Automatic Facial Paralysis Assessment via Computational Image Analysis

Abstract

Facial paralysis (FP) is a loss of facial movement due to nerve damage. Most existing diagnosis systems of FP are subjective, e.g., the House-Brackmann (HB) grading system, which highly depends on the skilled clinicians and lacks an automatic quantitative assessment. In this paper, we propose an efficient yet objective facial paralysis assessment approach via automatic computational image analysis. First, the facial blood flow of FP patients is measured by the technique of laser speckle contrast imaging to generate both RGB color images and blood flow images. Second, with an improved segmentation approach, the patient's face is divided into concerned regions to extract facial blood flow distribution characteristics. Finally, three HB score classifiers are employed to quantify the severity of FP patients. The proposed method has been validated on 80 FP patients, and quantitative results demonstrate that our method, achieving an accuracy of 97.14%, outperforms the state-of-the-art systems. Experimental evaluations also show that the proposed approach could yield objective and quantitative FP diagnosis results, which agree with those obtained by an experienced clinician.

Figure 1

The proposed framework of the automatic facial paralysis assessment system based on quantitative computational image analysis.

Figure 2

(a) Color image. (b) Color blood flow image. (c) Gray blood flow image.

Figure 3

(a) 68 face landmarks. (b) Face landmarks in red color.

Figure 4

3D face model (a) with color texture, (b) with facial blood texture, and (c) with premarked regions.

Figure 5

The premarked regions used in the quantitative assessment. L means left side, R means right side, and other abbreviations are listed in Table 2.

Figure 6

(a) The normal color facial blood flow image. (b) The wrong color blood flow image. The two images were saved during one continuous scanning, but image (b) is slightly blurred due to the movement of the participant's head, and the facial blood flow data are abnormally high.

Figure 7

The ratio of the affected side to the healthy side of orbital blood flow in patients with different HB scores. From left to right, the HB score is from I to VI. For the healthy samples with HB score (I), we use the right side as the affected side.