Abstract
Tongue coating can provide valuable diagnostic information to reveal the disorder of the internal body. However, tongue coating classification has long been a challenging task in Traditional Chinese Medicine (TCM) due to the fact that tongue coatings are polymorphous, different tongue coatings have different colors, shapes, textures and locations. Most existing analyses utilize handcrafted features extracted from a fixed location, which may lead to inconsistent performance when the size or location of the tongue coating region varies. To solve this problem, this paper proposes a novel paradigm by employing artificial intelligence to feature extraction and classification of tongue coating. It begins with exploiting prior knowledge of rotten-greasy tongue coating to obtain suspected tongue coating patches. Based on the resulting patches, tongue coating features extracted by Convolutional Neural Network (CNN) are used instead of handcrafted features. Moreover, a multiple-instance Support Vector Machine (MI-SVM) which can circumvent the uncertain location problem is applied to tongue coating classification. Experimental results demonstrate that the proposed method outperforms state-of-the-art tongue coating classification methods.
Highlights
Tongue diagnosis is an effective treatment in Traditional Chinese Medicine (TCM)
In this paper, we have presented a new method for tongue coating classification using multiple-instance learning (MIL) and deep features
Tongue coating is represented by a bag consisting of multiple feature vectors and multiple-instance Support Vector Machine (MI-Support Vector Machine (SVM)) is used to perform the final classification
Summary
Tongue diagnosis is an effective treatment in Traditional Chinese Medicine (TCM). The tongue is rich in geometric features and texture features, which are closely linked to the physiological information of human organs. As described in [14], the classification of tongue coatings maps naturally to a multiple-instance problem Along this line of thought, only coarsely labeled images are required to train a MIL model. According to the TCM perspective and our observation (as shown in Fig. 4), the rottengreasy coating always appears in the middle and rear of a tongue body, while the rest of the tongue can be ignored These patches selected from a rotten-greasy coating tongue can satisfy the assumption of multiple-instance binary classification that there exists at least one positive instance in a positive bag. Motivated by the characteristics of CNN which can naturally integrate low/mid/high level features, we use a CNN to extract fixed-length feature vectors of the tongue coating patches instead of traditional handcrafted methods.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
