The “MYOCYTER” – Convert cellular and cardiac contractions into numbers with ImageJ

Figures

• 

  Figure 1. Workflow and basic functions of MYOCYTER. Panel A represents the typical evaluation of a video using MYOCYTER. Videos are first pre-tested (1.) to determine correct settings (for threshold and particle size) for videos containing multiple cells to be evaluated independently. This can also be skipped if the video contains only a single cell and if the entire frame is evaluated - but for a more precise evaluation with “masking” of the contracting object only, the pre-test is necessary. Subsequently, evaluation is performed (2.). MYOCYTER outputs a large number of different numerical and graphical results, which allow both quality control of the evaluation and a wide range of further statistics. Should a change of the applied user parameters be necessary, re-analysis in real-time is also possible. First, the two main parameters “speed” and “amplitude” are extracted. “Speed” represents the differences between the successive images of a video and indicates the speed of a movement: high speed correlates with high differences between consecutive pictures (see panel B). “Amplitude” represents the difference between every single frame and a “reference image” (either user-defined or…

• 

  Figure 2. Extracted data, dynamic thresholding, systoles, diastoles and peak times. Panel A shows the main graphic output of MYOCYTER. Upper left image: optional animated output, the analyzed cell is animated synchronized to its amplitude. Identified structures (here, neonatal cardiomyocytes) are also indicated on a separate image (upper right), marked by a yellow outline and numbered individually. Furthermore, the amplitude of every single recognized cell/structure is plotted over time (bottom image). Panel B shows a few of the 43 extracted parameters (for details please see the manual, Online Supplement). The amplitude (thick black line) is “subdivided” by four thresholds (horizontal lines, light grey and dashed, representing the thresholds at 10, 20, 50, and 90%), that are re-calculated for every single recognized contraction. They represent the local minima + 10/20/50/90% of the difference between the local minima (dashed vertical spikes) and the next following maxima (continuous vertical spikes). Local (re)adjustment of dynamic thresholding enables “tracking” of even a shifted amplitude (thick black graph). The distance between two subsequent maxima determines the overall contraction…

• 

  Figure 3. Validation of MYOCYTER (MC) and comparison with MUSCLEMOTION (MM). To validate the reliability of MYOCYTER’s (MC) output, videos with known parameters were analyzed and the macro was also compared to the recently published MUSCLEMOTION (MM). Panel A: Analysis of a video inverting from frame to frame returned the expected values for speed (constant at 255) and amplitude (change of 255 from frame to frame). Video: “alternating bw.avi”, Settings (A-E): LowerThreshold = 0; RefFrame = 1. Panel B: Analysis of a black bar moving from left to right on a white background. Amplitude (solid line, difference between reference frame and current frame) increases absolutely linearly. Here, first frame was defined as reference. “Speed” (dotted line) indicates the difference between subsequent images. Video: “moving bar.avi”; Panel C: Same video as in panel B, but in the range from frame 1 to frame 325. After frame 100, the black bar no longer overlaps with its position in the reference image (frame 1), so the change in its linear motion tends linearly towards a limit (reached after 100 frames). Panel D,E: Analysis of a video following a known equation (“declining function.avi”, Online…

• 

  Figure 4. MYOCYTER’s dynamic thresholding and peak recognition. Panel A: Difference between static and dynamic thresholding. A shifted amplitude (red curve) of cardiac contractions is tracked easily by dynamic thresholding, able to adapt to the entire curve also including local maxima of different height. Horizontal solid lines represent the corresponding dynamic thresholds at 10 (black), 20 (blue), 50 (green), and 90% (orange) of the local peak height. Static thresholds, defined at the beginning of the video (dashed horizontal line: at 10%, dots and strokes: at 20% of the first maximum), failed and “lost” the amplitude completely after a short time (static threshold at 10% after 4 seconds, static threshold at 20% after 7 seconds). Vertical dashed lines indicate detected local maxima. Panel B: Influence of the parameter “Detection” on recognition of maxima (continuous vertical spikes) and minima (dotted vertical spikes). The amplitude (thick black curve) is also given by a known time-dependent function (see top figure of this panel). “Detection” determines how large the deviations of local maxima and minima may be from each other to be recognized as such. With higher values for “Detection”…

• 

  Figure 5. Impact of thapsigargin on isolated C57BL/6J-mouse cardiomyocytes. In this figure, the impact of different thapsigargin concentrations on the overall peak times, systoles and diastoles of isolated adult C57BL/6J-cardiomyocytes is depicted (paced measurement). The different panels present the results measured at four different thresholds: 10 (Panel A), 20 (Panel B), 50 (Panel C) and 90% (Panel D). White columns (“0”) indicate the untreated control, “6” (light grey columns) and “12” (dark grey columns), represent the according nanomolar concentrations of thapsigargin applied. The (overall) peak time is the sum of both its systolic and diastolic share of the according threshold. Panel E displays the normalized amplitude of a single contraction for untreated control cells (dotted black line) compared to cells exposed to 6 nM (dashed black line) and 12 nM (continuous black line) thapsigargin, respectively. The four thresholds (10, 20, 50, and 90%) are represented by accordingly labelled horizontal lines, while the vertical dotted line represents the detected local maximum. Statistics: significant differences are indicated by connecting lines; *(p < 0.05), **(p < 0.01), #(p < 0.005),…

• 

  Figure 6. Influence of the frame rate on the amplitude during evaluation with MYOCYTER. This figure shows the effects of the frame rate on the final plot of the amplitude in an evaluation with MYOCYTER. A representative cardiomyocyte (isolated from a C57BL/6J-mouse heart) contracts about 3 times per second, so at 120 fps there are about 40 data points/frames per contraction available (panel A). If the frame rate of the video is reduced (halving from panel B–F), correspondingly fewer measuring points are available describing the shape of a contraction. The shape of the curve loses detail until a complete contraction is only represented by a single measurement point (panel E). In panel F, the frame rate is so low that the information of complete contractions is missing.