Author name cluster

Qing Cai

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

1 author row

AAAI Conference 2026 Conference Paper

SEMC: Structure-Enhanced Mixture-of-Experts Contrastive Learning for Ultrasound Standard Plane Recognition

Qing Cai
GuihaoYan
Fan Zhang
Cheng Zhang
Zhi Liu

Ultrasound standard plane recognition is essential for clinical tasks such as disease screening, organ evaluation, and biometric measurement. However, existing methods fail to effectively exploit shallow structural information and struggle to capture fine-grained semantic differences through contrastive samples generated by image augmentations, leading to poor recognition of structural and discriminative details in ultrasound standard planes. To address these issues, we propose Structure-Enhanced Mixture-of-Experts Contrastive Learning (SEMC), a novel framework that combines structure-aware feature fusion with expert-guided contrastive learning. Specifically, we propose a Semantic-Structure Fusion Module (SSFM) to exploit multi-scale structural information and enhance the model's ability to perceive fine-grained structural details by effectively aligning shallow and deep features. Meanwhile, a Mixture-of-Experts Contrastive Recognition Module (MCRM) is designed to perform hierarchical contrastive learning and classification across multi-level features using a mixture-of-experts (MoE) mechanism, further improving class separability and overall recognition performance. More importantly, we also curate a large-scale and meticulously annotated liver ultrasound dataset containing six standard planes. Extensive experimental results on our in-house dataset and two public datasets demonstrate that SEMC outperforms recent state-of-the-art methods across various metrics.

PDF Details DOI

AAAI Conference 2025 Conference Paper

BSAFusion: A Bidirectional Stepwise Feature Alignment Network for Unaligned Medical Image Fusion

Huafeng Li
Dayong Su
Qing Cai
Yafei Zhang

If unaligned multimodal medical images can be simultaneously aligned and fused using a single-stage approach within a unified processing framework, it will not only achieve mutual promotion of dual tasks but also help reduce the complexity of the model. However, the design of this model faces the challenge of incompatible requirements for feature fusion and alignment. To address this challenge, this paper proposes an unaligned medical image fusion method called Bidirectional Stepwise Feature Alignment and Fusion (BSFA-F) strategy. To reduce the negative impact of modality differences on cross-modal feature matching, we incorporate the Modal Discrepancy-Free Feature Representation (MDF-FR) method into BSFA-F. MDF-FR utilizes a Modality Feature Representation Head (MFRH) to integrate the global information of the input image. By injecting the information contained in MFRH of the current image into other modality images, it effectively reduces the impact of modality differences on feature alignment while preserving the complementary information carried by different images. In terms of feature alignment, BSFA-F employs a bidirectional stepwise alignment deformation field prediction strategy based on the path independence of vector displacement between two points. This strategy solves the problem of large spans and inaccurate deformation field prediction in single-step alignment. Finally, Multi-Modal Feature Fusion block achieves the fusion of aligned features. The experimental results across multiple datasets demonstrate the effectiveness of our method.

PDF Details DOI

AAAI Conference 2025 Conference Paper

C2PD: Continuity-Constrained Pixelwise Deformation for Guided Depth Super-Resolution

Jiahui Kang
Qing Cai
Runqing Tan
Yimei Liu
Zhi Liu

Guided depth super-resolution (GDSR) has demonstrated impressive performance across a wide range of domains, with numerous methods being proposed. However, existing methods often treat depth maps as images, where shading values are computed discretely, making them struggle to effectively restore the continuity inherent in the depth map. In this paper, we propose a novel approach that maximizes the utilization of spatial characteristics in depth, coupled with human abstract perception of real-world substance, by transforming the GDSR issue into deformation of a roughcast with ideal plasticity, which can be deformed by force like a continuous object. Specifically, we firstly designed a cross-modal operation, Continuity-constrained Asymmetrical Pixelwise Operation (CAPO), which can mimic the process of deforming an isovolumetrically flexible object through external forces. Utilizing CAPO as the fundamental component, we develop the Pixelwise Cross Gradient Deformation (PCGD), which is capable of emulating operations on ideal plastic objects (without volume constraint). Notably, our approach demonstrates state-of-the-art performance across four widely adopted benchmarks for GDSR, with significant advantages in large-scale tasks and generalizability.

PDF Details DOI

YNIMG Journal 2025 Journal Article

Fiber length distribution characterizes the brain network maturation during early school-age

Yanlin Yu
Qing Cai
Longnian Lin
Chu-Chung Huang

Environmental and social changes during early school age have a profound impact on brain development. However, it remains unclear how the brains of typically-developing children adjust white matter to optimize network topology during this period. This study proposes fiber length distribution as a novel nodal metric to capture the continuous maturation of brain network. We acquired dMRI data from N = 30 typically developing children in their first year of primary school and a one-year follow-up. We assessed the longitudinal changes in fiber length distribution, characterized by the median length of connected fibers for each brain region. The length median was positively correlated with degree and betweenness centrality, while negatively correlated with clustering coefficient and local efficiency. From ages 7 to 8, we observed significant decreases in length median in the temporal, superior parietal, anterior cingulate, and medial prefrontal cortices, accompanied by a reduction in long-range connections and an increase in short-range connections. Meta-analytic decoding revealed that the widespread decrease in length median occurred in regions responsible for sensory processing, whereas a more localized increase in length median was observed in regions involved in memory and cognitive control. Finally, simulation tests on healthy adults further supported that the decrease in long-range connections and increase in short-range connections contributed to enhanced network segregation and integration, respectively. Our results suggest that the dual process of short- and long-range fiber changes reflects a cost-efficient strategy for optimizing network organization during this critical developmental stage.

Details DOI

JBHI Journal 2025 Journal Article

Geometric Deep Learning for Protein-Ligand Affinity Prediction with Hybrid Message Passing Strategies

Jiaren Li
Huasen Jiang
Wenjian Ma
Xiangpeng Bi
Rui Chen
Weigang Lu
Qing Cai
Fei Yang

Accurate prediction of protein-ligand affinity (PLA) is critical for drug discovery. Recent deep learning approaches have adopted data-driven models for PLA prediction by learning intrinsic patterns from one-dimensional (1D) sequential or two-dimensional (2D) graph representations of proteins and ligands. However, these low-dimensional methods overlook the three-dimensional (3D) geometric features, which are hypothesized to be critical in binding interaction. To address the above problem, we present a Geo metric deep learning approach with Hybrid message passing strategies- HybridGeo, for protein-ligand affinity prediction. We adopt dual-view graph learning to model the intra- and inter-molecular atomic interactions and propose to aggregate the spatial information with hybrid strategies. In addition, to fully model the inter-residue dependency upon message aggregation, we adopt a geometric graph transformer on the residue-scale graph of protein pockets. Extensive experiments on the PDBbind dataset show that HybridGeo achieves state-of-the-art performance with a Root Mean Square Error (RMSE) of 1. 172. HybridGeo also achieves the best among all baseline models on three external test sets, showcasing good generalizability and robustness. Through systematic ablation experiments, we validated the effectiveness of the proposed modules, and further demonstrated the superior performance of HybridGeo in predicting the binding affinity of macrocyclic compound complexes through case studies. Visualization analysis further indicates the biological interpretability of the model predictions. Our code is publicly available at https://github.com/anxiangbiye1231/HybridGeo

Details DOI

YNIMG Journal 2025 Journal Article

Mapping the mind’s landscape: Common neural encoding for spatial and morality concepts

Jing Wang
Miao Qian
Qing Cai

concepts such as justice are not directly tied to our sensory or motor experiences, yet they constitute an essential part of our knowledge. A longstanding question is how the brain, shaped by survival pressures, encodes these abstract concepts. This study investigated how vertical spatial representations relate to moral concept encoding in the brain (“good is up; bad is down”). We found that vertical positional processing and moral semantics elicited characteristic activation patterns, which enabled the learned neural distinctions between up and down to be generalized to decode the neural signatures of moral and immoral concepts, and vice versa, suggesting shared neural signatures between the two concept domains. Most of the vertical metaphorical representations of morality were independent of the encoding of pleasant vs. unpleasant affect, indicating the specificity of the vertical spatial representation that could not be attributed to the generic representation of arbitrary magnitude or polarity. Nonetheless, morality encoding did not rely solely on vertical spatial information, in that the morality of a word could also be decoded from neural signatures in non-spatial areas. These findings highlight both the spatial metaphorical associations and the domain-specific information in the neural representation of moral concepts.

Details DOI

AAAI Conference 2025 Conference Paper

SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation

Ke Yan
Qing Cai
Fan Zhang
Ziyan Cao
Zhi Liu

Although semi-supervised learning has made significant advances in the field of medical image segmentation, fully annotating a volumetric sample slice by slice remains a costly and time-consuming task. Even worse, most of the existing approaches pay much attention to image-level information and ignore semantic features, resulting in the inability to perceive weak boundaries. To address these issues, we propose a novel Semantic-Guided Triplet Co-training (SGTC) framework, which achieves high-end medical image segmentation by only annotating three orthogonal slices of a few volumetric samples, significantly alleviating the burden of radiologists. Our method consist of two main components. Specifically, to enable semantic-aware, fine-granular segmentation and enhance the quality of pseudo-labels, a novel semantic-guided auxiliary learning mechanism is proposed based on the pretrained CLIP. In addition, focusing on a more challenging but clinically realistic scenario, a new triple-view disparity training strategy is proposed, which uses sparse annotations (i.e., only three labeled slices of a few volumes) to perform co-training between three sub-networks, significantly improving the robustness. Extensive experiments on three public medical datasets demonstrate that our method outperforms most state-of-the-art semi-supervised counterparts under sparse annotation settings.

PDF Details DOI

JBHI Journal 2025 Journal Article

Transformer-Based Weakly Supervised Learning for Whole Slide Lung Cancer Image Classification

Jianpeng An
Yong Wang
Qing Cai
Gang Zhao
Stephan Dooper
Geert Litjens
Zhongke Gao

Image analysis can play an important role in supporting histopathological diagnoses of lung cancer, with deep learning methods already achieving remarkable results. However, due to the large scale of whole-slide images (WSIs), creating manual pixel-wise annotations from expert pathologists is expensive and time-consuming. In addition, the heterogeneity of tumors and similarities in the morphological phenotype of tumor subtypes have caused inter-observer variability in annotations, which limits optimal performance. Effective use of weak labels could potentially alleviate these issues. In this paper, we propose a two-stage transformer-based weakly supervised learning framework called Simple Shuffle-Remix Vision Transformer (SSRViT). Firstly, we introduce a Shuffle-Remix Vision Transformer (SRViT) to retrieve discriminative local tokens and extract effective representative features. Then, the token features are selected and aggregated to generate sparse representations of WSIs, which are fed into a simple transformer-based classifier (SViT) for slide-level prediction. Experimental results demonstrate that the performance of our proposed SSRViT is significantly improved compared with other state-of-the-art methods in discriminating between adenocarcinoma, pulmonary sclerosing pneumocytoma and normal lung tissue (accuracy of 96. 9 ${\%}$ and AUC of 99. 6 ${\%}$ ).

Details DOI

IJCAI Conference 2024 Conference Paper

Exploring Cross-Domain Few-Shot Classification via Frequency-Aware Prompting

Tiange Zhang
Qing Cai
Feng Gao
Lin Qi
Junyu Dong

Cross-Domain Few-Shot Learning has witnessed great stride with the development of meta-learning. However, most existing methods pay more attention to learning domain-adaptive inductive bias (meta-knowledge) through feature-wise manipulation or task diversity improvement while neglecting the phenomenon that deep networks tend to rely more on high-frequency cues to make the classification decision, which thus degenerates the robustness of learned inductive bias since high-frequency information is vulnerable and easy to be disturbed by noisy information. Hence in this paper, we make one of the first attempts to propose a Frequency-Aware Prompting method with mutual attention for Cross-Domain Few-Shot classification, which can let networks simulate the human visual perception of selecting different frequency cues when facing new recognition tasks. Specifically, a frequency-aware prompting mechanism is first proposed, in which high-frequency components of the decomposed source image are switched either with normal distribution sampling or zeroing to get frequency-aware augment samples. Then, a mutual attention module is designed to learn generalizable inductive bias under CD-FSL settings. More importantly, the proposed method is a plug-and-play module that can be directly applied to most off-the-shelf CD-FLS methods. Experimental results on CD-FSL benchmarks demonstrate the effectiveness of our proposed method as well as robustly improve the performance of existing CD-FLS methods. Resources at https: //github. com/tinkez/FAP_CDFSC.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Spherical Pseudo-Cylindrical Representation for Omnidirectional Image Super-resolution

Qing Cai
Mu Li
Dongwei Ren
Jun Lyu
Haiyong Zheng
Junyu Dong
Yee-Hong Yang

Omnidirectional images have attracted significant attention in recent years due to the rapid development of virtual reality technologies. Equirectangular projection (ERP), a naive form to store and transfer omnidirectional images, however, is challenging for existing two-dimensional (2D) image super-resolution (SR) methods due to its inhomogeneous distributed sampling density and distortion across latitude. In this paper, we make one of the first attempts to design a spherical pseudo-cylindrical representation, which not only allows pixels at different latitudes to adaptively adopt the best distinct sampling density but also is model-agnostic to most off-the-shelf SR methods, enhancing their performances. Specifically, we start by upsampling each latitude of the input ERP image and design a computationally tractable optimization algorithm to adaptively obtain a (sub)-optimal sampling density for each latitude of the ERP image. Addressing the distortion of ERP, we introduce a new viewport-based training loss based on the original 3D sphere format of the omnidirectional image, which inherently lacks distortion. Finally, we present a simple yet effective recursive progressive omnidirectional SR network to showcase the feasibility of our idea. The experimental results on public datasets demonstrate the effectiveness of the proposed method as well as the consistently superior performance of our method over most state-of-the-art methods both quantitatively and qualitatively.

PDF Details DOI

JBHI Journal 2021 Journal Article

COVID-19 Screening in Chest X-Ray Images Using Lung Region Priors

Jianpeng An
Qing Cai
Zhiyong Qu
Zhongke Gao

Early screening of COVID-19 is essential for pandemic control, and thus to relieve stress on the health care system. Lung segmentation from chest X-ray (CXR) is a promising method for early diagnoses of pulmonary diseases. Recently, deep learning has achieved great success in supervised lung segmentation. However, how to effectively utilize the lung region in screening COVID-19 still remains a challenge due to domain shift and lack of manual pixel-level annotations. We hereby propose a multi-appearance COVID-19 screening framework by using lung region priors derived from CXR images. Firstly, we propose a multi-scale adversarial domain adaptation network (MS-AdaNet) to boost the cross-domain lung segmentation task as the prior knowledge to the classification network. Then, we construct a multi-appearance network (MA-Net), which is composed of three sub-networks to realize multi-appearance feature extraction and fusion using lung region priors. At last, we can obtain prediction results from normal, viral pneumonia, and COVID-19 using the proposed MA-Net. We extend the proposed MS-AdaNet for lung segmentation task on three different public CXR datasets. The results suggest that the MS-AdaNet outperforms contrastive methods in cross-domain lung segmentation. Moreover, experiments reveal that the proposed MA-Net achieves accuracy of 98. 83 $\%$ and F1-score of 98. 71 $\%$ on COVID-19 screening. The results indicate that the proposed MA-Net can obtain significant performance on COVID-19 screening.

Details DOI

YNICL Journal 2017 Journal Article

Progressive medial temporal lobe atrophy during preclinical Alzheimer's disease

Corinne Pettigrew
Anja Soldan
Kelly Sloane
Qing Cai
Jiangxia Wang
Mei-Cheng Wang
Abhay Moghekar
Michael I. Miller

This study examined whether longitudinal MRI trajectories in medial temporal lobe (MTL) brain regions differed among groups of cognitively normal individuals defined by their cerebrospinal fluid (CSF) levels when they were first enrolled (N =207; mean clinical follow-up=13. 3years (max=20years), mean MRI follow-up=2. 4years (max=8years)). We first compared atrophy rates among groups defined by CSF amyloid and phosphorylated-tau (p-tau) vs. CSF amyloid and total tau (t-tau). We also examined whether, in the presence of amyloid or tau/p-tau, the atrophy rates differed based on whether the subjects ultimately progressed to a diagnosis of mild cognitive impairment (MCI), as well as whether apolipoprotein ε4 (Apoε4) status had an impact on the longitudinal MRI trajectories. The primary finding was that when the groups were defined using CSF amyloid and p-tau, individuals with low levels of CSF amyloid and high levels of CSF p-tau (referred to as Stage 2) showed a significantly greater rate of atrophy in a composite measure of MTL volumes compared to groups defined by evidence of abnormal CSF levels in only one of the brain proteins (but not both), or no evidence of CSF abnormality. In contrast, there were no differences in rate of MTL atrophy when the groups were defined by levels of CSF amyloid and t-tau (instead of p-tau). Additionally, the rate of MTL atrophy did not differ between subjects who progressed to MCI at follow-up vs. those who remained cognitively normal when CSF levels of amyloid, t-tau, or p-tau were covaried. Lastly, the presence of an APOE ε4 genotype did not modulate the degree of MTL atrophy once baseline levels of CSF amyloid, p-tau or t-tau were accounted for. These results suggest that abnormal levels of CSF amyloid and CSF p-tau (but not t-tau) maximize the likelihood of observing significant MTL atrophy over time among individuals with normal cognition at baseline, and emphasize the importance of differentiating biomarkers that primarily reflect neurofibrillary tangle pathology (CSF p-tau) compared with biomarkers of neuronal injury (CSF t-tau).

Details DOI