scholarly journals Comparison of manual and machine learning image processing approaches to determine fungiform papillae on the tongue

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Camilla Cattaneo ◽  
Jing Liu ◽  
Chenhao Wang ◽  
Ella Pagliarini ◽  
Jon Sporring ◽  
...  
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Taste Perception ◽  
Automated Image Analysis ◽  
Learning Approach ◽  
Fungiform Papillae ◽  
Automated Method ◽  
Manual Counting ◽  
Machine Learning Approach ◽  
Counting Methods

Abstract Human taste perception is associated with the papillae on the tongue as they contain a large proportion of chemoreceptors for basic tastes and other chemosensation. Especially the density of fungiform papillae (FP) is considered as an index for responsiveness to oral chemosensory stimuli. The standard procedure for FP counting involves visual identification and manual counting of specific parts of the tongue by trained operators. This is a tedious task and automated image analysis methods are desirable. In this paper a machine learning image processing method based on a convolutional neural network is presented. This automated method was compared with three standard manual FP counting procedures using tongue pictures from 132 subjects. Automated FP counts, within the selected areas and the whole tongue, significantly correlated with the manual counting methods (all ρs ≥ 0.76). When comparing the images for gender and PROP status, the density of FP predicted from automated analysis was in good agreement with data from the manual counting methods, especially in the case of gender. Moreover, the present results reinforce the idea that caution should be applied in considering the relationship between FP density and PROP responsiveness since this relationship can be an oversimplification of the complexity of phenomena arising at the central and peripherical levels. Indeed, no significant correlations were found between FP and PROP bitterness ratings using the automated method for selected areas or the whole tongue. Besides providing estimates of the number of FP, the machine learning approach used a tongue coordinate system that normalizes the size and shape of an individual tongue and generated a heat map of the FP position and normalized area they cover. The present study demonstrated that the machine learning approach could provide similar estimates of FP on the tongue as compared to manual counting methods and provide estimates of more difficult-to-measure parameters, such as the papillae's areas and shape.

scholarly journals Comparison of manual and machine learning image processing approaches to determine fungiform papillae on the tongue

2020 ◽  
Author(s):  
Camilla Cattaneo ◽  
Jing Liu ◽  
Chenhao Wang ◽  
Ella Pagliarini ◽  
Jon Sporring ◽  
...  
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Taste Perception ◽  
Automated Image Analysis ◽  
Learning Approach ◽  
Fungiform Papillae ◽  
Automated Method ◽  
Manual Counting ◽  
Machine Learning Approach ◽  
Counting Methods

AbstractHuman taste perception is associated with the papillae on the tongue as they contain a large proportion of chemoreceptors for basic tastes and other chemosensation. Especially the density of fungiform papillae (FP) is considered as an index for responsiveness to oral chemosensory stimuli. The standard procedure for FP counting involves visual identification and manual counting of specific parts of the tongue by trained operators. This is a tedious task and automated image analysis methods are desirable. In this paper a machine learning image processing method based on a convolutional neural network is presented. This automated method was compared with three standard manual FP counting procedures using tongue pictures from 132 subjects. Automated FP counts, within the selected areas and the whole tongue, significantly correlated with the manual counting methods (all ρs ≥ 0.76). When comparing the images for gender and PROP status, the density of FP predicted from automated analysis was in good agreement with data from the manual counting methods, especially in the case of gender. Moreover, the present results reinforce the idea that caution should be applied in considering the relationship between FP density and PROP responsiveness since this relationship can be an oversimplification of the complexity of phenomena arising at the central and peripherical levels. Indeed, no significant correlations were found between FP and PROP bitterness ratings using the automated method for selected areas or the whole tongue. Besides providing estimates of the number of FP, the machine learning approach used a tongue coordinate system that normalizes the size and shape of an individual tongue and generated a heat map of the FP position and normalized area they cover. The present study demonstrated that the machine learning approach could provide similar estimates of FP on the tongue as compared to manual counting methods and provide estimates of more difficult-to-measure parameters, such as the papillae’s areas and shape.

scholarly journals A Machine Learning Approach to Crater Classification from Topographic Data

2019 ◽  
Vol 11 (21) ◽  
pp. 2594
Author(s):  
Qiangyi Liu ◽  
Weiming Cheng ◽  
Guangjian Yan ◽  
Yunliang Zhao ◽  
Jianzhong Liu
Keyword(s):  
Machine Learning ◽  
Cross Validation ◽  
Random Forest Classifier ◽  
Learning Approach ◽  
Geological History ◽  
Automated Method ◽  
Machine Learning Approach ◽  
Absolute Age ◽  
The Impact ◽  
Crater Shape

Craters contain important information on geological history and have been widely used for dating absolute age and reconstructing impact history. The impact process results in a lot of ejected fragments and these fragments may form secondary craters. Studies on distinguishing primary craters from secondary craters are helpful in improving the accuracy of crater dating. However, previous studies about distinguishing primary craters from secondary craters were either conducted by manual identification or used approaches mainly concerning crater spatial distribution, which are time-consuming or have low accuracy. This paper presents a machine learning approach to distinguish primary craters from secondary craters. First, samples used for training and testing were identified and unified. The whole dataset contained 1032 primary craters and 4041 secondary craters. Then, considering the differences between primary and secondary craters, features mainly related to crater shape, depth, and density were calculated. Finally, a random forest classifier was trained and tested. This approach showed a favorable performance. The accuracy and F1-score for fivefold cross-validation were 0.939 and 0.839, respectively. The proposed machine learning approach enables an automated method of distinguishing primary craters from secondary craters, which results in better performance.

scholarly journals Ontology learning algorithm using weak functions

Open Physics ◽  
2018 ◽  
Vol 16 (1) ◽  
pp. 910-916 ◽  
Author(s):  
Linli Zhu ◽  
Gang Hua ◽  
Adnan Aslam
Keyword(s):  
Machine Learning ◽  
Image Processing ◽  
Learning Algorithm ◽  
Point Of View ◽  
Learning Approach ◽  
Theoretical Point ◽  
Ontology Learning ◽  
Machine Learning Approach ◽  
Similarity Calculation ◽  
Weak Ontology

AbstractOntology is widely used in information retrieval, image processing and other various disciplines. This article discusses how to use machine learning approach to solve the most essential similarity calculation problem in multi-dividing ontology setting. The ontology function is regarded as a combination of several weak ontology functions, and the optimal ontology function is obtained by an iterative algorithm. In addition, the performance of the algorithm is analyzed from a theoretical point of view by statistical methods, and several results are obtained.

Sign in / Sign up

Export Citation Format

Share Document