Recent advances in large models demonstrate significant prospects for transforming the field of medical imaging. These models, including large language models, large visual models, and multimodal large models, offer unprecedented capabilities in processing and interpreting complex medical data across various imaging modalities. By leveraging self-supervised pretraining on vast unlabeled datasets, cross-modal representation learning, and domain-specific medical knowledge adaptation through fine-tuning, large models can achieve higher diagnostic accuracy and more efficient workflows for key clinical tasks. This review summarizes the concepts, methods, and progress of large models in medical imaging, highlighting their potential in precision medicine. The article first outlines the integration of multimodal data under large model technologies, approaches for training large models with medical datasets, and the need for robust evaluation metrics. It then explores how large models can revolutionize applications in critical tasks such as image segmentation, disease diagnosis, personalized treatment strategies, and real-time interactive systems, thus pushing the boundaries of traditional imaging analysis. Despite their potential, the practical implementation of large models in medical imaging faces notable challenges, including the scarcity of high-quality medical data, the need for optimized perception of imaging phenotypes, safety considerations, and seamless integration with existing clinical workflows and equipment. As research progresses, the development of more efficient, interpretable, and generalizable models will be critical to ensuring their reliable deployment across diverse clinical environments. This review aims to provide insights into the current state of the field and provide directions for future research to facilitate the broader adoption of large models in clinical practice.
Humans ; Diagnostic Imaging/methods* ; Precision Medicine/methods* ; Image Processing, Computer-Assisted/methods*

With the emergence of deep learning techniques based on convolutional neural networks, artificial intelligence (AI) has driven transformative developments in the field of medical image analysis. Recently, large language models (LLMs) such as ChatGPT have also started to achieve distinction in this domain. Increasing research shows the undeniable role of AI in reshaping various aspects of medical image analysis, including processes such as image enhancement, segmentation, detection in image preprocessing, and postprocessing related to medical diagnosis and prognosis in clinical settings. However, despite the significant progress in AI research, studies investigating the recent advances in AI technology in the aforementioned aspects, the changes in research hotspot trajectories, and the performance of studies in addressing key clinical challenges in this field are limited. This article provides an overview of recent advances in AI for medical image analysis and discusses the methodological profiles, advantages, disadvantages, and future trends of AI technologies.
Artificial Intelligence ; Humans ; Image Processing, Computer-Assisted/methods* ; Neural Networks, Computer ; Deep Learning ; Diagnostic Imaging/methods*

A complete plant body consists of elements on different scales, including microscopic molecules, mesoscopic multicellular structures, and macroscopic tissues and organs, which are interconnected to form complex biological networks. The growth and development of plants involve the regulation of elements on different scales and their biological networks, which requires the coordinated operation of multiple molecules, cells, tissues, and organs. It is difficult to reveal the essence of multi-level life activities by a single method or technology. In recent years, the development of various novel imaging technologies has provided new approaches for revealing the complex life activities in plants. Using multi-modal imaging technologies to study the cross-scale network connections of plants from the microscopic, mesoscopic, and macroscopic levels is crucial for understanding the complex internal connections behind biological functions. This paper first summarizes multi-modal cross-scale imaging technologies, three-dimensional reconstruction, and image processing methods, outlines the basic framework of cross-scale network connection properties, and then summarizes the applications of multi-modal imaging technologies in elucidating plant multi-scale networks. Finally, this review systematically integrates the combined analysis of cross-scale 3D spatial structural data and single-cell omics, laying a theoretical foundation for the innovation of novel plant imaging technologies. Furthermore, it provides a new research paradigm for in-depth exploration of the interaction mechanisms among cross-scale elements and the principles of biological network connectivity in plant life activities.
Plants/metabolism* ; Imaging, Three-Dimensional/methods* ; Image Processing, Computer-Assisted/methods* ; Multimodal Imaging/methods* ; Plant Physiological Phenomena

Accurately obtaining the crop quantity and density is not only crucial for the demand-based input of water and fertilizer in the field but also vital for ensuring the yield and quality of crops. Aerial photography by unmanned aerial vehicles (UAVs) can quickly acquire the distribution image information of crops over a large area. However, the accurate recognition of a single type of dense targets is a huge challenge for most recognition algorithms. Taking banana seedlings as an example in this study, we captured the images of banana plantations by UAVs from high altitudes to explore an efficient recognition method for dense targets. We proposed a strategy of "cut-recognition-stitch" and constructed a counting method based on the improved Faster R-CNN algorithm. First, the images containing highly dense targets were cropped into a large number of image tiles according to different sizes (simulating different flight altitudes), and the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm was adopted to improve the image quality. A banana seedling dataset containing 36 000 image tiles was constructed. Then, the Faster R-CNN network with optimized parameters was used to train the banana seedling recognition model. Finally, the recognition results were reversely stitched together, and a boundary deduplication algorithm was designed to correct the final counting results to reduce the repeated recognition caused by image cropping. The results show that the recognition accuracy of the Faster R-CNN with optimized parameters for banana image datasets of different sizes can reach up to 0.99 at most. The deduplication algorithm can reduce the average counting error for the original aerial images from 1.60% to 0.60%, and the average counting accuracy of banana seedlings reaches 99.4%. The proposed method effectively addresses the challenge of recognizing dense small objects in high-resolution aerial images, providing an efficient and reliable technical solution for intelligent crop density monitoring in precision agriculture.
Musa/growth & development* ; Crops, Agricultural/growth & development* ; Algorithms ; Neural Networks, Computer ; Unmanned Aerial Devices ; Seedlings/growth & development* ; Image Processing, Computer-Assisted/methods* ; Photography ; Agriculture/methods*

Objective To explore the application value of multi-model adaptive statistical iterative reconstruction-Veo (ASiR-V) in ultra-low dose chest CT examination of children in the plateau area. Methods The children who underwent chest CT examination in Xizang Autonomous Region People's Hospital were enrolled in this study and assigned into two groups according to the scanning conditions.Group A underwent scanning at a tube voltage of 100 kV and ASiR-V 50% reconstruction,and group B underwent scanning at a tube voltage of 80 kV and ASiR-V 0 (Group B1) and ASiR-V 50% (Group B2) reconstruction.The image quality of each group was evaluated objectively and subjectively.The radiation dose and image quality were compared between groups. Results Groups A and B showed the volume CT dose indexes of (2.33±0.62) mGy and (0.86±0.01) mGy and the dose length products of (65.01±25.12) mGy·cm and (23.55±3.38) mGy·cm,respectively,which presented differences between groups (both P<0.001).The image noise in the bilateral upper and middle lung areas in group B2 was lower than that in group B1 but higher than that in group A (all P<0.001).There was no significant difference in image quality score of the lung window among groups (all P>0.05).Groups A,B1,and B2 had no significant differences in ascending aorta (P=0.538) or liver CT value (P=0.175) in the mediastinal window.The signal-to-noise ratios and contrast-to-noise ratios of ascending aorta and liver in group B2 were higher than those in group B1 (all P<0.001) and lower than those in group A (all P<0.05).The image quality score of the mediastinal window followed a descending order of group A>group B2>group B1 (all P<0.001)。Conclusion ASiR-V combined with low tube voltage can effectively reduce the radiation dose and guarantee the image quality of chest CT of children in the plateau area.
Humans ; Radiation Dosage ; Tomography, X-Ray Computed/methods* ; Child ; Male ; Female ; Child, Preschool ; Radiography, Thoracic/methods* ; Infant ; Models, Statistical ; Image Processing, Computer-Assisted/methods*

Objective To develop a diagnostic model combining the CT angiography(CCTA)-derived myocardial radiomics signatures with the CT-derived fractional flow reserve(CT-FFR)based on coronary CCTA and investigate the diagnostic accuracy of the hybrid model for hemodynamically significant coronary artery disease(CAD).Methods The patients presenting stable angina pectoris,diagnosed with CAD,and clinically referred for CCTA examination and invasive coronary angiography were prospectively recruited.Radiomics features of the left ventricular myocardium were extracted from the three main perfusion territories demarcated according to the coronary blood supply.The extracted features were first selected by the minimum redundancy maximum relevance feature ranking method.A least absolute shrinkage and selection operator Logistic regression algorithm with leave-one-out cross-validation was then employed to construct a radiomics model.The CT-FFR value was generated for each blood vessel.The area under the receiver operating characteristics curve(AUC_ROC),sensitivity,and specificity were adopted to evaluate the performance of each model against the reference standard invasive coronary angiography/FFR.Results A total of 70 patients[42 men and 28 women;(61±10) years old] were included in this study and complemented CCTA examination,with 175 vessels and the corresponding myocardial territories undergoing invasive coronary angiography/FFR.A total of 1 656 specific radiomics parameters were extracted,from which 14 features were selected to establish the radiomics model.The AUC_ROC,sensitivity,and specificity were 0.797(95%CI=0.732-0.861),77.1%,and 73.7%for the radiomics model,0.892(95%CI=0.841-0.943),81.4%,and 88.8%for the CT-FFR model,and 0.928(95%CI=0.890-0.965),83.3%,and 88.4%for the hybrid model,respectively.The hybrid model outperformed the radiomics model and CT-FFR alone(P=0.040).Conclusions The radiomics signatures of the vessel-related myocardium from CCTA could provide incremental value to the diagnostic performance of CT-FFR and improve vessel-specific ischemia detection.The hybrid model combining CT-FFR with radiomics signatures is potentially feasible for improving the diagnostic accuracy for hemodynamically significant CAD.
Coronary Angiography/methods* ; Tomography, X-Ray Computed ; Humans ; Hemodynamics ; Coronary Artery Disease/diagnostic imaging* ; Male ; Female ; Middle Aged ; Aged ; Radiomics ; Angina Pectoris/diagnostic imaging* ; China ; Image Processing, Computer-Assisted ; Coronary Vessels/diagnostic imaging*

Synthetic CT (sCT) generated from CBCT has proven effective in artifact reduction and CT number correction, facilitating precise radiation dose calculation. However, the quality of different regions in sCT images is severely imbalanced, with soft tissue region exhibiting notably inferior quality compared to others. To address this imbalance, we proposed a Multi-Task Attention Network (MuTA-Net) based on VGG-16, specifically focusing the enhancement of image quality in soft tissue region of sCT. First, we introduced a multi-task learning strategy that divides the sCT generation task into three sub-tasks: global image generation, soft tissue region generation and bone region segmentation. This approach ensured the quality of overall sCT image while enhancing the network's focus on feature extraction and generation for soft tissues region. The result of bone region segmentation task guided the fusion of sub-tasks results. Then, we designed an attention module to further optimize feature extraction capabilities of the network. Finally, by employing a results fusion module, the results of three sub-tasks were integrated, generating a high-quality sCT image. Experimental results on head and neck CBCT demonstrated that the sCT images generated by the proposed MuTA-Net exhibited a 12.52% reduction in mean absolute error in soft tissue region, compared to the best performance among the three comparative methods, including ResNet, U-Net, and U-Net++. It can be seen that MuTA-Net is suitable for high-quality sCT image generation and has potential application value in the field of CBCT guided adaptive radiation therapy.
Cone-Beam Computed Tomography/methods* ; Humans ; Image Processing, Computer-Assisted/methods* ; Artifacts ; Algorithms ; Bone and Bones/diagnostic imaging* ; Neural Networks, Computer

In the clinical stage, suspected hemolytic plasma may cause hemolysis illness, manifesting as symptoms such as heart failure, severe anemia, etc. Applying a deep learning method to plasma images significantly improves recognition accuracy, so that this paper proposes a plasma quality detection model based on improved "You Only Look Once" 5th version (YOLOv5). Then the model presented in this paper and the evaluation system ‌were introduced‌ into the plasma datasets, and ‌the average accuracy of the final classification reached 98.7%‌. The results of this paper's experiment were obtained through the combination of several key algorithm modules including‌ omni-dimensional dynamic convolution, pooling with separable kernel attention, residual bi-fusion feature pyramid network, ‌and‌ re-parameterization convolution. The method of this paper‌ obtains the feature information of spatial mapping efficiently, and enhances the average recognition accuracy of plasma quality detection. This paper presents a high-efficiency detection method for plasma images, aiming to provide a practical approach to prevent hemolysis illnesses caused by external factors.
Algorithms ; Humans ; Hemolysis ; Plasma ; Deep Learning ; Image Processing, Computer-Assisted/methods*

Due to its irregular shape and varying contour, pancreas segmentation is a recognized challenge in medical image segmentation. Convolutional neural network (CNN) and Transformer-based networks perform well but have limitations: CNN have constrained receptive fields, and Transformer underutilize image features. This work proposes an improved pancreas segmentation method by combining CNN and Transformer. Point-wise separable convolution was introduced in a stage-wise encoder to extract more features with fewer parameters. A densely connected ensemble decoder enabled multi-scale feature fusion, addressing the structural constraints of skip connections. Consistency terms and contrastive loss were integrated into deep supervision to ensure model accuracy. Extensive experiments on the Changhai and National Institute of Health (NIH) pancreas datasets achieved the highest Dice similarity coefficient (DSC) values of 76.32% and 86.78%, with superiority in other metrics. Ablation studies validated each component's contributions to performance and parameter reduction. Results demonstrate that the proposed loss function smooths training and optimizes performance. Overall, the method outperforms other advanced methods, enhances pancreas segmentation performance, supports physician diagnosis, and provides a reliable reference for future research.
Humans ; Neural Networks, Computer ; Pancreas/diagnostic imaging* ; Image Processing, Computer-Assisted/methods* ; Algorithms ; Deep Learning

The accurate segmentation of breast ultrasound images is an important precondition for the lesion determination. The existing segmentation approaches embrace massive parameters, sluggish inference speed, and huge memory consumption. To tackle this problem, we propose T ²KD Attention U-Net (dual-Teacher Knowledge Distillation Attention U-Net), a lightweight semantic segmentation method combined double-path joint distillation in breast ultrasound images. Primarily, we designed two teacher models to learn the fine-grained features from each class of images according to different feature representation and semantic information of benign and malignant breast lesions. Then we leveraged the joint distillation to train a lightweight student model. Finally, we constructed a novel weight balance loss to focus on the semantic feature of small objection, solving the unbalance problem of tumor and background. Specifically, the extensive experiments conducted on Dataset BUSI and Dataset B demonstrated that the T ²KD Attention U-Net outperformed various knowledge distillation counterparts. Concretely, the accuracy, recall, precision, Dice, and mIoU of proposed method were 95.26%, 86.23%, 85.09%, 83.59%and 77.78% on Dataset BUSI, respectively. And these performance indexes were 97.95%, 92.80%, 88.33%, 88.40% and 82.42% on Dataset B, respectively. Compared with other models, the performance of this model was significantly improved. Meanwhile, compared with the teacher model, the number, size, and complexity of student model were significantly reduced (2.2×10 ⁶ vs. 106.1×10 ⁶, 8.4 MB vs. 414 MB, 16.59 GFLOPs vs. 205.98 GFLOPs, respectively). Indeedy, the proposed model guarantees the performances while greatly decreasing the amount of computation, which provides a new method for the deployment of clinical medical scenarios.
Humans ; Breast Neoplasms/diagnostic imaging* ; Female ; Ultrasonography, Mammary/methods* ; Image Processing, Computer-Assisted/methods* ; Algorithms ; Neural Networks, Computer ; Breast/diagnostic imaging*