Choosing the best image processing method for masticatory performance assessment when using two‐coloured specimens

Abstract

Objective masticatory performance assessment using two-coloured specimens relies on image processing techniques; however, just a few approaches have been tested and no comparative studies are reported. The aim of this study was to present a selection procedure of the optimal image analysis method for masticatory performance assessment with a given two-coloured chewing gum. Dentate participants (n = 250; 25 ± 6·3 years) chewed red-white chewing gums for 3, 6, 9, 12, 15, 18, 21 and 25 cycles (2000 samples). Digitalised images of retrieved specimens were analysed using 122 image processing methods (IPMs) based on feature extraction algorithms (pixel values and histogram analysis). All IPMs were tested following the criteria of: normality of measurements (Kolmogorov-Smirnov), ability to detect differences among mixing states (anova corrected with post hoc Bonferroni) and moderate-to-high correlation with the number of cycles (Spearman's Rho). The optimal IPM was chosen using multiple criteria decision analysis (MCDA). Measurements provided by all IPMs proved to be normally distributed (P < 0·05), 116 proved sensible to mixing states (P < 0·05), and 35 showed moderate-to-high correlation with the number of cycles (|ρ| > 0·5; P < 0·05). The variance of the histogram of the Hue showed the highest correlation with the number of cycles (ρ = 0·792; P < 0·0001) and the highest MCDA score (optimal). The proposed procedure proved to be reliable and able to select the optimal approach among multiple IPMs. This experiment may be reproduced to identify the optimal approach for each case of locally available test foods.

Systematic review of measurement properties of methods for objectively assessing masticatory performance

“…Different domains and measurement properties were reported, of which, the most common was to report the validity of the method for assessing masticatory performance ( n = 36 studies; Abe et al, ; Felicio et al, ; Eberhard et al, ; Eberhard et al, ; Goto et al, ; Gunne, ; Halazonetis et al, ; Hama et al, ; Hama et al, ; Hayakawa et al, ; Huggare, ; Ikebe et al, ; Ikebe, Matsuda, Murai, Maeda, & Nokubi, ; Kamiyama et al, ; Khoury‐Ribas et al, ; Kobayashi et al, ; Mahmood et al, ; Matsui et al, ; Mowlana et al, ; Nokubi et al, ; Ohara et al, ; Prinz, ; Sanchez‐Ayala et al, ; Sato et al, ; Schimmel et al, ; Schimmel et al, ; Shiga et al, ; Slagter et al, ; Speksnijder, Abbink, van der Glas, Janssen, & van der Bilt, ; Sugiura et al, ; Vaccaro, Pelaez, & Gil, ; van der Bilt, Mojet, Tekamp, & Abbink, ; van der Bilt, Speksnijder, de Liz Pocztaruk, & Abbink, ; Wada et al, ; Weijenberg et al, ; Woda et al, ). Construct validity or hypotheses testing was frequently described ( n = 30 studies; Abe et al, ; Felicio et al, ; Goto et al, ; Gunne, ; Halazonetis et al, ; Hama et al, ; Hayakawa et al, ; Huggare, ; Ikebe et al, ; Ikebe et al, ; Khoury‐Ribas et al, ; Mahmood et al, ; Matsui et al, ; Mowlana et al, ; Nokubi et al, ; Ohara et al, ; Prinz, ; Sato et al, ; Schimmel et al, ; Schimmel et al, ; Shiga et al, ; Slagter et al, ; Speksnijder et al, ; Sugiura et al, …”

“…Methods for assessing masticatory performance may be categorized into three main categories; that is, comminution methods ( n = 21; Felicio et al, ; Eberhard et al, ; Eberhard et al, ; Escudeiro Santos et al, ; Fauzza & Lyons, ; Gunne, ; Huggare, ; Ikebe et al, ; Kapur et al, ; Khoury‐Ribas et al, ; Kobayashi et al, ; Lujan‐Climent et al, ; Mahmood et al, ; Mowlana et al, ; Nokubi et al, ; Ohara et al, ; Sanchez‐Ayala et al, ; Sanchez‐Ayala et al, ; Shiga et al, ; Slagter et al, ; Woda et al, ), mixing ability methods ( n = 23; Abe et al, ; Asakawa et al, ; Endo et al, ; Halazonetis et al, ; Hama et al, ; Hama et al, ; Hayakawa et al, ; Ishikawa et al, ; Kamiyama et al, ; Liedberg & Owall, ; Matsui et al, ; Prinz, ; Sato et al, ; Schimmel et al, ; Schimmel et al, ; Silva et al, ; Speksnijder et al, ; Sugiura et al, ; Vaccaro et al, ; van der Bilt et al, ; van der Bilt et al, ; Wada et al, ; Weijenberg et al, ), or other methods ( n = 2; Goto et al, ; Ikebe et al, ).…”

“…An excellent‐quality study reported construct validity regarding two‐color gum Using MathLab 2015b, [positive rating] (Vaccaro et al, ). Two fair‐quality studies reported construct validity regarding use of ViewGum© for assessing masticatory performance with various types of two‐color gums [positive rating] (Halazonetis et al, ; Schimmel et al, ).…”

“…To summarize, the studies reporting methods using two‐color chewing gums and digital analysis revealed moderate to strong level of evidence for construct validity (Halazonetis et al, ; Schimmel et al, ; Vaccaro et al, ), and moderate level of evidence for reliability using a visual scale (Silva et al, ). Moderate level of evidence was also reported for construct validity using two‐colored wax (Speksnijder et al, ; Sugiura et al, ; van der Bilt et al, ).…”

“…Regarding construct validity, six studies reported moderate level of evidence (Halazonetis et al, ; Sato et al, ; Schimmel et al, ; Speksnijder et al, ; Sugiura et al, ; van der Bilt et al, ), and one study reported strong level of evidence (Vaccaro et al, ). Regarding reliability, moderate level of evidence was reported in one study that used a visual bolus/color scale (Silva et al, ).…”