Xi Jiang
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
Xi Jiang has a total of 273 co-authors. Here are the top co-authors by number of shared publications:
- Tianming Liu
University of Georgia
16 shared publications - Tuo Zhang
School of Automation Northwestern Polytechnical University Xi'an China
16 shared publications - Keith Maurice Kendrick
University of Electronic Science and Technology of China
12 shared publications Jiadong Yan
11 shared publicationsTianming Liu
11 shared publicationsTuo Zhang
School of Automation, Northwestern Polytechnical University, Xi'an, China
11 shared publications
Latest Publications
- Open Access
- 29 September 2026
- Published by Elsevier in Developmental Cognitive Neuroscience
- Vol. 82
Frequency-specific gyro-sulcal functional differentiation in preterm neonates: Sulcal dominance and gene expression correlates
- 29 September 2026
- Published by Elsevier in Developmental Cognitive Neuroscience
- Vol. 82
Open AccessFrequency-specific gyro-sulcal functional differentiation in preterm neonates: Sulcal dominance and gene expression correlates
- 19 September 2026
- Published by Elsevier in Neural Networks
- Vol. 205
FL-SEHGT: Federated Learning-Based Heterogeneous Graph Transformer with Squeeze-and-Excitation for ASD Identification
- 19 September 2026
- Published by Elsevier in Neural Networks
- Vol. 205
FL-SEHGT: Federated Learning-Based Heterogeneous Graph Transformer with Squeeze-and-Excitation for ASD Identification
- Open Access
- 15 April 2026
- Published by Wiley in Human Brain Mapping
- Vol. 47 (6) , e70524
Structuro‐Functional Differentiation and Coupling of Gyri and Sulci in the Neonatal Cortex
Brain structure and function undergo rapid development during the neonatal period. Investigating neonatal brain structure, function, and their relationship is therefore crucial for understanding how the brain matures into a complex structuro‐functional system. As fundamental anatomical units of the cerebral cortex, gyri and sulci provide novel and valuable insights for such investigations. However, gyro‐sulcal differentiation and their structuro‐functional developmental relationship in neonates remain poorly explored. To address this gap, we used multi‐modal MRI data (structural T2w, diffusion‐weighted, and resting state functional MRI) from 438 neonatal brains in the public dHCP dataset. We systematically examined differences in functional connectivity (FC) and structural connectivity (SC) between gyri and sulci from 38 to 44 weeks postmenstrual age, as well as their FC‐SC coupling characteristics. From 38 to 44 weeks, both FC and SC were consistently strongest between gyro‐gyral regions and weakest between sulco‐sulcal regions, demonstrating that gyri act as global information processing hubs while sulci serve as local functional units in the neonatal brain. FC‐SC coupling exhibited distinct patterns across cortical lobes and over time, with a characteristic shift from coupling to decoupling around 41 weeks in most regions. This study provides a foundation for understanding early developmental mechanisms of brain structure–function relationships and establishes a normative reference of gyro‐sulcal differentiation as well as FC‐SC coupling in the neonatal period. These findings may inform future investigations of atypical neurodevelopment and contribute to the identification of early biomarkers for neurodevelopmental disorders.
- 15 April 2026
- Published by Wiley in Human Brain Mapping
- Vol. 47 (6) , e70524
Open AccessStructuro‐Functional Differentiation and Coupling of Gyri and Sulci in the Neonatal Cortex
Brain structure and function undergo rapid development during the neonatal period. Investigating neonatal brain structure, function, and their relationship is therefore crucial for understanding how the brain matures into a complex structuro‐functional system. As fundamental anatomical units of the cerebral cortex, gyri and sulci provide novel and valuable insights for such investigations. However, gyro‐sulcal differentiation and their structuro‐functional developmental relationship in neonates remain poorly explored. To address this gap, we used multi‐modal MRI data (structural T2w, diffusion‐weighted, and resting state functional MRI) from 438 neonatal brains in the public dHCP dataset. We systematically examined differences in functional connectivity (FC) and structural connectivity (SC) between gyri and sulci from 38 to 44 weeks postmenstrual age, as well as their FC‐SC coupling characteristics. From 38 to 44 weeks, both FC and SC were consistently strongest between gyro‐gyral regions and weakest between sulco‐sulcal regions, demonstrating that gyri act as global information processing hubs while sulci serve as local functional units in the neonatal brain. FC‐SC coupling exhibited distinct patterns across cortical lobes and over time, with a characteristic shift from coupling to decoupling around 41 weeks in most regions. This study provides a foundation for understanding early developmental mechanisms of brain structure–function relationships and establishes a normative reference of gyro‐sulcal differentiation as well as FC‐SC coupling in the neonatal period. These findings may inform future investigations of atypical neurodevelopment and contribute to the identification of early biomarkers for neurodevelopmental disorders.
- Top Cited Papers
- 1 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Engineering
- Vol. 73 (3) , 1050-1061
ChatRadio-Valuer: A Chat Large Language Model for Generalizable Radiology Impression Generation on Multi-Institution and Multi-System Data
Achieving clinical level performance and widespread deployment for generating radiology impressions encounters a giant challenge for conventional artificial intelligence models tailored to specific diseases and organs. Concurrent with the increasing accessibility of radiology reports and advancements in modern general AI techniques, the emergence and potential of deployable radiology AI exploration have been bolstered. Here, we present ChatRadio-Valuer, the first general radiology diagnosis large language model for localized deployment within hospitals and being close to clinical use for multi-institution and multi-system diseases. ChatRadio-Valuer achieved 15 state-of-the-art results across five human systems and six institutions in clinical-level events (n = 332,673) through rigorous and full-spectrum assessment, including engineering metrics, clinical validation, and efficiency evaluation. Notably, it exceeded OpenAI's GPT-3.5 and GPT-4 models, achieving superior performance in comprehensive disease diagnosis compared to the average level of radiology experts. Besides, ChatRadio-Valuer supports zero-shot transfer learning, greatly boosting its effectiveness as a radiology assistant, while ensuring adherence to privacy standards and being readily utilized for large-scale patient populations. Our expeditions suggest the development of localized LLMs would become an imperative avenue in hospital applications.
- 1 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Engineering
- Vol. 73 (3) , 1050-1061
Top Cited PapersChatRadio-Valuer: A Chat Large Language Model for Generalizable Radiology Impression Generation on Multi-Institution and Multi-System Data
Achieving clinical level performance and widespread deployment for generating radiology impressions encounters a giant challenge for conventional artificial intelligence models tailored to specific diseases and organs. Concurrent with the increasing accessibility of radiology reports and advancements in modern general AI techniques, the emergence and potential of deployable radiology AI exploration have been bolstered. Here, we present ChatRadio-Valuer, the first general radiology diagnosis large language model for localized deployment within hospitals and being close to clinical use for multi-institution and multi-system diseases. ChatRadio-Valuer achieved 15 state-of-the-art results across five human systems and six institutions in clinical-level events (n = 332,673) through rigorous and full-spectrum assessment, including engineering metrics, clinical validation, and efficiency evaluation. Notably, it exceeded OpenAI's GPT-3.5 and GPT-4 models, achieving superior performance in comprehensive disease diagnosis compared to the average level of radiology experts. Besides, ChatRadio-Valuer supports zero-shot transfer learning, greatly boosting its effectiveness as a radiology assistant, while ensuring adherence to privacy standards and being readily utilized for large-scale patient populations. Our expeditions suggest the development of localized LLMs would become an imperative avenue in hospital applications.
- Open Access
- 26 November 2025
- Published by Elsevier in Developmental Cognitive Neuroscience
- Vol. 77, 101647
Preterm birth differentially impacts structural and functional connectivity of cortical gyri and sulci
- 26 November 2025
- Published by Elsevier in Developmental Cognitive Neuroscience
- Vol. 77, 101647
Open AccessPreterm birth differentially impacts structural and functional connectivity of cortical gyri and sulci
- Open Access
- 1 September 2025
- Published by KEAI Publishing in Meta-Radiology
- Vol. 3 (3)
Intelligent analysis of chest X-ray based on multi-modal instruction tuning
- 1 September 2025
- Published by KEAI Publishing in Meta-Radiology
- Vol. 3 (3)
Open AccessIntelligent analysis of chest X-ray based on multi-modal instruction tuning
- Open Access
- 20 June 2025
- Published by Springer Nature in Molecular Psychiatry
- Vol. 30 (11) , 5034-5044
Identifying brain functional subtypes and corresponding task performance profiles in autism spectrum disorder
Refining the classification of autism spectrum disorder (ASD) subtypes is essential for advancing personalized interventions strategies, given the substantial heterogeneity in phenotypic clinical symptoms among individuals with ASD. Thus, the current study integrated normative modeling, resting-state fMRI data, clinical assessment, and eye-gaze patterns to investigate potential ASD subtypes. By incorporating both static and instant dynamic (strength and variability) functional connectivity as predictive variables within the normative models, we aimed to delineate multi-level functional developmental trajectories. Our comprehensive analysis of 1046 participants (479 with ASD, 567 typical development) identified two distinct neural ASD subtypes with unique functional brain network profiles despite comparable clinical presentations. One ASD subtype was characterized by positive deviations in the occipital network and cerebellar network, coupled with negative deviations in the frontoparietal network, default mode network, and cingulo-opercular network. Conversely, the other subtype exhibited an inverse pattern of functional deviations across these networks. Furthermore, an independent cohort of 21 ASD individuals revealed that these neural subtypes were also associated with distinct gaze patterns assessed by two autism-sensitive eye-tracking tasks focused on preference for social cues. These findings emphasize the complexity and heterogeneity of ASD, uncovering the presence of neurobehavioral subtypes that extend beyond simple neural variations and manifest in diverse functional developmental patterns and behavioral presentations. This study underscores the importance of adopting personalized intervention strategies that cater to the individual needs of each person, rather than relying on a standardized, one-size-fits-all approach, despite the presence of similar clinical symptoms.
- 20 June 2025
- Published by Springer Nature in Molecular Psychiatry
- Vol. 30 (11) , 5034-5044
Open AccessIdentifying brain functional subtypes and corresponding task performance profiles in autism spectrum disorder
Refining the classification of autism spectrum disorder (ASD) subtypes is essential for advancing personalized interventions strategies, given the substantial heterogeneity in phenotypic clinical symptoms among individuals with ASD. Thus, the current study integrated normative modeling, resting-state fMRI data, clinical assessment, and eye-gaze patterns to investigate potential ASD subtypes. By incorporating both static and instant dynamic (strength and variability) functional connectivity as predictive variables within the normative models, we aimed to delineate multi-level functional developmental trajectories. Our comprehensive analysis of 1046 participants (479 with ASD, 567 typical development) identified two distinct neural ASD subtypes with unique functional brain network profiles despite comparable clinical presentations. One ASD subtype was characterized by positive deviations in the occipital network and cerebellar network, coupled with negative deviations in the frontoparietal network, default mode network, and cingulo-opercular network. Conversely, the other subtype exhibited an inverse pattern of functional deviations across these networks. Furthermore, an independent cohort of 21 ASD individuals revealed that these neural subtypes were also associated with distinct gaze patterns assessed by two autism-sensitive eye-tracking tasks focused on preference for social cues. These findings emphasize the complexity and heterogeneity of ASD, uncovering the presence of neurobehavioral subtypes that extend beyond simple neural variations and manifest in diverse functional developmental patterns and behavioral presentations. This study underscores the importance of adopting personalized intervention strategies that cater to the individual needs of each person, rather than relying on a standardized, one-size-fits-all approach, despite the presence of similar clinical symptoms.
- Top Cited Papers
- 17 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (10) , 17635-17649
ChatABL: Abductive Learning via Natural Language Interaction With ChatGPT
Large language models (LLMs) such as ChatGPT have recently demonstrated significant potential in mathematical abilities, providing a valuable reasoning paradigm consistent with human natural language. However, LLMs currently have difficulty in bridging perception, language understanding, and reasoning (PLR) capabilities due to incompatibility of the underlying information flow among them, making their reasoning ability not fully elicited and challenging to accomplish complicated reasoning tasks autonomously. To resolve the above problem, a novel method called ChatABL is proposed by integrating LLMs into an abductive learning (ABL) framework, capable of unifying the three abilities effectively in a more user-friendly and understandable manner. Initially, the proposed method uses LLMs to correct the incomplete logical facts for optimizing the perception module, by summarizing and reorganizing domain knowledge represented in natural language format. Then, the perception module also provides necessary logical reasoning materials for feeding LLMs. Finally, these parts are integrated into a dynamic closed-loop system by introducing the feedback form and automatic learning strategies to mutually promote their performance. As a testbed, the variable-length handwritten equation decipherment (HED), an abstract expression of the Mayan calendar decoding, is used to demonstrate that ChatABL has reasoning ability beyond most existing state-of-the-art methods, which has been well-supported by comparative studies. To the best of authors’ knowledge, the proposed ChatABL is the first attempt to explore a possible and novel avenue to approaching human-level cognitive ability via natural language interaction by means of ChatGPT.
- 17 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (10) , 17635-17649
Top Cited PapersChatABL: Abductive Learning via Natural Language Interaction With ChatGPT
Large language models (LLMs) such as ChatGPT have recently demonstrated significant potential in mathematical abilities, providing a valuable reasoning paradigm consistent with human natural language. However, LLMs currently have difficulty in bridging perception, language understanding, and reasoning (PLR) capabilities due to incompatibility of the underlying information flow among them, making their reasoning ability not fully elicited and challenging to accomplish complicated reasoning tasks autonomously. To resolve the above problem, a novel method called ChatABL is proposed by integrating LLMs into an abductive learning (ABL) framework, capable of unifying the three abilities effectively in a more user-friendly and understandable manner. Initially, the proposed method uses LLMs to correct the incomplete logical facts for optimizing the perception module, by summarizing and reorganizing domain knowledge represented in natural language format. Then, the perception module also provides necessary logical reasoning materials for feeding LLMs. Finally, these parts are integrated into a dynamic closed-loop system by introducing the feedback form and automatic learning strategies to mutually promote their performance. As a testbed, the variable-length handwritten equation decipherment (HED), an abstract expression of the Mayan calendar decoding, is used to demonstrate that ChatABL has reasoning ability beyond most existing state-of-the-art methods, which has been well-supported by comparative studies. To the best of authors’ knowledge, the proposed ChatABL is the first attempt to explore a possible and novel avenue to approaching human-level cognitive ability via natural language interaction by means of ChatGPT.
- 22 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. PP (2162237X) , 1-15
A Novel Dynamic Neural Network for Heterogeneity-Aware Structural Brain Network Exploration and Alzheimer’s Disease Diagnosis
Heterogeneity is a fundamental characteristic of brain diseases, distinguished by variability not only in brain atrophy but also in the complexity of neural connectivity and brain networks. However, existing data-driven methods fail to provide a comprehensive analysis of brain heterogeneity. Recently, dynamic neural networks (DNNs) have shown significant advantages in capturing sample-wise heterogeneity. Therefore, in this article, we first propose a novel dynamic heterogeneity-aware network (DHANet) to identify critical heterogeneous brain regions, explore heterogeneous connectivity between them, and construct a heterogeneous-aware structural brain network (HGA-SBN) using structural magnetic resonance imaging (sMRI). Specifically, we develop a 3-D dynamic convmixer to extract abundant heterogeneous features from sMRI first. Subsequently, the critical brain atrophy regions are identified by dynamic prototype learning with embedding the hierarchical brain semantic structure. Finally, we employ a joint dynamic edge-correlation (JDE) modeling approach to construct the heterogeneous connectivity between these regions and analyze the HGA-SBN. To evaluate the effectiveness of the DHANet, we conduct elaborate experiments on three public datasets and the method achieves state-of-the-art (SOTA) performance on two classification tasks.
- 22 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. PP (2162237X) , 1-15
A Novel Dynamic Neural Network for Heterogeneity-Aware Structural Brain Network Exploration and Alzheimer’s Disease Diagnosis
Heterogeneity is a fundamental characteristic of brain diseases, distinguished by variability not only in brain atrophy but also in the complexity of neural connectivity and brain networks. However, existing data-driven methods fail to provide a comprehensive analysis of brain heterogeneity. Recently, dynamic neural networks (DNNs) have shown significant advantages in capturing sample-wise heterogeneity. Therefore, in this article, we first propose a novel dynamic heterogeneity-aware network (DHANet) to identify critical heterogeneous brain regions, explore heterogeneous connectivity between them, and construct a heterogeneous-aware structural brain network (HGA-SBN) using structural magnetic resonance imaging (sMRI). Specifically, we develop a 3-D dynamic convmixer to extract abundant heterogeneous features from sMRI first. Subsequently, the critical brain atrophy regions are identified by dynamic prototype learning with embedding the hierarchical brain semantic structure. Finally, we employ a joint dynamic edge-correlation (JDE) modeling approach to construct the heterogeneous connectivity between these regions and analyze the HGA-SBN. To evaluate the effectiveness of the DHANet, we conduct elaborate experiments on three public datasets and the method achieves state-of-the-art (SOTA) performance on two classification tasks.