• Something wrong with this record ?

Predicting Chronic Hyperplastic Candidiasis Retro-Angular Mucosa Using Machine Learning

O. Moztarzadeh, J. Liska, V. Liskova, A. Skalova, O. Topolcan, A. Jamshidi, L. Hauer

. 2023 ; 13 (6) : 1335-1351. [pub] 20231028

Status not-indexed Language English Country Switzerland

Document type Journal Article

Persistent link   https://www.medvik.cz/link/bmc23022639

Chronic hyperplastic candidiasis (CHC) presents a distinctive and relatively rare form of oral candidal infection characterized by the presence of white or white-red patches on the oral mucosa. Often mistaken for leukoplakia or erythroleukoplakia due to their appearance, these lesions display nonhomogeneous textures featuring combinations of white and red hyperplastic or nodular surfaces. Predominant locations for such lesions include the tongue, retro-angular mucosa, and buccal mucosa. This paper aims to investigate the potential influence of specific anatomical locations, retro-angular mucosa, on the development and occurrence of CHC. By examining the relationship between risk factors, we present an approach based on machine learning (ML) to predict the location of CHC occurrence. In this way, we employ Gradient Boosting Regression (GBR) to classify CHC lesion locations based on important risk factors. This estimator can serve both research and diagnostic purposes effectively. The findings underscore that the proposed ML technique can be used to predict the occurrence of CHC in retro-angular mucosa compared to other locations. The results also show a high rate of accuracy in predicting lesion locations. Performance assessment relies on Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R-squared (R2), and Mean Absolute Error (MAE), consistently revealing favorable results that underscore the robustness and dependability of our classification method. Our research contributes valuable insights to the field, enhancing diagnostic accuracy and informing treatment strategies.

References provided by Crossref.org

000      
00000naa a2200000 a 4500
001      
bmc23022639
003      
CZ-PrNML
005      
20240116163107.0
007      
ta
008      
240105s2023 sz f 000 0|eng||
009      
AR
024    7_
$a 10.3390/clinpract13060120 $2 doi
035    __
$a (PubMed)37987421
040    __
$a ABA008 $b cze $d ABA008 $e AACR2
041    0_
$a eng
044    __
$a sz
100    1_
$a Moztarzadeh, Omid $u Department of Stomatology, University Hospital Pilsen, Faculty of Medicine in Pilsen, Charles University, Alej Svobody 80, 30460 Pilsen, Czech Republic $u Department of Anatomy, Faculty of Medicine in Pilsen, Charles University, 32300 Pilsen, Czech Republic $1 https://orcid.org/0000000247682382
245    10
$a Predicting Chronic Hyperplastic Candidiasis Retro-Angular Mucosa Using Machine Learning / $c O. Moztarzadeh, J. Liska, V. Liskova, A. Skalova, O. Topolcan, A. Jamshidi, L. Hauer
520    9_
$a Chronic hyperplastic candidiasis (CHC) presents a distinctive and relatively rare form of oral candidal infection characterized by the presence of white or white-red patches on the oral mucosa. Often mistaken for leukoplakia or erythroleukoplakia due to their appearance, these lesions display nonhomogeneous textures featuring combinations of white and red hyperplastic or nodular surfaces. Predominant locations for such lesions include the tongue, retro-angular mucosa, and buccal mucosa. This paper aims to investigate the potential influence of specific anatomical locations, retro-angular mucosa, on the development and occurrence of CHC. By examining the relationship between risk factors, we present an approach based on machine learning (ML) to predict the location of CHC occurrence. In this way, we employ Gradient Boosting Regression (GBR) to classify CHC lesion locations based on important risk factors. This estimator can serve both research and diagnostic purposes effectively. The findings underscore that the proposed ML technique can be used to predict the occurrence of CHC in retro-angular mucosa compared to other locations. The results also show a high rate of accuracy in predicting lesion locations. Performance assessment relies on Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R-squared (R2), and Mean Absolute Error (MAE), consistently revealing favorable results that underscore the robustness and dependability of our classification method. Our research contributes valuable insights to the field, enhancing diagnostic accuracy and informing treatment strategies.
590    __
$a NEINDEXOVÁNO
655    _2
$a časopisecké články $7 D016428
700    1_
$a Liska, Jan $u Department of Stomatology, University Hospital Pilsen, Faculty of Medicine in Pilsen, Charles University, Alej Svobody 80, 30460 Pilsen, Czech Republic $1 https://orcid.org/0000000273151681
700    1_
$a Liskova, Veronika $u Department of Stomatology, University Hospital Pilsen, Faculty of Medicine in Pilsen, Charles University, Alej Svobody 80, 30460 Pilsen, Czech Republic $1 https://orcid.org/0000000200780487
700    1_
$a Skalova, Alena $u Sikl's Department of Pathology, Faculty of Medicine in Pilsen, Charles University, Ed. Beneše 13, 30599 Pilsen, Czech Republic $u Biopticka Laboratory, Mikulasske namesti 628, 32600 Pilsen, Czech Republic
700    1_
$a Topolcan, Ondrej $u Central Laboratory of Immunoanalysis, University Hospital Pilsen, Faculty of Medicine in Pilsen, Charles University, Ed. Beneše 13, 30599 Pilsen, Czech Republic
700    1_
$a Jamshidi, Alireza $u Dentistry School, Babol University of Medical Sciences, Babol 4717647745, Iran
700    1_
$a Hauer, Lukas $u Department of Stomatology, University Hospital Pilsen, Faculty of Medicine in Pilsen, Charles University, Alej Svobody 80, 30460 Pilsen, Czech Republic
773    0_
$w MED00201253 $t Clinics and practice $x 2039-7275 $g Roč. 13, č. 6 (2023), s. 1335-1351
856    41
$u https://pubmed.ncbi.nlm.nih.gov/37987421 $y Pubmed
910    __
$a ABA008 $b sig $c sign $y - $z 0
990    __
$a 20240105 $b ABA008
991    __
$a 20240116163104 $b ABA008
999    __
$a ok $b bmc $g 2036345 $s 1209084
BAS    __
$a 3
BAS    __
$a PreBMC-PubMed-not-MEDLINE
BMC    __
$a 2023 $b 13 $c 6 $d 1335-1351 $e 20231028 $i 2039-7275 $m Clinics and practice $n Clin Pract $x MED00201253
LZP    __
$a Pubmed-20240105

Find record

Citation metrics

Archiving options