1
|
Küchler EC, Kirschneck C, Marañón-Vásquez GA, Schroder ÂGD, Baratto-Filho F, Romano FL, Stuani MBS, Matsumoto MAN, de Araujo CM. Mandibular and dental measurements for sex determination using machine learning. Sci Rep 2024; 14:9587. [PMID: 38671054 PMCID: PMC11053013 DOI: 10.1038/s41598-024-59556-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2023] [Accepted: 04/12/2024] [Indexed: 04/28/2024] Open
Abstract
The present study tested the combination of mandibular and dental dimensions for sex determination using machine learning. Lateral cephalograms and dental casts were used to obtain mandibular and mesio-distal permanent teeth dimensions, respectively. Univariate statistics was used for variables selection for the supervised machine learning model (alpha = 0.05). The following algorithms were trained: logistic regression, gradient boosting classifier, k-nearest neighbors, support vector machine, multilayer perceptron classifier, decision tree, and random forest classifier. A threefold cross-validation approach was adopted to validate each model. The areas under the curve (AUC) were computed, and ROC curves were constructed. Three mandibular-related measurements and eight dental size-related dimensions were used to train the machine learning models using data from 108 individuals. The mandibular ramus height and the lower first molar mesio-distal size exhibited the greatest predictive capability in most of the evaluated models. The accuracy of the models varied from 0.64 to 0.74 in the cross-validation stage, and from 0.58 to 0.79 when testing the data. The logistic regression model exhibited the highest performance (AUC = 0.84). Despite the limitations of this study, the results seem to show that the integration of mandibular and dental dimensions for sex prediction would be a promising approach, emphasizing the potential of machine learning techniques as valuable tools for this purpose.
Collapse
Affiliation(s)
- Erika Calvano Küchler
- Department of Orthodontics, Medical Faculty, University Hospital Bonn, Welschnonnenstr. 17, 53111, Bonn, Germany.
| | - Christian Kirschneck
- Department of Orthodontics, Medical Faculty, University Hospital Bonn, Welschnonnenstr. 17, 53111, Bonn, Germany
| | - Guido Artemio Marañón-Vásquez
- Department of Pediatric Dentistry, School of Dentistry of Ribeirão Preto, University of São Paulo, Av. do Café s/n, Ribeirão Preto, São Paulo, 14040-904, Brazil
| | - Ângela Graciela Deliga Schroder
- Postgraduate Program in Communication Disorders, Tuiuti University of Paraná, R. Padre Ladislau Kula 395, Curitiba, Paraná, 82010-210, Brazil
- School of Dentistry, Tuiuti University of Paraná, R. Padre Ladislau Kula 395, Curitiba, Paraná, 82010-210, Brazil
| | - Flares Baratto-Filho
- School of Dentistry, Tuiuti University of Paraná, R. Padre Ladislau Kula 395, Curitiba, Paraná, 82010-210, Brazil
- Department of Dentistry, University of the Region of Joinville (Univille), R. Paulo Malschitzki 10, Joinville, Santa Catarina, 89219-710, Brazil
| | - Fábio Lourenço Romano
- Department of Pediatric Dentistry, School of Dentistry of Ribeirão Preto, University of São Paulo, Av. do Café s/n, Ribeirão Preto, São Paulo, 14040-904, Brazil
| | - Maria Bernadete Sasso Stuani
- Department of Pediatric Dentistry, School of Dentistry of Ribeirão Preto, University of São Paulo, Av. do Café s/n, Ribeirão Preto, São Paulo, 14040-904, Brazil
| | - Mírian Aiko Nakane Matsumoto
- Department of Pediatric Dentistry, School of Dentistry of Ribeirão Preto, University of São Paulo, Av. do Café s/n, Ribeirão Preto, São Paulo, 14040-904, Brazil
| | - Cristiano Miranda de Araujo
- Postgraduate Program in Communication Disorders, Tuiuti University of Paraná, R. Padre Ladislau Kula 395, Curitiba, Paraná, 82010-210, Brazil
- School of Dentistry, Tuiuti University of Paraná, R. Padre Ladislau Kula 395, Curitiba, Paraná, 82010-210, Brazil
| |
Collapse
|
2
|
Bhardwaj P, Tyagi A, Tyagi S, Antão J, Deng Q. Machine learning model for classification of predominantly allergic and non-allergic asthma among preschool children with asthma hospitalization. J Asthma 2023; 60:487-495. [PMID: 35344453 DOI: 10.1080/02770903.2022.2059763] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/18/2022]
Abstract
OBJECTIVE Asthma is the most frequent chronic airway illness in preschool children and is difficult to diagnose due to the disease's heterogeneity. This study aimed to investigate different machine learning models and suggested the most effective one to classify two forms of asthma in preschool children (predominantly allergic asthma and non-allergic asthma) using a minimum number of features. METHODS After pre-processing, 127 patients (70 with non-allergic asthma and 57 with predominantly allergic asthma) were chosen for final analysis from the Frankfurt dataset, which had asthma-related information on 205 patients. The Random Forest algorithm and Chi-square were used to select the key features from a total of 63 features. Six machine learning models: random forest, extreme gradient boosting, support vector machines, adaptive boosting, extra tree classifier, and logistic regression were then trained and tested using 10-fold stratified cross-validation. RESULTS Among all features, age, weight, C-reactive protein, eosinophilic granulocytes, oxygen saturation, pre-medication inhaled corticosteroid + long-acting beta2-agonist (PM-ICS + LABA), PM-other (other pre-medication), H-Pulmicort/celestamine (Pulmicort/celestamine during hospitalization), and H-azithromycin (azithromycin during hospitalization) were found to be highly important. The support vector machine approach with a linear kernel was able to diffrentiate between predominantly allergic asthma and non-allergic asthma with higher accuracy (77.8%), precision (0.81), with a true positive rate of 0.73 and a true negative rate of 0.81, a F1 score of 0.81, and a ROC-AUC score of 0.79. Logistic regression was found to be the second-best classifier with an overall accuracy of 76.2%. CONCLUSION Predominantly allergic and non-allergic asthma can be classified using machine learning approaches based on nine features. Supplemental data for this article is available online at at www.tandfonline.com/ijas .
Collapse
Affiliation(s)
- Piyush Bhardwaj
- Centre for Advanced Computational Solutions (C-fACS), Department of Molecular Biosciences, Lincoln University, Lincoln, Christchurch, New Zealand
| | - Ashish Tyagi
- Department of Forensic Medicine & Toxicology, SHKM Govt. Medical College, Nuh, Haryana, India
| | - Shashank Tyagi
- Department of Forensic Medicine & Toxicology, Lady Hardinge Medical College & Associated Hospitals, New Delhi, India
| | - Joana Antão
- Lab3R-Respiratory Research and Rehabilitation Laboratory, School of Health Sciences (ESSUA), Department of Medical Sciences, Institute of Biomedicine (iBiMED), University of Aveiro, Aveiro, Portugal.,Department of Research and Education, CIRO, Horn, The Netherlands
| | - Qichen Deng
- Department of Research and Education, CIRO, Horn, The Netherlands.,Department of Respiratory Medicine, NUTRIM School of Nutrition and Translational Research in Metabolism, Maastricht University Medical Centre, Maastricht, The Netherlands.,Faculty of Health, Medicine and Life Sciences, Maastricht University Medical Centre, Limburg, The Netherlands
| |
Collapse
|