this information is located in the FHM overview folder.

This information is provided in the form of a comma separated value file called overview csv. This file can be opened in most spreadsheet programs such as Microsoft Excel. The output statistics include the heart rate, heart period diastolic and systolic intervals, diameters of the heart during diastolic and systole, and the percent fractional shortening.

The program has an update histogram module, which will display all of the data on heart intervals for a dataset in histogram format. Because heart rates tend to vary from animal to animal, the data is normalized. Since the distribution of the heart period intervals is not symmetric around the mean, we use the median interval value for the normalization.

The values for all heartbeat intervals in the dataset are also provided as a TXT file that can be imported into a spreadsheet. This capability proved very useful in demonstrating differences in rhythmicity between wild type fly hearts and hearts from flies that had mutations in a potassium channel. The distribution of all the measured intervals was very dispersed in the mutants compared to controls.

This is one of several ways we have attempted to quantify heart arrhythmic in our program. The M mode module generates an edge trace, also called an M mode that shows the movement of the heart edges over time. Here we show five second M mode traces from a wild type fruit fly heart two fly hearts with mutations in the muscle protein myosin and a zebrafish larval heart.

M modes are useful to show qualitatively what the contraction pattern of a heart is. Over time. Here, an M mode is being generated from a movie taken of a heart.

In a developing zebrafish larvae, we can change the line of pixels that will be used to produce the M mode by clicking on different locations in the frame. A preview of the M mode at that location is then displayed. Once a suitable location is identified, you can choose how many of the frames you want to include or simply choose the length of time.

Here we will select 10 seconds. The M Mode is produced as a TIFF file located in the FHMM mode subdirectory. The movement movie module automatically produces a slowed 22nd version of any analyzed movie showing all the pixels that are identified as changing in red.

This feature is primarily used for illustration purposes. These movies are located in the FHM Movement movie Subdirectory. This is the movement movie made from the original young fly heart that we used in the beginning of this video.

This analytical system can be applied to high speed movies of other models with small hearts such as the embryonic mouse heart. This is a movie of a heart from an eight day old mouse embryo. The small size makes it difficult to analyze using traditional methodologies with our methodology.

The same parameters that were quantified for the fly heart can also be obtained for the mouse heart. We've just shown you how to use our semi-automated optical heart analysis program. Remember, when doing this analysis to click on discard data if the program doesn't accurately identify all of your heartbeats in your movie, but for those that are accurately identified, the program will calculate parameters for every beat in the movie, and we'll provide you with a wealth of information concerning the heart function in your model system.

So that's it. Thanks for watching. Good luck with your analysis.