The density of fungiform papillae (FPD) on the human tongue is currently taken as index for responsiveness to oral chemosensory stimuli. Visual analysis of digital tongue picture and manual counting by trained operators represents the most popular technique for FPD assessment. Methodological issues mainly due to operator bias are considered among factors accounting for the uncertainty about the relationships between FPD and responsiveness to chemosensory stimuli. The present study describes a novel automated method to count fungiform papillae (FP) from image analysis of tongue pictures. The method was applied to tongue pictures from 133 subjects. Taking the manual count as reference method, a partial least squares regression model was developed to predict FPD from tongue automated analysis output. FPD from manual and automated count showed the same normal distribution and comparable descriptive statistic values. Consistent subject classifications as low and high FPD were obtained according to the median values from manual and automated count. The same results on the effect of FPD variation on taste perception were obtained both using predicted and counted values. The proposed method overcomes count uncertainties due to researcher bias in manual counting and is suited for large population studies. Additional information is provided such as FP size class distribution which would help for a better understanding of the relationships between FPD variation and taste functions.

Comparing Manual Counting to Automated Image Analysis for the Assessment of Fungiform Papillae Density on Human Tongue / Piochi, Maria; Monteleone, Erminio; Torri, Luisa; Masi, Camilla; Almli, Valérie Lengard; Wold, Jens Petter; Dinnella, Caterina. - In: CHEMICAL SENSES. - ISSN 0379-864X. - STAMPA. - 42:(2017), pp. 553-561. [10.1093/chemse/bjx035]

Comparing Manual Counting to Automated Image Analysis for the Assessment of Fungiform Papillae Density on Human Tongue

PIOCHI, MARIA;MONTELEONE, ERMINIO;MASI, CAMILLA;DINNELLA, CATERINA
2017

Abstract

The density of fungiform papillae (FPD) on the human tongue is currently taken as index for responsiveness to oral chemosensory stimuli. Visual analysis of digital tongue picture and manual counting by trained operators represents the most popular technique for FPD assessment. Methodological issues mainly due to operator bias are considered among factors accounting for the uncertainty about the relationships between FPD and responsiveness to chemosensory stimuli. The present study describes a novel automated method to count fungiform papillae (FP) from image analysis of tongue pictures. The method was applied to tongue pictures from 133 subjects. Taking the manual count as reference method, a partial least squares regression model was developed to predict FPD from tongue automated analysis output. FPD from manual and automated count showed the same normal distribution and comparable descriptive statistic values. Consistent subject classifications as low and high FPD were obtained according to the median values from manual and automated count. The same results on the effect of FPD variation on taste perception were obtained both using predicted and counted values. The proposed method overcomes count uncertainties due to researcher bias in manual counting and is suited for large population studies. Additional information is provided such as FP size class distribution which would help for a better understanding of the relationships between FPD variation and taste functions.
2017
42
553
561
Piochi, Maria; Monteleone, Erminio; Torri, Luisa; Masi, Camilla; Almli, Valérie Lengard; Wold, Jens Petter; Dinnella, Caterina
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1099549
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 12
  • ???jsp.display-item.citation.isi??? 9
social impact