Extraction of Temporal Motor Activity Signals From Video Recordings of Neonatal Seizures By Improved Feature Tracking Methods Based on Translation Motion Models

This paper presents a new method for tracking features in video. This method estimates the displacement of a feature between two successive frames by minimizing an error function defined in terms of the feature intensities at these frames. The minimization problem is made analytically tractable by approximating the error function using a second-order Taylor expansion. The displacement between two successive frames is computed in an iterative fashion using gradient descent. The improved reliability of the proposed method is illustrated by its application in the extraction of temporal motor activity signals from video recordings of neonatal seizures. successive frames is estimated by minimizing an error function defined in terms of the intensity functions at these frames. In the proposed procedure, the error function is approximated by using a second-order Taylor expansion for the intensity function at the next frame. The proposed feature tracking method is used to extract motor activity signals from video recordings of neonatal seizures. Extraction of Temporal Motor Activity Signals From Video Recordings of Neonatal Seizures By Improved Feature Tracking Methods Based on Translation Motion Models Nicolaos B. Karayiannis and Yaohua Xiong Department of Electrical and Computer Engineering, University of Houston, Houston, TX 77204-4005 II. EXTRACTION OF MOTOR ACTIVITY SIGNALS FROM VIDEO Motor activity signals can be extracted by projecting the location of selected anatomical sites to the horizontal and vertical axes. As the seizure progresses in time, these projections will produce temporal signals recording motor activity of the body parts of interest. Keywords—Feature tracking, motor activity signal, translation motion model Figure 1 illustrates the mechanism that can be used for generating temporal signals tracking the movements of different parts of the infant’s body during focal clonic and myoclonic seizures. Focal clonic and myoclonic seizures are manifested as repetitive and rapid movements of the infants’ extremities, respectively [1], [5], [6], [10]. Figure 1 depicts a single frame containing the sketch of an infant’s body with four selected anatomical sites. In this particular configuration, XLL and YLL represent the projections of the site located at the left leg to the horizontal and vertical axes, respectively. The projections of the sites located at the right leg, left hand, and right hand are denoted by XRL and YRL, XLH and YLH, and XRH and YRH, respectively. As the infant moves its extremities, the locations of the sites in the frame will change, as will the projections of the sites to the horizontal and vertical axes. Recording the values of the projections from frame to frame of the videotaped seizure will generate four pairs of temporal signals, namely the signals XLL(t) and YLL(t) for the left leg, the signals XRL(t) and YRL(t) for the right leg, the signals XLH(t) and YLH(t) for the left hand, and the signals XRH(t) and YRH(t) for the right hand. For a given set of anatomical sites, each seizure will produce signature signals depending on its type and location.