Assessing lymph node metastasis (LNM) involvement in patients with rectal cancer (RC) is fundamental in disease management. In this study, we used artificial intelligence (AI) technology to develop a segmentation model that automatically segments the tumor core area and mesangial tissue from magnetic resonance T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) images collected from 122 RC patients to improve the accuracy of LNM prediction, after which omics machine modeling was performed on the segmented ROI. An automatic segmentation model was developed using nn-UNet. This pipeline integrates deep learning (DL), specifically 3D U-Net, for semantic segmentation and image processing techniques such as resampling, normalization, connected component analysis, image registration, and radiomics features coupled with machine learning. The results showed that the DL segmentation method could effectively segment the tumor and mesangial areas from MR sequences (the median dice coefficient: 0.90 ± 0.08; mesorectum segmentation: 0.85 ± 0.36), and the radiological characteristics of rectal and mesangial tissues in T2WI and ADC images could help distinguish RC treatments. The nn-UNet model demonstrated promising preliminary results, achieving the highest area under the curve (AUC) values in various scenarios. In the evaluation encompassing both tumor lesions and mesorectum involvement, the model exhibited an AUC of 0.743, highlighting its strong discriminatory ability to predict a combined outcome involving both elements. Specifically targeting tumor lesions, the model achieved an AUC of 0.731, emphasizing its effectiveness in distinguishing between positive and negative cases of tumor lesions. In assessing the prediction of mesorectum involvement, the model displayed moderate predictive utility with an AUC of 0.753. The nn-UNet model demonstrated impressive performance across all evaluated scenarios, including combined tumor lesions and mesorectum involvement, tumor lesions alone, and mesorectum involvement alone. The online version contains supplementary material available at 10.1186/s12880-025-01878-9.

Ultrasound (US) imaging is a critical tool in medical diagnostics, offering real-time visualization of physiological processes. One of its major advantages is its ability to capture temporal dynamics, which is essential for assessing motion patterns in applications such as cardiac monitoring, fetal development, and vascular imaging. Despite its importance, current deep learning models often overlook the temporal continuity of ultrasound sequences, analyzing frames independently and missing key temporal dependencies. To address this gap, we propose a method for learning effective temporal representations from ultrasound videos, with a focus on echocardiography-based ejection fraction (EF) estimation. EF prediction serves as an ideal case study to demonstrate the necessity of temporal learning, as it requires capturing the rhythmic contraction and relaxation of the heart. Our approach leverages temporally consistent masking and contrastive learning to enforce temporal coherence across video frames, enhancing the model's ability to represent motion patterns. Evaluated on the EchoNet-Dynamic dataset, our method achieves a substantial improvement in EF prediction accuracy, highlighting the importance of temporally-aware representation learning for real-time ultrasound analysis.

Rotator cuff muscle pathology affects outcomes following total shoulder arthroplasty, yet current assessment methods lack reliability in quantifying muscle atrophy and fat infiltration. We developed a deep learning-based model for automated segmentation of rotator cuff muscles on computed tomography (CT) and propose a T-score classification of volumetric muscle atrophy. We further characterized distinct atrophy phenotypes, 3D fat infiltration percentage (3DFI%), and anterior-posterior (AP) balance, which were compared between healthy controls, anatomic total shoulder arthroplasty (aTSA), and reverse total shoulder arthroplasty (rTSA) patients. 952 shoulder CT scans were included (762 controls, 103 undergoing aTSA for glenohumeral osteoarthritis, and 87 undergoing rTSA for cuff tear arthropathy. A deep learning model was developed to allow automated segmentation of supraspinatus (SS), subscapularis (SC), infraspinatus (IS) and teres minor (TM). Muscle volumes were normalized to scapula volume, and control muscle volumes were referenced to calculate T-scores for each muscle. T-scores were classified as no atrophy (>-1.0), moderate atrophy (-1 to -2.5), and severe atrophy (<-2.5). 3DFI% was quantified as the proportion of fat within each muscle using Hounsfield unit thresholds. The T-scores, 3DFI%, and AP balance were compared between the three cohorts. The aTSA cohort had significantly greater atrophy in all muscles compared to control (p<0.001), whereas the rTSA cohort had significantly greater atrophy in SS, SC, and IS than aTSA (p<0.001). In the aTSA cohort, the most common phenotype was SS_severe/SC_moderate/IS+TM_moderate, while in the rTSA cohort it was SS_severe/SC_moderate/IS+TM_severe. The aTSA group had significantly higher 3DFI% compared to controls for all muscles (p<0.001), while the rTSA cohort had significantly higher 3DFI% than aTSA and control cohorts for all muscles (p<0.001). Additionally, the aTSA cohort had a significantly lower AP muscle volume ratio (1.06 vs. 1.14, p<0.001), whereas the rTSA group had a significantly higher AP muscle volume ratio than the control cohort (1.31 vs. 1.14, p<0.001). Our study demonstrates successful development of a deep learning model for automated volumetric assessment of rotator cuff muscle atrophy, 3DFI% and AP balance on shoulder CT scans. We found that aTSA patients had significantly greater muscle atrophy and 3DFI% than controls, while the rTSA patients had the most severe muscle atrophy and 3DFI%. Additionally, distinct phenotypes of muscle atrophy and AP muscle balance exist in aTSA and rTSA that warrant further investigation with regards to shoulder arthroplasty outcomes.

Abnormality detection in medical imaging is a critical task requiring both high efficiency and accuracy to support effective diagnosis. While convolutional neural networks (CNNs) and Transformer-based models are widely used, both face intrinsic challenges: CNNs have limited receptive fields, restricting their ability to capture broad contextual information, and Transformers encounter prohibitive computational costs when processing high-resolution medical images. Mamba, a recent innovation in natural language processing, has gained attention for its ability to process long sequences with linear complexity, offering a promising alternative. Building on this foundation, we present SpectMamba, the first Mamba-based architecture designed for medical image detection. A key component of SpectMamba is the Hybrid Spatial-Frequency Attention (HSFA) block, which separately learns high- and low-frequency features. This approach effectively mitigates the loss of high-frequency information caused by frequency bias and correlates frequency-domain features with spatial features, thereby enhancing the model's ability to capture global context. To further improve long-range dependencies, we propose the Visual State-Space Module (VSSM) and introduce a novel Hilbert Curve Scanning technique to strengthen spatial correlations and local dependencies, further optimizing the Mamba framework. Comprehensive experiments show that SpectMamba achieves state-of-the-art performance while being both effective and efficient across various medical image detection tasks.

Aortic dissection (AD) is a lethal emergency requiring prompt diagnosis. Current computed tomography angiography (CTA)-based diagnosis requires contrast agents, which expends time, whereas existing deep learning (DL) models only support single-modality inputs [non-contrast computed tomography (CT) or CTA]. In this study, we propose a bimodal DL framework to independently process both types, enabling dual-path detection and improving diagnostic efficiency. Patients who underwent non-contrast CT and CTA from February 2016 to September 2021 were retrospectively included from three institutions, including the First Affiliated Hospital, Zhejiang University School of Medicine (Center I), Zhejiang Hospital (Center II), and Yiwu Central Hospital (Center III). A two-stage DL model for predicting AD was developed. The first stage used an aorta detection network (AoDN) to localize the aorta in non-contrast CT or CTA images. Image patches that contained detected aorta were cut from CT images and combined to form an image patch sequence, which was inputted to an aortic dissection diagnosis network (ADDiN) to diagnose AD in the second stage. The following performances were assessed: aorta detection and diagnosis using average precision at the intersection over union threshold 0.5 ([email protected]) and area under the receiver operating characteristic curve (AUC). The first cohort, comprising 102 patients (53±15 years, 80 men) from two institutions, was used for the AoDN, whereas the second cohort, consisting of 861 cases (55±15 years, 623 men) from three institutions, was used for the ADDiN. For the AD task, the AoDN achieved [email protected] 99.14% on the non-contrast CT test set and 99.34% on the CTA test set, respectively. For the AD diagnosis task, the ADDiN obtained an AUCs of 0.98 on the non-contrast CT test set and 0.99 on the CTA test set. The proposed bimodal CT data-driven DL model accurately diagnoses AD, facilitating prompt hospital diagnosis and treatment of AD.

Ki-67 labelling index (LI), a critical marker of tumor proliferation, is vital for grading adult-type diffuse gliomas and predicting patient survival. However, its accurate assessment currently relies on invasive biopsy or surgical resection. This makes it challenging to non-invasively predict Ki-67 LI and subsequent prognosis. Therefore, this study aimed to investigate whether histogram analysis of multi-parametric diffusion model metrics-specifically diffusion tensor imaging (DTI), diffusion kurtosis imaging (DKI), and neurite orientation dispersion and density imaging (NODDI)-could help predict Ki-67 LI in adult-type diffuse gliomas and further predict patient survival. A total of 123 patients with diffuse gliomas who underwent preoperative bipolar spin-echo diffusion magnetic resonance imaging (MRI) were included. Diffusion metrics (DTI, DKI and NODDI) and their histogram features were extracted and used to develop a nomogram model in the training set (n=86), and the performance was verified in the test set (n=37). Area under the receiver operating characteristics curve of the nomogram model was calculated. The outcome cohort, including 123 patients, was used to evaluate the predictive value of the diffusion nomogram model for overall survival (OS). Cox proportion regression was performed to predict OS. Among 123 patients, 87 exhibited high Ki-67 LI (Ki-67 LI >5%). The patients had a mean age of 46.08±13.24 years, with 39 being female. Tumor grading showed 46 cases of grade 2, 21 cases of grade 3, and 56 cases of grade 4. The nomogram model included eight histogram features from diffusion MRI and showed good performance for prediction Ki-67 LI, with area under the receiver operating characteristic curves (AUCs) of 0.92 [95% confidence interval (CI): 0.85-0.98, sensitivity =0.85, specificity =0.84] and 0.84 (95% CI: 0.64-0.98, sensitivity =0.77, specificity =0.73) in the training set and test set, respectively. Further nomogram incorporating these variables showed good discrimination in Ki-67 LI predicting and glioma grading. A low nomogram model score relative to the median value in the outcomes cohort was independently associated with OS (P<0.01). Accurate prediction of the Ki-67 LI in adult-type diffuse glioma patients was achieved by using multi-modal diffusion MRI histogram radiomics model, which also reliably and accurately determined survival. ClinicalTrials.gov Identifier: NCT06572592.

Deep learning-driven super resolution (SR) aims to enhance the quality and resolution of images, offering potential benefits in dental imaging. Although extensive research has focused on deep learning based dental classification tasks, the impact of applying SR techniques on classification remains underexplored. This study seeks to address this gap by evaluating and comparing the performance of deep learning classification models on dental images with and without SR enhancement. An open-source dental image dataset was utilized to investigate the impact of SR on image classification performance. SR was applied by 2 models with a scaling ratio of 2 and 4, while classification was performed by 4 deep learning models. Performances were evaluated by well-accepted metrics like structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), accuracy, recall, precision, and F1 score. The effect of SR on classification performance is interpreted through 2 different approaches. Two SR models yielded average SSIM and PSNR values of 0.904 and 36.71 for increasing resolution with 2 scaling ratios. Average accuracy and F-1 score for the classification trained and tested with 2 SR-generated images were 0.859 and 0.873. In the first of the comparisons carried out with 2 different approaches, it was observed that the accuracy increased in at least half of the cases (8 out of 16) when different models and scaling ratios were considered, while in the second approach, SR showed a significantly higher performance for almost all cases (12 out of 16). This study demonstrated that the classification with SR-generated images significantly improved outcomes. For the first time, the classification performance of dental radiographs with improved resolution by SR has been investigated. Significant performance improvement was observed compared to the case without SR.

The nnUNet segmentation framework adeptly adjusts most hyperparameters in training scripts automatically, but it overlooks the tuning of internal hyperparameters within the segmentation network itself, which constrains the model's ability to generalize. Addressing this limitation, this study presents a novel Self-Adaptive Convolution Module that dynamically adjusts the size of the convolution kernels depending on the unique fingerprints of different datasets. This adjustment enables the MSA2-Net, when equipped with this module, to proficiently capture both global and local features within the feature maps. Self-Adaptive Convolution Module is strategically integrated into two key components of the MSA2-Net: the Multi-Scale Convolution Bridge and the Multi-Scale Amalgamation Decoder. In the MSConvBridge, the module enhances the ability to refine outputs from various stages of the CSWin Transformer during the skip connections, effectively eliminating redundant data that could potentially impair the decoder's performance. Simultaneously, the MSADecoder, utilizing the module, excels in capturing detailed information of organs varying in size during the decoding phase. This capability ensures that the decoder's output closely reproduces the intricate details within the feature maps, thus yielding highly accurate segmentation images. MSA2-Net, bolstered by this advanced architecture, has demonstrated exceptional performance, achieving Dice coefficient scores of 86.49\%, 92.56\%, 93.37\%, and 92.98\% on the Synapse, ACDC, Kvasir, and Skin Lesion Segmentation (ISIC2017) datasets, respectively. This underscores MSA2-Net's robustness and precision in medical image segmentation tasks across various datasets.

Cone-beam computed tomography (CBCT) is a three-dimensional (3D) imaging method designed for routine target verification of cancer patients during radiotherapy. The images are reconstructed from a sequence of projection images obtained by the on-board imager attached to a radiotherapy machine. CBCT images are usually stored in a health information system, but the projection images are mostly abandoned due to their massive volume. To store them economically, in this study, a deep learning (DL)-based super-resolution (SR) method for compressing the projection images was investigated. In image compression, low-resolution (LR) images were down-sampled by a factor from the high-resolution (HR) projection images and then encoded to the video file. In image restoration, LR images were decoded from the video file and then up-sampled to HR projection images via the DL network. Three SR DL networks, convolutional neural network (CNN), residual network (ResNet), and generative adversarial network (GAN), were tested along with three video coding-decoding (CODEC) algorithms: Advanced Video Coding (AVC), High Efficiency Video Coding (HEVC), and AOMedia Video 1 (AV1). Based on the two databases of the natural and projection images, the performance of the SR networks and video codecs was evaluated with the compression ratio (CR), peak signal-to-noise ratio (PSNR), video quality metric (VQM), and structural similarity index measure (SSIM). The codec AV1 achieved the highest CR among the three codecs. The CRs of AV1 were 13.91, 42.08, 144.32, and 289.80 for the down-sampling factor (DSF) 0 (non-SR) 2, 4, and 6, respectively. The SR network, ResNet, achieved the best restoration accuracy among the three SR networks. Its PSNRs were 69.08, 41.60, 37.08, and 32.44 dB for the four DSFs, respectively; its VQMs were 0.06%, 3.65%, 6.95%, and 13.03% for the four DSFs, respectively; and its SSIMs were 0.9984, 0.9878, 0.9798, and 0.9518 for the four DSFs, respectively. As the DSF increased, the CR increased proportionally with the modest degradation of the restored images. The application of the SR model can further improve the CR based on the current result achieved by the video encoders. This compression method is not only effective for the two-dimensional (2D) projection images, but also applicable to the 3D images used in radiotherapy.

This study explores the application of artificial intelligence (AI), specifically deep learning, in the detection and classification of mandibular fractures using CT scans. Data from 459 patients were retrospectively obtained from West China Hospital of Stomatology, Sichuan University, spanning from 2020 to 2023. The CT scans were divided into training, testing, and independent validation sets. This research focuses on training and validating a deep learning model using the nnU-Net segmentation framework for pixel-level accuracy in identifying fracture locations. Additionally, a 3D-ResNet with pre-trained weights was employed to classify fractures into 3 types based on severity. Performance metrics included sensitivity, precision, specificity, and area under the receiver operating characteristic curve (AUC). The study achieved high diagnostic accuracy in mandibule fracture detection, with sensitivity >0.93, precision >0.79, and specificity >0.80. For mandibular fracture classification, accuracies were all above 0.718, with a mean AUC of 0.86. Detection and classification of mandibular fractures in CT images can be significantly enhanced using the nnU-Net segmentation framework, aiding in clinical diagnosis.