Video Analysis for Clinical Trials

Atom5^TM video analysis locates specific points on the body and computes the speed and trajectory of each movement to explore smoothness of motion and erratic patterns, enabling clinicians to assess features such as gait and posture.

Part of our eCOA suite of tools, video capture has multiple applications but is especially helpful for movement disorders. Our pioneering work with Duchenne UK on DMDhome demonstrates the potential that video assessments have for more meaningful endpoints as documented in our paper, “Patient-led development of digital endpoints and the use of computer vision analysis in assessment of motor function in rare diseases“.

Let’s show you what Aparito’s video analysis can do!

Turn & Sit Action

The plot shows the location of the shoulders through the turn & sit motion. Each dot is the location in the image of the shoulders in the frame of the video. Dot size illustrates time; the larger the dot, the later in the video. In the clip and in the plot, the turn and then lower to sit can be observed as relatively distinct actions.

Leg Alignment

The plot shows the inner angle at the right knee. It’s difficult to interpret data from the entire duration of the walk, as the depth of the subject relative to the camera is constantly changing, and even if stationary, there is depth to the subject’s body (e.g. the distance between the feet mid-stride) and we have a 2D representation of this 3D reality. However, a point of interest is to see whether the leg becomes straight (the 180-degree dashed line) when the foot is placed down on the ground!

Arm Swing

The plot shows the angle between the shoulder and the wrist. The 0-degree dashed line implies that the arm is straight down. The oscillation peaks illustrate the extent of the arm swing. Two associated videos show an observable difference in magnitude, which is clearly seen in the plot.