Pub Date : 2024-08-09DOI: 10.3389/frai.2024.1336320
Azat Absadyk, Olzhas Turar, Darkhan Akhmed-Zaki
The COVID-19 pandemic highlighted the need for accurate virtual sizing in e-commerce to reduce returns and waste. Existing methods for extracting anthropometric data from images have limitations. This study aims to develop a semantic segmentation model trained on synthetic data that can accurately determine body shape from real images, accounting for clothing.A synthetic dataset of over 22,000 images was created using NVIDIA Omniverse Replicator, featuring human models in various poses, clothing, and environments. Popular CNN architectures (U-Net, SegNet, DeepLabV3, PSPNet) with different backbones were trained on this dataset for semantic segmentation. Models were evaluated on accuracy, precision, recall, and IoU metrics. The best performing model was tested on real human subjects and compared to actual measurements.U-Net with EfficientNet backbone showed the best performance, with 99.83% training accuracy and 0.977 IoU score. When tested on real images, it accurately segmented body shape while accounting for clothing. Comparison with actual measurements on 9 subjects showed average deviations of −0.24 cm for neck, −0.1 cm for shoulder, 1.15 cm for chest, −0.22 cm for thallium, and 0.17 cm for hip measurements.The synthetic dataset and trained models enable accurate extraction of anthropometric data from real images while accounting for clothing. This approach has significant potential for improving virtual fitting and reducing returns in e-commerce. Future work will focus on refining the algorithm, particularly for thallium and hip measurements which showed higher variability.
{"title":"Using synthetic dataset for semantic segmentation of the human body in the problem of extracting anthropometric data","authors":"Azat Absadyk, Olzhas Turar, Darkhan Akhmed-Zaki","doi":"10.3389/frai.2024.1336320","DOIUrl":"https://doi.org/10.3389/frai.2024.1336320","url":null,"abstract":"The COVID-19 pandemic highlighted the need for accurate virtual sizing in e-commerce to reduce returns and waste. Existing methods for extracting anthropometric data from images have limitations. This study aims to develop a semantic segmentation model trained on synthetic data that can accurately determine body shape from real images, accounting for clothing.A synthetic dataset of over 22,000 images was created using NVIDIA Omniverse Replicator, featuring human models in various poses, clothing, and environments. Popular CNN architectures (U-Net, SegNet, DeepLabV3, PSPNet) with different backbones were trained on this dataset for semantic segmentation. Models were evaluated on accuracy, precision, recall, and IoU metrics. The best performing model was tested on real human subjects and compared to actual measurements.U-Net with EfficientNet backbone showed the best performance, with 99.83% training accuracy and 0.977 IoU score. When tested on real images, it accurately segmented body shape while accounting for clothing. Comparison with actual measurements on 9 subjects showed average deviations of −0.24 cm for neck, −0.1 cm for shoulder, 1.15 cm for chest, −0.22 cm for thallium, and 0.17 cm for hip measurements.The synthetic dataset and trained models enable accurate extraction of anthropometric data from real images while accounting for clothing. This approach has significant potential for improving virtual fitting and reducing returns in e-commerce. Future work will focus on refining the algorithm, particularly for thallium and hip measurements which showed higher variability.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"75 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141922330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-08DOI: 10.3389/frai.2024.1404940
Haoyuan Chen, Nuobei Shi, Ling Chen, Raymond S. T. Lee
Online question-and-answer (Q&A) platforms are frequently replete with extensive human resource support. This study proposes a novel methodology of a customized large language model (LLM) called Chaotic LLM-based Educational Q&A System (CHAQS) to navigate the complexities associated with intelligent Q&A systems for the educational sector.It uses an expansive dataset comprising over 383,000 educational data pairs, an intricate fine-tuning process encompassing p-tuning v2, low-rank adaptation (LRA), and strategies for parameter freezing at an open-source large language model ChatGLM as a baseline model. In addition, Fuzzy Logic is implemented to regulate parameters and the system's adaptability with the Lee Oscillator to refine the model's response variability and precision.Experiment results showed a 5.12% improvement in precision score, an 11% increase in recall metric, and an 8% improvement in the F1 score as compared to other models.These results suggest that the CHAQS methodology significantly enhances the performance of educational Q&A systems, demonstrating the effectiveness of combining advanced tuning techniques and fuzzy logic for improved model precision and adaptability.
{"title":"Enhancing educational Q&A systems using a Chaotic Fuzzy Logic-Augmented large language model","authors":"Haoyuan Chen, Nuobei Shi, Ling Chen, Raymond S. T. Lee","doi":"10.3389/frai.2024.1404940","DOIUrl":"https://doi.org/10.3389/frai.2024.1404940","url":null,"abstract":"Online question-and-answer (Q&A) platforms are frequently replete with extensive human resource support. This study proposes a novel methodology of a customized large language model (LLM) called Chaotic LLM-based Educational Q&A System (CHAQS) to navigate the complexities associated with intelligent Q&A systems for the educational sector.It uses an expansive dataset comprising over 383,000 educational data pairs, an intricate fine-tuning process encompassing p-tuning v2, low-rank adaptation (LRA), and strategies for parameter freezing at an open-source large language model ChatGLM as a baseline model. In addition, Fuzzy Logic is implemented to regulate parameters and the system's adaptability with the Lee Oscillator to refine the model's response variability and precision.Experiment results showed a 5.12% improvement in precision score, an 11% increase in recall metric, and an 8% improvement in the F1 score as compared to other models.These results suggest that the CHAQS methodology significantly enhances the performance of educational Q&A systems, demonstrating the effectiveness of combining advanced tuning techniques and fuzzy logic for improved model precision and adaptability.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"4 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141926935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-25DOI: 10.3389/frai.2024.1328530
Ali Ahmad, Anderson X. W. Liew, Francesca Venturini, Athanasios Kalogeras, Alessandro Candiani, Giacomo di Benedetto, Segun Ajibola, Pedro Cartujo, Pablo Romero, Aspasia Lykoudi, Michelangelo Mastrorocco De Grandis, Christos Xouris, Riccardo Lo Bianco, Irawan Doddy, Isa Elegbede, Giuseppe Falvo D'Urso Labate, Luis F. García del Moral, Vanessa M. Martos
Food and nutrition are a steadfast essential to all living organisms. With specific reference to humans, the sufficient and efficient supply of food is a challenge as the world population continues to grow. Artificial Intelligence (AI) could be identified as a plausible technology in this 5th industrial revolution in bringing us closer to achieving zero hunger by 2030—Goal 2 of the United Nations Sustainable Development Goals (UNSDG). This goal cannot be achieved unless the digital divide among developed and underdeveloped countries is addressed. Nevertheless, developing and underdeveloped regions fall behind in economic resources; however, they harbor untapped potential to effectively address the impending demands posed by the soaring world population. Therefore, this study explores the in-depth potential of AI in the agriculture sector for developing and under-developed countries. Similarly, it aims to emphasize the proven efficiency and spin-off applications of AI in the advancement of agriculture. Currently, AI is being utilized in various spheres of agriculture, including but not limited to crop surveillance, irrigation management, disease identification, fertilization practices, task automation, image manipulation, data processing, yield forecasting, supply chain optimization, implementation of decision support system (DSS), weed control, and the enhancement of resource utilization. Whereas AI supports food safety and security by ensuring higher crop yields that are acquired by harnessing the potential of multi-temporal remote sensing (RS) techniques to accurately discern diverse crop phenotypes, monitor land cover dynamics, assess variations in soil organic matter, predict soil moisture levels, conduct plant biomass modeling, and enable comprehensive crop monitoring. The present study identifies various challenges, including financial, infrastructure, experts, data availability, customization, regulatory framework, cultural norms and attitudes, access to market, and interdisciplinary collaboration, in the adoption of AI for developing nations with their subsequent remedies. The identification of challenges and opportunities in the implementation of AI could ignite further research and actions in these regions; thereby supporting sustainable development.
{"title":"AI can empower agriculture for global food security: challenges and prospects in developing nations","authors":"Ali Ahmad, Anderson X. W. Liew, Francesca Venturini, Athanasios Kalogeras, Alessandro Candiani, Giacomo di Benedetto, Segun Ajibola, Pedro Cartujo, Pablo Romero, Aspasia Lykoudi, Michelangelo Mastrorocco De Grandis, Christos Xouris, Riccardo Lo Bianco, Irawan Doddy, Isa Elegbede, Giuseppe Falvo D'Urso Labate, Luis F. García del Moral, Vanessa M. Martos","doi":"10.3389/frai.2024.1328530","DOIUrl":"https://doi.org/10.3389/frai.2024.1328530","url":null,"abstract":"Food and nutrition are a steadfast essential to all living organisms. With specific reference to humans, the sufficient and efficient supply of food is a challenge as the world population continues to grow. Artificial Intelligence (AI) could be identified as a plausible technology in this 5th industrial revolution in bringing us closer to achieving zero hunger by 2030—Goal 2 of the United Nations Sustainable Development Goals (UNSDG). This goal cannot be achieved unless the digital divide among developed and underdeveloped countries is addressed. Nevertheless, developing and underdeveloped regions fall behind in economic resources; however, they harbor untapped potential to effectively address the impending demands posed by the soaring world population. Therefore, this study explores the in-depth potential of AI in the agriculture sector for developing and under-developed countries. Similarly, it aims to emphasize the proven efficiency and spin-off applications of AI in the advancement of agriculture. Currently, AI is being utilized in various spheres of agriculture, including but not limited to crop surveillance, irrigation management, disease identification, fertilization practices, task automation, image manipulation, data processing, yield forecasting, supply chain optimization, implementation of decision support system (DSS), weed control, and the enhancement of resource utilization. Whereas AI supports food safety and security by ensuring higher crop yields that are acquired by harnessing the potential of multi-temporal remote sensing (RS) techniques to accurately discern diverse crop phenotypes, monitor land cover dynamics, assess variations in soil organic matter, predict soil moisture levels, conduct plant biomass modeling, and enable comprehensive crop monitoring. The present study identifies various challenges, including financial, infrastructure, experts, data availability, customization, regulatory framework, cultural norms and attitudes, access to market, and interdisciplinary collaboration, in the adoption of AI for developing nations with their subsequent remedies. The identification of challenges and opportunities in the implementation of AI could ignite further research and actions in these regions; thereby supporting sustainable development.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140653871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-25DOI: 10.3389/frai.2023.1237285
Sirinant Khunakornbodintr
China's commitment to achieving carbon neutrality by 2060 has sparked scholars' interest in examining the environmental ramifications of green technologies in the digital era. While plenty of them provide eco-efficiency policy such as increasing R&D investment or stimulating green exports, little attention has been paid to the firm-level technological management and recombination strategies such as differentiation/specialization of green portfolios along with AI integration, which can significantly impact the pace of net-zero transitions. To address these gaps, this study investigates the moderating effect of technological specialization on levels of AI integration into green technologies estimated by green-AI technological distance and enterprises' innovation performance in Chinese contemporary contexts. Regression results of fixed-effect model in Chinese patent data (2011–2020) indicate that enterprises' green innovation performance is significantly improved as AI integrates more into the green technologies due to the legitimacy and the inability to appropriate more green values. Interestingly, specialized green-technological enterprises demonstrate superior performance in integrating distant AI technologies. This occurrence could potentially be driven by the governments' incentives and the organization's risk attitudes, shaping green innovation outcomes. Hence, the study underscores the importance of considering both the AI integration and green specialization in shaping innovation outcomes amidst green transitions.
{"title":"Examining the impact of green technological specialization and the integration of AI technologies on green innovation performance: evidence from China","authors":"Sirinant Khunakornbodintr","doi":"10.3389/frai.2023.1237285","DOIUrl":"https://doi.org/10.3389/frai.2023.1237285","url":null,"abstract":"China's commitment to achieving carbon neutrality by 2060 has sparked scholars' interest in examining the environmental ramifications of green technologies in the digital era. While plenty of them provide eco-efficiency policy such as increasing R&D investment or stimulating green exports, little attention has been paid to the firm-level technological management and recombination strategies such as differentiation/specialization of green portfolios along with AI integration, which can significantly impact the pace of net-zero transitions. To address these gaps, this study investigates the moderating effect of technological specialization on levels of AI integration into green technologies estimated by green-AI technological distance and enterprises' innovation performance in Chinese contemporary contexts. Regression results of fixed-effect model in Chinese patent data (2011–2020) indicate that enterprises' green innovation performance is significantly improved as AI integrates more into the green technologies due to the legitimacy and the inability to appropriate more green values. Interestingly, specialized green-technological enterprises demonstrate superior performance in integrating distant AI technologies. This occurrence could potentially be driven by the governments' incentives and the organization's risk attitudes, shaping green innovation outcomes. Hence, the study underscores the importance of considering both the AI integration and green specialization in shaping innovation outcomes amidst green transitions.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140658016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-23DOI: 10.3389/frai.2024.1377337
Yiting Li, Sichao Tian, Haiyue Zhu, Yeying Jin, Keqing Wang, Jun Ma, Cheng Xiang, P. Vadakkepat
This study aims at addressing the challenging incremental few-shot object detection (iFSOD) problem toward online adaptive detection. iFSOD targets to learn novel categories in a sequential manner, and eventually, the detection is performed on all learned categories. Moreover, only a few training samples are available for all sequential novel classes in these situations. In this study, we propose an efficient yet suitably simple framework, Expandable-RCNN, as a solution for the iFSOD problem, which allows online sequentially adding new classes with zero retraining of the base network. We achieve this by adapting the Faster R-CNN to the few-shot learning scenario with two elegant components to effectively address the overfitting and category bias. First, an IOU-aware weight imprinting strategy is proposed to directly determine the classifier weights for incremental novel classes and the background class, which is with zero training to avoid the notorious overfitting issue in few-shot learning. Second, since the above zero-retraining imprinting approach may lead to undesired category bias in the classifier, we develop a bias correction module for iFSOD, named the group soft-max layer (GSL), that efficiently calibrates the biased prediction of the imprinted classifier to organically improve classification performance for the few-shot classes, preventing catastrophic forgetting. Extensive experiments on MS-COCO show that our method can significantly outperform the state-of-the-art method ONCE by 5.9 points in commonly encountered few-shot classes.
{"title":"Expandable-RCNN: toward high-efficiency incremental few-shot object detection","authors":"Yiting Li, Sichao Tian, Haiyue Zhu, Yeying Jin, Keqing Wang, Jun Ma, Cheng Xiang, P. Vadakkepat","doi":"10.3389/frai.2024.1377337","DOIUrl":"https://doi.org/10.3389/frai.2024.1377337","url":null,"abstract":"This study aims at addressing the challenging incremental few-shot object detection (iFSOD) problem toward online adaptive detection. iFSOD targets to learn novel categories in a sequential manner, and eventually, the detection is performed on all learned categories. Moreover, only a few training samples are available for all sequential novel classes in these situations. In this study, we propose an efficient yet suitably simple framework, Expandable-RCNN, as a solution for the iFSOD problem, which allows online sequentially adding new classes with zero retraining of the base network. We achieve this by adapting the Faster R-CNN to the few-shot learning scenario with two elegant components to effectively address the overfitting and category bias. First, an IOU-aware weight imprinting strategy is proposed to directly determine the classifier weights for incremental novel classes and the background class, which is with zero training to avoid the notorious overfitting issue in few-shot learning. Second, since the above zero-retraining imprinting approach may lead to undesired category bias in the classifier, we develop a bias correction module for iFSOD, named the group soft-max layer (GSL), that efficiently calibrates the biased prediction of the imprinted classifier to organically improve classification performance for the few-shot classes, preventing catastrophic forgetting. Extensive experiments on MS-COCO show that our method can significantly outperform the state-of-the-art method ONCE by 5.9 points in commonly encountered few-shot classes.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"117 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140669513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-19DOI: 10.3389/frai.2024.1385871
A. Mencattini, Elena Daprati, David Della-Morte, Fiorella Guadagni, Federica Sangiuolo, Eugenio Martinelli
{"title":"Assembloid learning: opportunities and challenges for personalized approaches to brain functioning in health and disease","authors":"A. Mencattini, Elena Daprati, David Della-Morte, Fiorella Guadagni, Federica Sangiuolo, Eugenio Martinelli","doi":"10.3389/frai.2024.1385871","DOIUrl":"https://doi.org/10.3389/frai.2024.1385871","url":null,"abstract":"","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":" 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140683257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Editorial: Human-centered AI at work: common ground in theories and methods","authors":"Annette Kluge, Uta Wilkens, Verena Nitsch, Corinna Peifer","doi":"10.3389/frai.2024.1411795","DOIUrl":"https://doi.org/10.3389/frai.2024.1411795","url":null,"abstract":"","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":" 71","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140691895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-16DOI: 10.3389/frai.2024.1396160
Vatsala Anand, Deepika Koundal, Wael Y. Alghamdi, Bayan M. Alsharbi
Diabetic retinopathy is a condition that affects the retina and causes vision loss due to blood vessel destruction. The retina is the layer of the eye responsible for visual processing and nerve signaling. Diabetic retinopathy causes vision loss, floaters, and sometimes blindness; however, it often shows no warning signals in the early stages. Deep learning-based techniques have emerged as viable options for automated illness classification as large-scale medical imaging datasets have become more widely available. To adapt to medical image analysis tasks, transfer learning makes use of pre-trained models to extract high-level characteristics from natural images. In this research, an intelligent recommendation-based fine-tuned EfficientNetB0 model has been proposed for quick and precise assessment for the diagnosis of diabetic retinopathy from fundus images, which will help ophthalmologists in early diagnosis and detection. The proposed EfficientNetB0 model is compared with three transfer learning-based models, namely, ResNet152, VGG16, and DenseNet169. The experimental work is carried out using publicly available datasets from Kaggle consisting of 3,200 fundus images. Out of all the transfer learning models, the EfficientNetB0 model has outperformed with an accuracy of 0.91, followed by DenseNet169 with an accuracy of 0.90. In comparison to other approaches, the proposed intelligent recommendation-based fine-tuned EfficientNetB0 approach delivers state-of-the-art performance on the accuracy, recall, precision, and F1-score criteria. The system aims to assist ophthalmologists in early detection, potentially alleviating the burden on healthcare units.
{"title":"Smart grading of diabetic retinopathy: an intelligent recommendation-based fine-tuned EfficientNetB0 framework","authors":"Vatsala Anand, Deepika Koundal, Wael Y. Alghamdi, Bayan M. Alsharbi","doi":"10.3389/frai.2024.1396160","DOIUrl":"https://doi.org/10.3389/frai.2024.1396160","url":null,"abstract":"Diabetic retinopathy is a condition that affects the retina and causes vision loss due to blood vessel destruction. The retina is the layer of the eye responsible for visual processing and nerve signaling. Diabetic retinopathy causes vision loss, floaters, and sometimes blindness; however, it often shows no warning signals in the early stages. Deep learning-based techniques have emerged as viable options for automated illness classification as large-scale medical imaging datasets have become more widely available. To adapt to medical image analysis tasks, transfer learning makes use of pre-trained models to extract high-level characteristics from natural images. In this research, an intelligent recommendation-based fine-tuned EfficientNetB0 model has been proposed for quick and precise assessment for the diagnosis of diabetic retinopathy from fundus images, which will help ophthalmologists in early diagnosis and detection. The proposed EfficientNetB0 model is compared with three transfer learning-based models, namely, ResNet152, VGG16, and DenseNet169. The experimental work is carried out using publicly available datasets from Kaggle consisting of 3,200 fundus images. Out of all the transfer learning models, the EfficientNetB0 model has outperformed with an accuracy of 0.91, followed by DenseNet169 with an accuracy of 0.90. In comparison to other approaches, the proposed intelligent recommendation-based fine-tuned EfficientNetB0 approach delivers state-of-the-art performance on the accuracy, recall, precision, and F1-score criteria. The system aims to assist ophthalmologists in early detection, potentially alleviating the burden on healthcare units.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"23 38","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140696505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-16DOI: 10.3389/frai.2024.1339193
Alexander Broll, Martin Rosentritt, Thomas Schlegl, Markus Goldhacker
Due to the high prevalence of dental caries, fixed dental restorations are regularly required to restore compromised teeth or replace missing teeth while retaining function and aesthetic appearance. The fabrication of dental restorations, however, remains challenging due to the complexity of the human masticatory system as well as the unique morphology of each individual dentition. Adaptation and reworking are frequently required during the insertion of fixed dental prostheses (FDPs), which increase cost and treatment time. This article proposes a data-driven approach for the partial reconstruction of occlusal surfaces based on a data set that comprises 92 3D mesh files of full dental crown restorations.A Generative Adversarial Network (GAN) is considered for the given task in view of its ability to represent extensive data sets in an unsupervised manner with a wide variety of applications. Having demonstrated good capabilities in terms of image quality and training stability, StyleGAN-2 has been chosen as the main network for generating the occlusal surfaces. A 2D projection method is proposed in order to generate 2D representations of the provided 3D tooth data set for integration with the StyleGAN architecture. The reconstruction capabilities of the trained network are demonstrated by means of 4 common inlay types using a Bayesian Image Reconstruction method. This involves pre-processing the data in order to extract the necessary information of the tooth preparations required for the used method as well as the modification of the initial reconstruction loss.The reconstruction process yields satisfactory visual and quantitative results for all preparations with a root mean square error (RMSE) ranging from 0.02 mm to 0.18 mm. When compared against a clinical procedure for CAD inlay fabrication, the group of dentists preferred the GAN-based restorations for 3 of the total 4 inlay geometries.This article shows the effectiveness of the StyleGAN architecture with a downstream optimization process for the reconstruction of 4 different inlay geometries. The independence of the reconstruction process and the initial training of the GAN enables the application of the method for arbitrary inlay geometries without time-consuming retraining of the GAN.
{"title":"A data-driven approach for the partial reconstruction of individual human molar teeth using generative deep learning","authors":"Alexander Broll, Martin Rosentritt, Thomas Schlegl, Markus Goldhacker","doi":"10.3389/frai.2024.1339193","DOIUrl":"https://doi.org/10.3389/frai.2024.1339193","url":null,"abstract":"Due to the high prevalence of dental caries, fixed dental restorations are regularly required to restore compromised teeth or replace missing teeth while retaining function and aesthetic appearance. The fabrication of dental restorations, however, remains challenging due to the complexity of the human masticatory system as well as the unique morphology of each individual dentition. Adaptation and reworking are frequently required during the insertion of fixed dental prostheses (FDPs), which increase cost and treatment time. This article proposes a data-driven approach for the partial reconstruction of occlusal surfaces based on a data set that comprises 92 3D mesh files of full dental crown restorations.A Generative Adversarial Network (GAN) is considered for the given task in view of its ability to represent extensive data sets in an unsupervised manner with a wide variety of applications. Having demonstrated good capabilities in terms of image quality and training stability, StyleGAN-2 has been chosen as the main network for generating the occlusal surfaces. A 2D projection method is proposed in order to generate 2D representations of the provided 3D tooth data set for integration with the StyleGAN architecture. The reconstruction capabilities of the trained network are demonstrated by means of 4 common inlay types using a Bayesian Image Reconstruction method. This involves pre-processing the data in order to extract the necessary information of the tooth preparations required for the used method as well as the modification of the initial reconstruction loss.The reconstruction process yields satisfactory visual and quantitative results for all preparations with a root mean square error (RMSE) ranging from 0.02 mm to 0.18 mm. When compared against a clinical procedure for CAD inlay fabrication, the group of dentists preferred the GAN-based restorations for 3 of the total 4 inlay geometries.This article shows the effectiveness of the StyleGAN architecture with a downstream optimization process for the reconstruction of 4 different inlay geometries. The independence of the reconstruction process and the initial training of the GAN enables the application of the method for arbitrary inlay geometries without time-consuming retraining of the GAN.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"342 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140698211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-16DOI: 10.3389/frai.2024.1374148
Dayan Liu, Tao Song, Kang Na, Shudong Wang
Alzheimer's disease (AD) is a gradually advancing neurodegenerative disorder characterized by a concealed onset. Acetylcholinesterase (AChE) is an efficient hydrolase that catalyzes the hydrolysis of acetylcholine (ACh), which regulates the concentration of ACh at synapses and then terminates ACh-mediated neurotransmission. There are inhibitors to inhibit the activity of AChE currently, but its side effects are inevitable. In various application fields where Al have gained prominence, neural network-based models for molecular design have recently emerged and demonstrate encouraging outcomes. However, in the conditional molecular generation task, most of the current generation models need additional optimization algorithms to generate molecules with intended properties which make molecular generation inefficient. Consequently, we introduce a cognitive-conditional molecular design model, termed PED, which leverages the variational auto-encoder. Its primary function is to adeptly produce a molecular library tailored for specific properties. From this library, we can then identify molecules that inhibit AChE activity without adverse effects. These molecules serve as lead compounds, hastening AD treatment and concurrently enhancing the AI's cognitive abilities. In this study, we aim to fine-tune a VAE model pre-trained on the ZINC database using active compounds of AChE collected from Binding DB. Different from other molecular generation models, the PED can simultaneously perform both property prediction and molecule generation, consequently, it can generate molecules with intended properties without additional optimization process. Experiments of evaluation show that proposed model performs better than other methods benchmarked on the same data sets. The results indicated that the model learns a good representation of potential chemical space, it can well generate molecules with intended properties. Extensive experiments on benchmark datasets confirmed PED's efficiency and efficacy. Furthermore, we also verified the binding ability of molecules to AChE through molecular docking. The results showed that our molecular generation system for AD shows excellent cognitive capacities, the molecules within the molecular library could bind well to AChE and inhibit its activity, thus preventing the hydrolysis of ACh.
{"title":"PED: a novel predictor-encoder-decoder model for Alzheimer drug molecular generation","authors":"Dayan Liu, Tao Song, Kang Na, Shudong Wang","doi":"10.3389/frai.2024.1374148","DOIUrl":"https://doi.org/10.3389/frai.2024.1374148","url":null,"abstract":"Alzheimer's disease (AD) is a gradually advancing neurodegenerative disorder characterized by a concealed onset. Acetylcholinesterase (AChE) is an efficient hydrolase that catalyzes the hydrolysis of acetylcholine (ACh), which regulates the concentration of ACh at synapses and then terminates ACh-mediated neurotransmission. There are inhibitors to inhibit the activity of AChE currently, but its side effects are inevitable. In various application fields where Al have gained prominence, neural network-based models for molecular design have recently emerged and demonstrate encouraging outcomes. However, in the conditional molecular generation task, most of the current generation models need additional optimization algorithms to generate molecules with intended properties which make molecular generation inefficient. Consequently, we introduce a cognitive-conditional molecular design model, termed PED, which leverages the variational auto-encoder. Its primary function is to adeptly produce a molecular library tailored for specific properties. From this library, we can then identify molecules that inhibit AChE activity without adverse effects. These molecules serve as lead compounds, hastening AD treatment and concurrently enhancing the AI's cognitive abilities. In this study, we aim to fine-tune a VAE model pre-trained on the ZINC database using active compounds of AChE collected from Binding DB. Different from other molecular generation models, the PED can simultaneously perform both property prediction and molecule generation, consequently, it can generate molecules with intended properties without additional optimization process. Experiments of evaluation show that proposed model performs better than other methods benchmarked on the same data sets. The results indicated that the model learns a good representation of potential chemical space, it can well generate molecules with intended properties. Extensive experiments on benchmark datasets confirmed PED's efficiency and efficacy. Furthermore, we also verified the binding ability of molecules to AChE through molecular docking. The results showed that our molecular generation system for AD shows excellent cognitive capacities, the molecules within the molecular library could bind well to AChE and inhibit its activity, thus preventing the hydrolysis of ACh.","PeriodicalId":508738,"journal":{"name":"Frontiers in Artificial Intelligence","volume":"7 5‐6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140698798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：