Alberto Argentiero
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Alberto Argentiero has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Michele Maffia
Department of Experimental Medicine, University of Salento, Via Lecce-Monteroni, 73100 Lecce, Italy
4 shared publications - Andrea Buccoliero
Department of Research and Development (R&D), GPI SpA, 38123 Trento, Italy
3 shared publications - Angela Lupo
Division of Neurology, Vito Fazzi Hospital, 73100 Lecce, Italy
3 shared publications - Chiara Coppola
Department of Experimental Medicine, University of Salento, Via Lecce-Monteroni, 73100 Lecce, Italy
3 shared publications - Filomena My
Division of Neurology, Vito Fazzi Hospital, 73100 Lecce, Italy
3 shared publications - Marcello Donzella
Department of Research and Development (R&D), GPI SpA, 38123 Trento, Italy
3 shared publications
Latest Publications
- Open Access
- 11 September 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (9) , 1058
A Pilot Study of VGGish-CNN as a Model for the Classification of Parkinson’s Disease from a Control Group Using Speech Impairment, Integrated with Explainable Artificial Intelligence
Introduction: Voice-based digital biomarkers have emerged as a promising, non-invasive approach for the early detection and monitoring of neurodegenerative disorders, particularly Parkinson’s disease (PD). Although voice recordings can be acquired easily using mobile health technologies, their integration into routine clinical practice remains limited due to challenges related to model interpretability and clinical validation. This study aimed to investigate the diagnostic potential of voice recordings acquired through the TALIA smartphone and web-based platform and to improve model transparency using explainable artificial intelligence (XAI) techniques. Methods: Voice recordings from participants with PD and control group (CG) were acquired over a six-month period using the TALIA digital-health platform at Vito Fazzi Hospital in Lecce, Italy. The data were evaluated cross-sectionally at the recording level. Sustained vowel (/a/) phonations were preprocessed and transformed into spectrogram images. A transfer-learning framework based on the pre-trained VGGish convolutional neural network (VGGish-CNN) was developed to classify PD and CG voice recordings. To enhance interpretability, explainable artificial intelligence (XAI) methods, including Local Interpretable Model-Agnostic Explanations (LIME) and Occlusion Sensitivity, were integrated to identify the spectro-temporal regions contributing most strongly to model predictions. Results: The proposed VGGish-CNN framework demonstrated excellent classification performance in distinguishing PD from CG. The model achieved an accuracy of 0.93, precision of 0.91, recall of 0.95, F1-score of 0.93, loss of 0.16, and an area under the receiver operating characteristic curve (AUC) of 0.987. XAI analyses provided qualitative, sample-specific visualizations of the spectro-temporal regions contributing to individual model predictions, thereby improving the transparency and interpretability of the deep-learning model. Conclusions: The findings demonstrate that a transfer-learning approach based on VGGish-CNN could achieve promising recording-level classification performance in distinguishing voice recordings from participants with PD and CG within this pilot dataset. Furthermore, XAI techniques provide qualitative insight into the model’s decision-making process. These preliminary findings support further investigation of voice-based approaches for PD screening in larger cohorts using participant-level and external validation.
- 11 September 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (9) , 1058
Open AccessA Pilot Study of VGGish-CNN as a Model for the Classification of Parkinson’s Disease from a Control Group Using Speech Impairment, Integrated with Explainable Artificial Intelligence
Introduction: Voice-based digital biomarkers have emerged as a promising, non-invasive approach for the early detection and monitoring of neurodegenerative disorders, particularly Parkinson’s disease (PD). Although voice recordings can be acquired easily using mobile health technologies, their integration into routine clinical practice remains limited due to challenges related to model interpretability and clinical validation. This study aimed to investigate the diagnostic potential of voice recordings acquired through the TALIA smartphone and web-based platform and to improve model transparency using explainable artificial intelligence (XAI) techniques. Methods: Voice recordings from participants with PD and control group (CG) were acquired over a six-month period using the TALIA digital-health platform at Vito Fazzi Hospital in Lecce, Italy. The data were evaluated cross-sectionally at the recording level. Sustained vowel (/a/) phonations were preprocessed and transformed into spectrogram images. A transfer-learning framework based on the pre-trained VGGish convolutional neural network (VGGish-CNN) was developed to classify PD and CG voice recordings. To enhance interpretability, explainable artificial intelligence (XAI) methods, including Local Interpretable Model-Agnostic Explanations (LIME) and Occlusion Sensitivity, were integrated to identify the spectro-temporal regions contributing most strongly to model predictions. Results: The proposed VGGish-CNN framework demonstrated excellent classification performance in distinguishing PD from CG. The model achieved an accuracy of 0.93, precision of 0.91, recall of 0.95, F1-score of 0.93, loss of 0.16, and an area under the receiver operating characteristic curve (AUC) of 0.987. XAI analyses provided qualitative, sample-specific visualizations of the spectro-temporal regions contributing to individual model predictions, thereby improving the transparency and interpretability of the deep-learning model. Conclusions: The findings demonstrate that a transfer-learning approach based on VGGish-CNN could achieve promising recording-level classification performance in distinguishing voice recordings from participants with PD and CG within this pilot dataset. Furthermore, XAI techniques provide qualitative insight into the model’s decision-making process. These preliminary findings support further investigation of voice-based approaches for PD screening in larger cohorts using participant-level and external validation.
- Open Access
- 10 July 2025
- Published by MDPI AG in Brain Sciences
- Vol. 15 (7) , 739
Prediction of Parkinson Disease Using Long-Term, Short-Term Acoustic Features Based on Machine Learning
Background: Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease, affecting countless individuals worldwide. PD is characterized by the onset of a marked motor symptomatology in association with several non-motor manifestations. The clinical phase of the disease is usually preceded by a long prodromal phase, devoid of overt motor symptomatology but often showing some conditions such as sleep disturbance, constipation, anosmia, and phonatory changes. To date, speech analysis appears to be a promising digital biomarker to anticipate even 10 years before the onset of clinical PD, as well serving as a useful prognostic tool for patient follow-up. That is why, the voice can be nominated as the non-invasive method to detect PD from healthy subjects (HS). Methods: Our study was based on cross-sectional study to analysis voice impairment. A dataset comprising 81 voice samples (41 from healthy individuals and 40 from PD patients) was utilized to train and evaluate common machine learning (ML) models using various types of features, including long-term (jitter, shimmer, and cepstral peak prominence (CPP)), short-term features (Mel-frequency cepstral coefficient (MFCC)), and non-standard measurements (pitch period entropy (PPE) and recurrence period density entropy (RPDE)). The study adopted multiple machine learning (ML) algorithms, including random forest (RF), K-nearest neighbors (KNN), decision tree (DT), naïve Bayes (NB), support vector machines (SVM), and logistic regression (LR). Cross-validation technique was applied to ensure the reliability of performance metrics on train and test subsets. These metrics (accuracy, recall, and precision), help determine the most effective models for distinguishing PD from healthy subjects. Result: Among all the algorithms used in this research, random forest (RF) was the best-performing model, achieving an accuracy of 82.72% with a ROC-AUC score of 89.65%. Although other models, such as support vector machine (SVM), could be considered with an accuracy of 75.29% and a ROC-AUC score of 82.63%, RF was by far the best one when evaluated across all metrics. The K-nearest neighbor (KNN) and decision tree (DT) performed the worst. Notably, by combining a comprehensive set of long-term, short-term, and non-standard acoustic features, unlike previous studies that typically focused on only a subset, our study achieved higher predictive performance, offering a more robust model for early PD detection. Conclusions: This study highlights the potential of combining advanced acoustic analysis with ML algorithms to develop non-invasive and reliable tools for early PD detection, offering substantial benefits for the healthcare sector.
- 10 July 2025
- Published by MDPI AG in Brain Sciences
- Vol. 15 (7) , 739
Open AccessPrediction of Parkinson Disease Using Long-Term, Short-Term Acoustic Features Based on Machine Learning
Background: Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease, affecting countless individuals worldwide. PD is characterized by the onset of a marked motor symptomatology in association with several non-motor manifestations. The clinical phase of the disease is usually preceded by a long prodromal phase, devoid of overt motor symptomatology but often showing some conditions such as sleep disturbance, constipation, anosmia, and phonatory changes. To date, speech analysis appears to be a promising digital biomarker to anticipate even 10 years before the onset of clinical PD, as well serving as a useful prognostic tool for patient follow-up. That is why, the voice can be nominated as the non-invasive method to detect PD from healthy subjects (HS). Methods: Our study was based on cross-sectional study to analysis voice impairment. A dataset comprising 81 voice samples (41 from healthy individuals and 40 from PD patients) was utilized to train and evaluate common machine learning (ML) models using various types of features, including long-term (jitter, shimmer, and cepstral peak prominence (CPP)), short-term features (Mel-frequency cepstral coefficient (MFCC)), and non-standard measurements (pitch period entropy (PPE) and recurrence period density entropy (RPDE)). The study adopted multiple machine learning (ML) algorithms, including random forest (RF), K-nearest neighbors (KNN), decision tree (DT), naïve Bayes (NB), support vector machines (SVM), and logistic regression (LR). Cross-validation technique was applied to ensure the reliability of performance metrics on train and test subsets. These metrics (accuracy, recall, and precision), help determine the most effective models for distinguishing PD from healthy subjects. Result: Among all the algorithms used in this research, random forest (RF) was the best-performing model, achieving an accuracy of 82.72% with a ROC-AUC score of 89.65%. Although other models, such as support vector machine (SVM), could be considered with an accuracy of 75.29% and a ROC-AUC score of 82.63%, RF was by far the best one when evaluated across all metrics. The K-nearest neighbor (KNN) and decision tree (DT) performed the worst. Notably, by combining a comprehensive set of long-term, short-term, and non-standard acoustic features, unlike previous studies that typically focused on only a subset, our study achieved higher predictive performance, offering a more robust model for early PD detection. Conclusions: This study highlights the potential of combining advanced acoustic analysis with ML algorithms to develop non-invasive and reliable tools for early PD detection, offering substantial benefits for the healthcare sector.
- Open Access
- 9 April 2025
- Published by MDPI AG in Preprints.org
Prediction of Parkinson Disease Using Long-term, Short-Term Acoustic Features Based on Machine Learning
Background: Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease affecting countless individuals worldwide. PD is characterized by the onset of a marked motor symptomatology in association with several nonmotor manifestations. The clinical phase of the disease is usually preceded by a long prodromal phase, devoid of overt motor symptomatology but often showing some conditions such as sleep disturbance, constipation, anosmia, and phonatory changes. To date, speech analysis appears to be a promising digital biomarker to anticipate even 10 years the onset of clinical PD, as well serving as a useful prognostic tool for patient follow-up. That is why, the voice can be nominated as the non-invasive method to detect PD from healthy subjects (HS).
- 9 April 2025
- Published by MDPI AG in Preprints.org
Open AccessPrediction of Parkinson Disease Using Long-term, Short-Term Acoustic Features Based on Machine Learning
Background: Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease affecting countless individuals worldwide. PD is characterized by the onset of a marked motor symptomatology in association with several nonmotor manifestations. The clinical phase of the disease is usually preceded by a long prodromal phase, devoid of overt motor symptomatology but often showing some conditions such as sleep disturbance, constipation, anosmia, and phonatory changes. To date, speech analysis appears to be a promising digital biomarker to anticipate even 10 years the onset of clinical PD, as well serving as a useful prognostic tool for patient follow-up. That is why, the voice can be nominated as the non-invasive method to detect PD from healthy subjects (HS).
- Open Access
- 28 October 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 19 (21) , 14023
Trends of Phase I Clinical Trials in the Latest Ten Years across Five European Countries
Background: Phase 1 clinical trials represent a critical phase of drug development because new candidate therapeutic agents are tested for the first time on humans. Therefore, international guidelines and local laws have been released to mitigate and control possible risks for human health in agreement with the declaration of Helsinki and the international Good Clinical Practice principles. Despite numerous scientific works characterizing the registered clinical trials on ClinicalTrials.gov, the main features and trends of registered phase 1 clinical trials in Europe have not been investigated. This study is aimed at assessing the features and the temporal trend of distribution of phase 1 clinical studies, carried out in the five largest European countries over a ten-year period (2012–2021), and to evaluate the impact of the Italian regulatory framework on the activation of such studies. Methods: The main data and characteristics of phase 1 clinical studies registered on the ClinicalTrials.gov database for France, Germany, Italy, Spain and the United Kingdom have been investigated and subsequently compared. The above-mentioned countries were selected based on similarities in terms of demographic and Gross Domestic Product (GDP) data available on official government websites. (3) Results: A total number of 6878 phase 1 clinical trials were registered for the five selected countries in the ClinicalTrials.gov database during the ten years analyzed; the studies were predominantly randomized (39.33%) and for-profit (76.64%). The most represented area of investigations was oncology (52.15%), followed by hematology (24.99%) and immunology (12.04%). The variability observed between the analyzed countries showed that the UK, Germany and France presented the highest reduction in the number of phase 1 clinical trials, while for Spain and Italy, a stable/increased trend was observed, although with a lower number of trials registered on the ClinicalTrials.gov database. (4) Conclusions: Italy displayed the lowest number of registered phase 1 clinical trials, even though it showed a stable trend over the years. In this regard, the Italian regulatory framework must urgently be adapted to that of other European countries (Spain has been the first country to implement the new Regulation (EU) No 536/2014) and streamline the process of clinical trial application to increase the attractiveness of the country. Moreover, nonprofit phase 1 clinical trials (which represent 19.81% of the total number of phase 1 clinical trials registered in Italy vs. 80.19% of profit phase 1 clinical studies) should be promoted and supported by the institutions, even from a financial point of view, to allow independent researchers to develop new therapeutic drugs.
- 28 October 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 19 (21) , 14023
Open AccessTrends of Phase I Clinical Trials in the Latest Ten Years across Five European Countries
Background: Phase 1 clinical trials represent a critical phase of drug development because new candidate therapeutic agents are tested for the first time on humans. Therefore, international guidelines and local laws have been released to mitigate and control possible risks for human health in agreement with the declaration of Helsinki and the international Good Clinical Practice principles. Despite numerous scientific works characterizing the registered clinical trials on ClinicalTrials.gov, the main features and trends of registered phase 1 clinical trials in Europe have not been investigated. This study is aimed at assessing the features and the temporal trend of distribution of phase 1 clinical studies, carried out in the five largest European countries over a ten-year period (2012–2021), and to evaluate the impact of the Italian regulatory framework on the activation of such studies. Methods: The main data and characteristics of phase 1 clinical studies registered on the ClinicalTrials.gov database for France, Germany, Italy, Spain and the United Kingdom have been investigated and subsequently compared. The above-mentioned countries were selected based on similarities in terms of demographic and Gross Domestic Product (GDP) data available on official government websites. (3) Results: A total number of 6878 phase 1 clinical trials were registered for the five selected countries in the ClinicalTrials.gov database during the ten years analyzed; the studies were predominantly randomized (39.33%) and for-profit (76.64%). The most represented area of investigations was oncology (52.15%), followed by hematology (24.99%) and immunology (12.04%). The variability observed between the analyzed countries showed that the UK, Germany and France presented the highest reduction in the number of phase 1 clinical trials, while for Spain and Italy, a stable/increased trend was observed, although with a lower number of trials registered on the ClinicalTrials.gov database. (4) Conclusions: Italy displayed the lowest number of registered phase 1 clinical trials, even though it showed a stable trend over the years. In this regard, the Italian regulatory framework must urgently be adapted to that of other European countries (Spain has been the first country to implement the new Regulation (EU) No 536/2014) and streamline the process of clinical trial application to increase the attractiveness of the country. Moreover, nonprofit phase 1 clinical trials (which represent 19.81% of the total number of phase 1 clinical trials registered in Italy vs. 80.19% of profit phase 1 clinical studies) should be promoted and supported by the institutions, even from a financial point of view, to allow independent researchers to develop new therapeutic drugs.