Hierarchical Network for Facial Palsy Detection

Hsu, Gee-Sern; Huang, Wen-Fong; Kang, Jiunn Horng

doi:10.1109/cvprw.2018.00100

Cited by 22 publications

(32 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It first uses an ordinary camera to take pictures of the patient's face when it is at rest or performing specified facial expressions. Then, computational techniques [11]- [14] in various areas such as computer vision, image processing and machine learning are employed to objectively and quantitatively assess the facial nerve function within a certain feature space. The resulting solution can significantly reduce the subjective bias in assessment and would be easily ported to ubiquitous mobile devices such as smartphones and tablets, hence has promising applicability in facial palsy diagnosis and therapy.…”

Section: Automated Assessment From Visual Face Capturementioning

confidence: 99%

“…NGO et al further extended the facial texture analysis from spatial domain to frequency domain by using Gabor filters [50], circular Gabor filters [44]. More recent studies [14], [51] turned to deep learning methods such as convolutional neural networks (CNNs) which have revolutionized the visual imagery analysis to extract high-level features from the face image. The extracted features are supposed to embed the most prominent image patterns probably including the facial abnormality into a compact numerical vector.…”

Section: A Computational Measures In 2dmentioning

confidence: 99%

“…The most frequently used grading scale is House-Brackmann scale (HBS) which divides the facial nerve function into six levels [11], [12], [30], [51], followed by Yanagihara scale (YGS) [44], [50] and Sunnybrook scale (SGS, see Table I) [47]. The grade could also simply be a binary value indicating whether the subject has facial palsy or not [37], [46], or if a specific face region is paralyzed [14]. The reported classification accuracy (by comparing the predicted grade against that from the clinician) varies a lot, ranging from 49.9% [12] to 95.5% [47].…”

Section: Assessment Outcomesmentioning

confidence: 99%

“…With ubiquitous visual face captureimages/videos, it is more widely accessible than those using obtrusive physical interventions such as electroneurography (ENoG) and electromyography (EMG). This motivated a branch of study [6]- [14] in this field to employ computational measures on biomedical visual face capture to objectively and quantitatively evaluate the facial nerve function. Such a solution is capable of automatically quantifying facial nerve function in millimetric precision [7], [13] or with semantic grades [15]- [19] based on a machine learning model trained on clinician labelled data [6], [11], [12].…”

Section: Introductionmentioning

confidence: 99%

“…This provides a highly efficient and cost effective means whereby facial nerve function can be appraised in an objective manner. With the development of techniques in image processing, computer vision and machine learning, especially those of 2D/3D face tracking [8]- [10] and feature learning [14], the field of automated facial nerve function assessment has witnessed promising progress and developed various instruments in recent years. However, to date, none of these instruments has gained a wide clinical use.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Review on Automated Facial Nerve Function Assessment From Visual Face Capture

Lou

Wang

2020

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Section: Automated Assessment From Visual Face Capturementioning

confidence: 99%

Section: A Computational Measures In 2dmentioning

confidence: 99%

Section: Assessment Outcomesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Review on Automated Facial Nerve Function Assessment From Visual Face Capture

Lou

Wang

2020

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

An Overview of Recent Developments in Convolutional Neural Network (CNN) Based Face Detector

Yadav

Priyanka

2020

Computational Methods and Data Engineering

View full text Add to dashboard Cite

Deep Hierarchical Network With Line Segment Learning for Quantitative Analysis of Facial Palsy

2019

Self Cite

View full text Add to dashboard Cite

We propose the deep hierarchical network (DHN) for the quantitative analysis of facial palsy. Facial palsy, also known as Bell's palsy, is the most common type of facial nerve palsy that results in the loss of muscle control in the affected facial regions. Typical symptoms include facial deformity and facial expression dysfunction. To the best of our best knowledge, all approaches for the automatic detection of facial palsy consider hand-crafted features. This paper reports the first deep-learning-based approach developed for the real-time quantitative analysis of facial palsy. The proposed DHN consists of three component networks: the first detects the subject's face, the second detects the facial landmarks and line segments on the detected face, and the third detects the local palsy regions. The first component network is built on the YOLO2 detector. The second component network is developed on a fused network architecture that incorporates a line segment learning network for locating the facial landmarks and line segments. The third component network is developed on an object detection network with the line-segment-embedded input that combines the landmarked region and the line segments detected by the second component network. The novelties of this research include: 1) the modification of a state-of-the-art edge detector for extracting the facial line segments; 2) the embedding of the line segment learning for the detection of facial landmarks and local palsy regions; 3) the quantitative description of the facial palsy syndrome intensity; and 4) the release of the first clinically labeled database, the YouTube Facial Palsy (YFP) database. The making of the YFP database solves the issue that previous methods were all evaluated on proprietary databases, making the comparison of different methods extremely difficult. The YFP database includes 32 videos of 21 patients collected from YouTube and labeled by clinic specialists. To enhance the robustness against facial expression variations, we include the CK+ facial expression database in the training. We show that the proposed DHN not only just detects the local palsy regions but also captures the intensity of the facial palsy syndrome over time, enabling the quantitative description of the syndrome. The experiments show that the proposed approach offers an accurate and efficient real-time solution for facial palsy analysis.INDEX TERMS Facial palsy, bell palsy, medical image diagnosis, face alignment, face recognition.

show abstract

Hierarchical Network for Facial Palsy Detection

Cited by 22 publications

References 7 publications

A Review on Automated Facial Nerve Function Assessment From Visual Face Capture

A Review on Automated Facial Nerve Function Assessment From Visual Face Capture

An Overview of Recent Developments in Convolutional Neural Network (CNN) Based Face Detector

Deep Hierarchical Network With Line Segment Learning for Quantitative Analysis of Facial Palsy

Contact Info

Product

Resources

About