Xindong Cui
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
Research Subjects
Xindong Cui has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Min Zhu
Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences
5 shared publications - Yemao Hou
Key Laboratory of Vertebrate Evolution and Human Origins, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing, China
2 shared publications Huijuan Mai
Yunnan Key Laboratory for Palaeobiology, Institute of Palaeontology, Yunnan University, Kunming 650091, China
1 shared publicationMario Canul-Ku
Centro de Investigacisn en Matematicas (CIMAT), Guanajuato 36023, Mexico
1 shared publication- Mario Canul‐Ku
Virtual University of the State of Guanajuato, Purísima del Rincón Guanajuato Mexico
1 shared publication - Matt Friedman
Museum of Paleontology and Department of Earth and Environmental Sciences University of Michigan Ann Arbor MI USA
1 shared publication
Latest Publications
- 2 August 2023
- Published by Wiley in X-Ray Spectrometry
- Vol. 53 (5) , 405-414
Super‐resolution reconstruction of vertebrate microfossil computed tomography images based on deep learning
Micropaleontologists use the fine structures of microfossils to extract evolutionary information. These structures could not be directly observed with the naked eye. Recently, paleontologists resort to computed tomography (CT) images to mine the information, and pursue higher resolution CT images with in‐depth research. Therefore, we propose a new model, weighted super‐resolution generative adversarial network (WSRGAN), for the super‐resolution reconstruction of CT images. The model proposed herein (WSRGAN) obtained higher LPIPS (0.0757) on the experimental dataset, compared with Bilinear (0.4289), Bicubic (0.4166), EDSR (0.2281), WDSR (0.2640), and SRGAN (0.0815). WSRGAN meets the requirements of paleontologists for reconstructing fish microfossils. We hope that more super‐resolution reconstruction methods based on deep learning could be applied to paleontology and achieve better performance.
- 2 August 2023
- Published by Wiley in X-Ray Spectrometry
- Vol. 53 (5) , 405-414
Super‐resolution reconstruction of vertebrate microfossil computed tomography images based on deep learning
Micropaleontologists use the fine structures of microfossils to extract evolutionary information. These structures could not be directly observed with the naked eye. Recently, paleontologists resort to computed tomography (CT) images to mine the information, and pursue higher resolution CT images with in‐depth research. Therefore, we propose a new model, weighted super‐resolution generative adversarial network (WSRGAN), for the super‐resolution reconstruction of CT images. The model proposed herein (WSRGAN) obtained higher LPIPS (0.0757) on the experimental dataset, compared with Bilinear (0.4289), Bicubic (0.4166), EDSR (0.2281), WDSR (0.2640), and SRGAN (0.0815). WSRGAN meets the requirements of paleontologists for reconstructing fish microfossils. We hope that more super‐resolution reconstruction methods based on deep learning could be applied to paleontology and achieve better performance.
- Open Access
- 2 May 2022
- Published by Springer Nature in Nature Communications
- Vol. 13 (1) , 1-9
The rapid evolution of lungfish durophagy
Innovations relating to the consumption of hard prey are implicated in ecological shifts in marine ecosystems as early as the mid-Paleozoic. Lungfishes represent the first and longest-ranging lineage of durophagous vertebrates, but how and when the various feeding specializations of this group arose remain unclear. Two exceptionally preserved fossils of the Early Devonian lobe-finned fishYoungolepisreveal the origin of the specialized lungfish feeding mechanism.Youngolepishas a radically restructured palate, reorienting jaw muscles for optimal force transition, coupled with radiating entopterygoid tooth rows like those of lungfish toothplates. This triturating surface occurs in conjunction with marginal dentition and blunt coronoid fangs, suggesting a role in crushing rather than piercing prey. Bayesian tip-dating analyses incorporating these morphological data indicate that the complete suite of lungfish feeding specializations may have arisen in as little as 7 million years, representing one of the most striking episodes of innovation during the initial evolutionary radiations of bony fishes.
- 2 May 2022
- Published by Springer Nature in Nature Communications
- Vol. 13 (1) , 1-9
Open AccessThe rapid evolution of lungfish durophagy
Innovations relating to the consumption of hard prey are implicated in ecological shifts in marine ecosystems as early as the mid-Paleozoic. Lungfishes represent the first and longest-ranging lineage of durophagous vertebrates, but how and when the various feeding specializations of this group arose remain unclear. Two exceptionally preserved fossils of the Early Devonian lobe-finned fishYoungolepisreveal the origin of the specialized lungfish feeding mechanism.Youngolepishas a radically restructured palate, reorienting jaw muscles for optimal force transition, coupled with radiating entopterygoid tooth rows like those of lungfish toothplates. This triturating surface occurs in conjunction with marginal dentition and blunt coronoid fangs, suggesting a role in crushing rather than piercing prey. Bayesian tip-dating analyses incorporating these morphological data indicate that the complete suite of lungfish feeding specializations may have arisen in as little as 7 million years, representing one of the most striking episodes of innovation during the initial evolutionary radiations of bony fishes.
- 17 September 2021
- Published by Taylor & Francis in Journal of Systematic Palaeontology
- Vol. 19 (18) , 1303-1315
A new dipnoan genus from the Middle Devonian of Huize, Yunnan, China
A new genus and species of tooth-plated lungfishes, Dianodipterus huizeensis gen. et sp. nov., is described from the Qujing Formation (late Eifelian, Middle Devonian) of Huize County, Qujing, north-eastern Yunnan, China. It differs from other lungfishes in the small proportion of the tooth-bearing part on the pterygoid (less than one-quarter), the extensively developed anteromedial edentulous portion of the pterygoid, and tooth rows few in number (only four). Phylogenetic analysis reveals that Dianodipterus is more plesiomorphic than Dipterus, yet crownward of Melanognathus. As the ninth Devonian lungfish genus from China, the new taxon displays a novel schema of lungfish dentition and increases the diversity of early lungfishes. http://zoobank.org/urn:lsid:zoobank.org:pub:207B0E6E-EBBF-4E92-83F4-78FDAD6EBB60
- 17 September 2021
- Published by Taylor & Francis in Journal of Systematic Palaeontology
- Vol. 19 (18) , 1303-1315
A new dipnoan genus from the Middle Devonian of Huize, Yunnan, China
A new genus and species of tooth-plated lungfishes, Dianodipterus huizeensis gen. et sp. nov., is described from the Qujing Formation (late Eifelian, Middle Devonian) of Huize County, Qujing, north-eastern Yunnan, China. It differs from other lungfishes in the small proportion of the tooth-bearing part on the pterygoid (less than one-quarter), the extensively developed anteromedial edentulous portion of the pterygoid, and tooth rows few in number (only four). Phylogenetic analysis reveals that Dianodipterus is more plesiomorphic than Dipterus, yet crownward of Melanognathus. As the ninth Devonian lungfish genus from China, the new taxon displays a novel schema of lungfish dentition and increases the diversity of early lungfishes. http://zoobank.org/urn:lsid:zoobank.org:pub:207B0E6E-EBBF-4E92-83F4-78FDAD6EBB60
- 4 March 2021
- Published by Taylor & Francis in Journal of Vertebrate Paleontology
- Vol. 41 (2)
Modeling scale morphogenesis in a Devonian chondrichthyan and scale growth patterns in crown gnathostomes
Early chondrichthyans (cartilaginous fishes) possess a rich record of scale fossils that have impacted on taxonomic and phylogenetic studies of early jawed vertebrates, but the significance of these scales remains elusive due to the lack of well-defined characters. Gualepis, a scale-based genus from the Lower Devonian of Yunnan, China, was initially recognized as an early chondrichthyan of uncertain affinity. We investigate new material of Gualepis elegans from the type locality. By combining microtomographic data with morphological and histological examination of specimens at successive growth stages, we have reconstructed the 3D growth pattern of the scales. During growth, Gualepis scales added new odontodes appositionally, which has previously been documented in the dermal skeletons of Seretolepis, Kathemacanthus, Parexus, and other crownward members of the stem chondrichthyans. Phylogenetic analysis resolves Gualepis as a non-acanthodid stem chondrichthyan near the crown chondrichthyan node. Our 3D approach is potentially useful to render more reliable scale-based characters, especially regarding scale growth patterns. This study provides the first 3D virtual growth model of stem chondrichthyan scales with a Seretolepis-type morphogenesis, offering new insights into the early evolution of chondrichthyan dermal skeletons.
- 4 March 2021
- Published by Taylor & Francis in Journal of Vertebrate Paleontology
- Vol. 41 (2)
Modeling scale morphogenesis in a Devonian chondrichthyan and scale growth patterns in crown gnathostomes
Early chondrichthyans (cartilaginous fishes) possess a rich record of scale fossils that have impacted on taxonomic and phylogenetic studies of early jawed vertebrates, but the significance of these scales remains elusive due to the lack of well-defined characters. Gualepis, a scale-based genus from the Lower Devonian of Yunnan, China, was initially recognized as an early chondrichthyan of uncertain affinity. We investigate new material of Gualepis elegans from the type locality. By combining microtomographic data with morphological and histological examination of specimens at successive growth stages, we have reconstructed the 3D growth pattern of the scales. During growth, Gualepis scales added new odontodes appositionally, which has previously been documented in the dermal skeletons of Seretolepis, Kathemacanthus, Parexus, and other crownward members of the stem chondrichthyans. Phylogenetic analysis resolves Gualepis as a non-acanthodid stem chondrichthyan near the crown chondrichthyan node. Our 3D approach is potentially useful to render more reliable scale-based characters, especially regarding scale growth patterns. This study provides the first 3D virtual growth model of stem chondrichthyan scales with a Seretolepis-type morphogenesis, offering new insights into the early evolution of chondrichthyan dermal skeletons.
- Open Access
- 12 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 148744-148756
ADMorph: A 3D Digital Microfossil Morphology Dataset for Deep Learning
Microfossils, tiny fossils whose study requires the use of a microscope, have been widely applied in many fields of earth, life, and environmental sciences. The abundance and high diversity of microfossils, as well as the need for rapid identification, call for automated methods to classify microfossils. In this study, we constructed an open dataset of three-dimensional (3D) microfossils and proposed a deep learning-based approach for microfossil classification. The dataset, named ‘Archives of Digital Morphology’ (ADMorph), currently contains more than ten thousand 3D models from five classes of 410 million-year-old fishes. The deep learning-based method includes data preprocessing, feature extraction, and 3D microfossil model classification. To assess the method performance and dataset representability, we performed extensive experiments. Compared with multiview convolutional neural networks (MVCNN) (91.54%), PointNet (64.13%), and VoxNet (78.15%), the method proposed herein had higher accuracy (97.60%) on the experimental dataset. We also verified data preprocessing (92.36%) and feature extraction (97.10%). We combined them to obtain the macroaveraging accuracy of 97.60%, the highest accuracy of 100%, and the lowest accuracy of 88.78%. We suggest that the proposed method can be applied to other 3D fossils and biomorphological research fields. The fast-accumulating 3D fossil models might become a source of information-rich datasets for deep learning.
- 12 August 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 148744-148756
Open AccessADMorph: A 3D Digital Microfossil Morphology Dataset for Deep Learning
Microfossils, tiny fossils whose study requires the use of a microscope, have been widely applied in many fields of earth, life, and environmental sciences. The abundance and high diversity of microfossils, as well as the need for rapid identification, call for automated methods to classify microfossils. In this study, we constructed an open dataset of three-dimensional (3D) microfossils and proposed a deep learning-based approach for microfossil classification. The dataset, named ‘Archives of Digital Morphology’ (ADMorph), currently contains more than ten thousand 3D models from five classes of 410 million-year-old fishes. The deep learning-based method includes data preprocessing, feature extraction, and 3D microfossil model classification. To assess the method performance and dataset representability, we performed extensive experiments. Compared with multiview convolutional neural networks (MVCNN) (91.54%), PointNet (64.13%), and VoxNet (78.15%), the method proposed herein had higher accuracy (97.60%) on the experimental dataset. We also verified data preprocessing (92.36%) and feature extraction (97.10%). We combined them to obtain the macroaveraging accuracy of 97.60%, the highest accuracy of 100%, and the lowest accuracy of 88.78%. We suggest that the proposed method can be applied to other 3D fossils and biomorphological research fields. The fast-accumulating 3D fossil models might become a source of information-rich datasets for deep learning.