TRACKING MARKERS WITH MISSING DATA BY LOWER RANK APPROXIMATION

Motion and deformation of an object may be quantified by following att ached markers in video or cine frame sequences. When recording cardiac motion by video (256 x 256 pixels, 50 Hz), generally no more than app roximately 20 markers can be followed due to difficulties in proper id entification of marker images. In the present study we developed the l ower rank (LR) tracking method which can automatically follow consider ably more than 20 markers. The performance of the method was evaluated in computer simulations of naturally moving myocardial markers observ ed in a sequence of 60 video frames. White noise was added to the mark er coordinates. Realistic loss of data due to detection failure was si mulated by deleting a generated marker image when the distance to anot her marker image was below a given minimum value. In a test, realistic values were substituted for the noise level sigma (0.5 pixel) and the minimum marker distance d(m) (4 pixels). For numbers of markers rangi ng from 50 to 100, 95-90% of the detected marker images was correctly tracked. Less than 0.7% was part of a false track, i.e. a track contai ning images of different markers. Under less favourable conditions (si gma = 1 pixel; d(m) = 8 pixels) the method was robust: for 75 markers with 40% of the marker images missing, still 70% of the detected image s was correctly tracked, while the fraction in false tracks did not in crease. The LR tracking method appears reliable for automatic tracking of large amounts of moving markers in a sequence of video or cine fra mes. Copyright (C) 1996 Elsevier Science Ltd.