Valon Nushi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Valon Nushi has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Cristiana Palmela Pereira
Faculdade de Medicina Dentária da Universidade de Lisboa, Lisbon, Portugal
5 shared publications - Rui Santos
Centro de Estatística e Aplicações, Universidade de Lisboa (CEAUL), Lisboa, Portugal
5 shared publications - Adriana Santos
Faculdade de Medicina Dentária, Universidade de Lisboa, Lisboa, Portugal
2 shared publications - Alexandre P. Francisco
INESC-ID, Lisbon, Portugal
2 shared publications - António Figueiras
Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal
2 shared publications - Mohamed Elbawab
Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal
2 shared publications
Latest Publications
- 10 November 2026
- Published by Springer Nature in International journal of legal medicine
- p. 1-17
Deep learning for dental age assessment from panoramic radiographs: regression-based age estimation and threshold-based age classification
Introduction and objectives Deep learning has significantly advanced medical image analysis by enabling automated extraction of complex imaging features with high predictive performance. In forensic dental age assessment, however, most artificial intelligence studies focus on either chronological age estimation or legal age threshold classification, limiting direct comparisons between predictive tasks. This study aimed to evaluate deep learning models for both continuous age estimation and judicial age threshold classification using panoramic radiographs, assessing their performance, robustness and potential applicability as decision-support tools in forensic practice. Materials and methods A dataset of 9,814 anonymised digital panoramic radiographs from individuals with known chronological age was retrospectively collected and standardised through image preprocessing. Transfer learning applied to convolutional neural network architectures to address two complementary prediction tasks: chronological age estimation as a regression problem and legal age threshold classification at 10, 12, 14, 16, 18 and 21 years as independent binary classification problems. Model performance was evaluated using mean absolute error (MAE), coefficient of determination (R²), median absolute error and mean absolute percentage error (MAPE) for regression, and accuracy, precision, recall and F1-score for classification. Results The regression model achieved an MAE of 1.122 years, an R² of 0.917 and a MAPE of 7.67% on the independent test dataset, demonstrating good predictive performance across the analysed age range. Classification models achieved accuracy exceeding 90% with F1-scores above 0.90 for the 10-, 12-, 14- and 16-year legal thresholds. Lower predictive performance at the 18- and 21-year thresholds reflected increased biological variability and class imbalance associated with later stages of dental development. Conclusions Deep learning provides good predictive performance from panoramic radiographs for both continuous age estimation and legal age threshold classification. The complementary evaluation of regression and classification demonstrates the versatility of convolutional neural networks for forensic dental image analysis and supports their integration into trustworthy artificial intelligence systems for clinical and judicial decision support.
- 10 November 2026
- Published by Springer Nature in International journal of legal medicine
- p. 1-17
Deep learning for dental age assessment from panoramic radiographs: regression-based age estimation and threshold-based age classification
Introduction and objectives Deep learning has significantly advanced medical image analysis by enabling automated extraction of complex imaging features with high predictive performance. In forensic dental age assessment, however, most artificial intelligence studies focus on either chronological age estimation or legal age threshold classification, limiting direct comparisons between predictive tasks. This study aimed to evaluate deep learning models for both continuous age estimation and judicial age threshold classification using panoramic radiographs, assessing their performance, robustness and potential applicability as decision-support tools in forensic practice. Materials and methods A dataset of 9,814 anonymised digital panoramic radiographs from individuals with known chronological age was retrospectively collected and standardised through image preprocessing. Transfer learning applied to convolutional neural network architectures to address two complementary prediction tasks: chronological age estimation as a regression problem and legal age threshold classification at 10, 12, 14, 16, 18 and 21 years as independent binary classification problems. Model performance was evaluated using mean absolute error (MAE), coefficient of determination (R²), median absolute error and mean absolute percentage error (MAPE) for regression, and accuracy, precision, recall and F1-score for classification. Results The regression model achieved an MAE of 1.122 years, an R² of 0.917 and a MAPE of 7.67% on the independent test dataset, demonstrating good predictive performance across the analysed age range. Classification models achieved accuracy exceeding 90% with F1-scores above 0.90 for the 10-, 12-, 14- and 16-year legal thresholds. Lower predictive performance at the 18- and 21-year thresholds reflected increased biological variability and class imbalance associated with later stages of dental development. Conclusions Deep learning provides good predictive performance from panoramic radiographs for both continuous age estimation and legal age threshold classification. The complementary evaluation of regression and classification demonstrates the versatility of convolutional neural networks for forensic dental image analysis and supports their integration into trustworthy artificial intelligence systems for clinical and judicial decision support.
- Open Access
- 13 May 2026
- Published by Elsevier in Archives of Oral Biology
- Vol. 188, 106626
Assessing convolutional neural networks architectures for orthopantomography image analysis: Judicial dental age threshold classification
- 13 May 2026
- Published by Elsevier in Archives of Oral Biology
- Vol. 188, 106626
Open AccessAssessing convolutional neural networks architectures for orthopantomography image analysis: Judicial dental age threshold classification
- 28 March 2024
- Published by Springer Nature in International journal of legal medicine
- Vol. 138 (4) , 1533-1557
Dental age assessment and dental scoring systems: combined different statistical methods
Introduction Age Estimation has been considered as a human basic right, carried out through the use of tables for dental age assessment based on the chronology of tooth eruption. As such, the final aim of this investigation is to create tables with applicability to the Portuguese population, for the different scoring systems used and combined different statistical approaches. Materials and methods For this purpose, dental age assessment was achieved in all four third molars, using different scoring systems, in a total sample of 626 orthopantomograms (324 females, 302 males), aged between 12 and 25 years old, from the database population of Lisbon North University Hospital Center, approved by the Ethic Committee. Results The values of validation showed excellent results both on precision and on reproducibility. Mostly all methods showed statistically significant differences between the estimated age and the chronological age and, therefore, the presence of estimation errors. Kullman’s and Mincer’s methods are the ones with best applicability in the Portuguese population, in the lower third molars. The reliability measures (sensitivity, specificity and accuracy) values decrease as age increases. Conclusion A combination of the scoring systems as a protocol for dental age assessment in Portuguese nationality was established. Tables, for all the scoring systems used, were made with applicability in the Portuguese population.
- 28 March 2024
- Published by Springer Nature in International journal of legal medicine
- Vol. 138 (4) , 1533-1557
Dental age assessment and dental scoring systems: combined different statistical methods
Introduction Age Estimation has been considered as a human basic right, carried out through the use of tables for dental age assessment based on the chronology of tooth eruption. As such, the final aim of this investigation is to create tables with applicability to the Portuguese population, for the different scoring systems used and combined different statistical approaches. Materials and methods For this purpose, dental age assessment was achieved in all four third molars, using different scoring systems, in a total sample of 626 orthopantomograms (324 females, 302 males), aged between 12 and 25 years old, from the database population of Lisbon North University Hospital Center, approved by the Ethic Committee. Results The values of validation showed excellent results both on precision and on reproducibility. Mostly all methods showed statistically significant differences between the estimated age and the chronological age and, therefore, the presence of estimation errors. Kullman’s and Mincer’s methods are the ones with best applicability in the Portuguese population, in the lower third molars. The reliability measures (sensitivity, specificity and accuracy) values decrease as age increases. Conclusion A combination of the scoring systems as a protocol for dental age assessment in Portuguese nationality was established. Tables, for all the scoring systems used, were made with applicability in the Portuguese population.
- 8 May 2023
- Published by Springer Nature in International journal of legal medicine
- Vol. 137 (5) , 1497-1504
Dental age assessment: scoring systems and models from the past until the present—how is it presented in the court?
Objective The aim of this study is to evaluate the accuracy of the methods from the past and present, such as Demirjian, Cameriere, and AlQahtani, as well as to determine the most reliable method for human age estimation under the request from the courts according to the effect size from each method in age estimation. Design Four hundred eighty-three orthopantomographic images were selected from 318 patients from Centro Hospitalar Universitário Lisboa Norte aged between 6 and 15 years old. Several measurements of widths and lengths and classification of tooth development stages were performed, according to each age estimation method. The listing of patients and orthopantomographic images were consulted through SECTRA®. All data was entered and analysed using SPSS version 28. Observations’ validity was verified through inter- and intraobserver validation. Results The correlation coefficients between age and age estimates by the three methods on both sides were close to 90%. Regarding the estimation error correlation coefficient, Demirjian and AlQahtani values were low, whereas Cameriere’s was significantly negative; i.e., underestimation increases as age increases. Between left and right, any significant difference in age estimation in AlQahtani and Cameriere methods was not observed, but there were great variability and large effect for the Demirjian method. Comparing females and males, the statistical analysis showed no significant differences and quite small effects in the precision of the estimates for any of the methods. Finally, although significant differences were detected when comparing estimated values and age, small effects were revealed except for the Demirjian method which attained a medium effect and, therefore, less consistency of estimation. Conclusions Since it was not possible to determine the most reliable method for age estimation, a combined application of different age estimation methodologies is recommended to be use in the courts with relevant statistical data such as effect size.
- 8 May 2023
- Published by Springer Nature in International journal of legal medicine
- Vol. 137 (5) , 1497-1504
Dental age assessment: scoring systems and models from the past until the present—how is it presented in the court?
Objective The aim of this study is to evaluate the accuracy of the methods from the past and present, such as Demirjian, Cameriere, and AlQahtani, as well as to determine the most reliable method for human age estimation under the request from the courts according to the effect size from each method in age estimation. Design Four hundred eighty-three orthopantomographic images were selected from 318 patients from Centro Hospitalar Universitário Lisboa Norte aged between 6 and 15 years old. Several measurements of widths and lengths and classification of tooth development stages were performed, according to each age estimation method. The listing of patients and orthopantomographic images were consulted through SECTRA®. All data was entered and analysed using SPSS version 28. Observations’ validity was verified through inter- and intraobserver validation. Results The correlation coefficients between age and age estimates by the three methods on both sides were close to 90%. Regarding the estimation error correlation coefficient, Demirjian and AlQahtani values were low, whereas Cameriere’s was significantly negative; i.e., underestimation increases as age increases. Between left and right, any significant difference in age estimation in AlQahtani and Cameriere methods was not observed, but there were great variability and large effect for the Demirjian method. Comparing females and males, the statistical analysis showed no significant differences and quite small effects in the precision of the estimates for any of the methods. Finally, although significant differences were detected when comparing estimated values and age, small effects were revealed except for the Demirjian method which attained a medium effect and, therefore, less consistency of estimation. Conclusions Since it was not possible to determine the most reliable method for age estimation, a combined application of different age estimation methodologies is recommended to be use in the courts with relevant statistical data such as effect size.
- Open Access
- 15 March 2023
- Published by University of Zagreb School of Dental Medicine in Acta Stomatologica Croatica
- Vol. 57 (1) , 43-51
Retrospective Study of Oral and Maxillofacial Trauma in Portuguese Population
Objective: The increasing significance of medicolegal evaluation following maxillofacial traumatic events constitutes a complex issue. This clinical research aimed to assess the current etiology of oral and maxillofacial injuries in Portuguese population. Material and methods: An epidemiological clin-ical observational study was conducted in Centro Hospitalar Lisboa Norte on a sample of 384 sub-jects diagnosed with oral and maxillofacial trauma, between 2018 and 2020. Data were collected through clinical reports and analyzed using SPSS version 27. Results: Women and men were nearly identical in the number and distribution, with 49.5% females and 50.5% males. In 2020, there was a decrease in the number of traumatic incidents compared to other years. Falls or accidental descents were found to be the most common cause of injury, accounting for 44.3%, followed by assaults ac-counting for 24.7%. A total of 84 subjects exhibited soft tissue injuries related to periodontal region. The upper central incisors (174) were the most frequently affected teeth with uncomplicated frac-tures, and the predominant form of treatment was the administration of pain medication. Conclu-sion: A correlation between falls or accidental descents, female subjects, and advancing age, as well as between assaults, male subjects, and adults, has been established. Falls or accidental descents and assault were the predominant etiologies, and the year 2020 exhibited a decrease in the incidents of traumatic events.
- 15 March 2023
- Published by University of Zagreb School of Dental Medicine in Acta Stomatologica Croatica
- Vol. 57 (1) , 43-51
Open AccessRetrospective Study of Oral and Maxillofacial Trauma in Portuguese Population
Objective: The increasing significance of medicolegal evaluation following maxillofacial traumatic events constitutes a complex issue. This clinical research aimed to assess the current etiology of oral and maxillofacial injuries in Portuguese population. Material and methods: An epidemiological clin-ical observational study was conducted in Centro Hospitalar Lisboa Norte on a sample of 384 sub-jects diagnosed with oral and maxillofacial trauma, between 2018 and 2020. Data were collected through clinical reports and analyzed using SPSS version 27. Results: Women and men were nearly identical in the number and distribution, with 49.5% females and 50.5% males. In 2020, there was a decrease in the number of traumatic incidents compared to other years. Falls or accidental descents were found to be the most common cause of injury, accounting for 44.3%, followed by assaults ac-counting for 24.7%. A total of 84 subjects exhibited soft tissue injuries related to periodontal region. The upper central incisors (174) were the most frequently affected teeth with uncomplicated frac-tures, and the predominant form of treatment was the administration of pain medication. Conclu-sion: A correlation between falls or accidental descents, female subjects, and advancing age, as well as between assaults, male subjects, and adults, has been established. Falls or accidental descents and assault were the predominant etiologies, and the year 2020 exhibited a decrease in the incidents of traumatic events.