Most methods for classifying hyperspectral data only consider the local spatial relationship among samples, ignoring the important non-local topological relationship. However, the non-local topological relationship is better at representing the structure of hyperspectral data. This paper proposes a deep learning model called Topology and semantic information fusion classification network (TSFnet) that incorporates a topology structure and semantic information transmission network to accurately classify traditional Chinese medicine in hyperspectral images. TSFnet uses a convolutional neural network (CNN) to extract features and a graph convolution network (GCN) to capture potential topological relationships among different types of Chinese herbal medicines. The results show that TSFnet outperforms other state-of-the-art deep learning classification algorithms in two different scenarios of herbal medicine datasets. Additionally, the proposed TSFnet model is lightweight and can be easily deployed for mobile herbal medicine classification.

Brain midline delineation can facilitate the clinical evaluation of brain midline shift, which has a pivotal role in the diagnosis and prognosis of various brain pathology. However, there are still challenges for brain midline delineation: 1) the largely deformed midline is hard to localize if mixed with severe cerebral hemorrhage; 2) the predicted midlines of recent methods are not smooth and continuous which violates the structural priority. To overcome these challenges, we propose an anisotropic three dimensional (3D) network with context-aware refinement (A3D-CAR) for brain midline modeling. The proposed network fuses 3D context from different two dimensional (2D) slices through asymmetric context fusion. To exploit the elongated structure of the midline, an anisotropic block is designed to balance the difference between the adjacent pixels in the horizontal and vertical directions. For maintaining the structural priority of a brain midline, we present a novel 3D connectivity regular loss (3D CRL) to penalize the disconnectivity between nearby coordinates. Extensive experiments on the CQ dataset and one in-house dataset show that the proposed method outperforms three state-of-the-art methods on four evaluation metrics without excessive computational burden.

This study is to compare three-dimensional (3D) isotropic T2-weighted magnetic resonance imaging (MRI) with compressed sensing-sampling perfection with application optimized contrast (CS-SPACE) and the conventional image (3D-SPACE) sequence in terms of image quality, estimated signal-to-noise ratio (SNR), relative contrast-to-noise ratio (CNR), and the lesions’ conspicuous of the female pelvis. Thirty-six females (age: 51, 28–73) with cervical carcinoma (n=20), rectal carcinoma (n=7), or uterine fibroid (n=9) were included. Patients underwent magnetic resonance (MR) imaging at a 3T scanner with the sequences of 3D-SPACE, CS-SPACE, and two-dimensional (2D) T2-weighted turbo-spin echo (TSE). Quantitative analyses of estimated SNR and relative CNR between tumors and other tissues, image quality, and tissue conspicuity were performed. Two radiologists assessed the difference in diagnostic findings for carcinoma. Quantitative values and qualitative scores were analyzed, respectively. The estimated SNR and the relative CNR of tumor-to-muscle obturator internus, tumor-to-myometrium, and myometrium-to-muscle obturator internus was comparable between 3D-SPACE and CS-SPACE. The overall image quality and the conspicuity of the lesion scores of the CS-SPACE were higher than that of the 3D-SPACE (P < 0.01). The CS-SPACE sequence offers shorter scan time, fewer artifacts, and comparable SNR and CNR to conventional 3D-SPACE, and has the potential to improve the performance of T2-weighted images.

Accurate histopathology classification is a crucial factor in the diagnosis and treatment of Cholangiocarcinoma (CCA). Hyperspectral images (HSI) provide rich spectral information than ordinary RGB images, making them more useful for medical diagnosis. The Convolutional Neural Network (CNN) is commonly employed in hyperspectral image classification due to its remarkable capacity for feature extraction and image classification. However, many existing CNN-based HSI classification methods tend to ignore the importance of image spatial context information and the interdependence between spectral channels, leading to unsatisfied classification performance. Thus, to address these issues, this paper proposes a Spatial-Spectral Joint Network (SSJN) model for hyperspectral image classification that utilizes spatial self-attention and spectral feature extraction. The SSJN model is derived from the ResNet18 network and implemented with the non-local and Coordinate Attention (CA) modules, which extract long-range dependencies on image space and enhance spatial features through the Branch Attention (BA) module to emphasize the region of interest. Furthermore, the SSJN model employs Conv-LSTM modules to extract long-range dependencies in the image spectral domain. This addresses the gradient disappearance/explosion phenomena and enhances the model classification accuracy. The experimental results show that the proposed SSJN model is more efficient in leveraging the spatial and spectral information of hyperspectral images on multidimensional microspectral datasets of CCA, leading to higher classification accuracy, and may have useful references for medical diagnosis of CCA.

Diffusion magnetic resonance imaging (dMRI) is a noninvasive method to capture the anisotropic pattern of water displacement in the neuronal tissue. The soma and neurite density imaging (SANDI) model introduced soma size and density to biophysical model for the first time. In addition to neurite density, it can achieve their joint estimation non-invasively using dMRI. In the traditional method, parameters of the SANDI are estimated in a maximum likelihood framework, where the nonlinear model fitting is computationally intensive. Also, the present methods require a large number of diffusion gradients. Efficient and accurate algorithms for tissue microstructure estimation of SANDI is still a challenge currently. Consequently, we introduce deep learning method for tissue microstructure estimation of the SANDI model. The model comprises two functional components. The first component produces the sparse representation of diffusion signals of input patches. The second component computes tissue microstructure from the sparse representation given by the first component. The deep network can produce not only tissue microstructure estimates but also the uncertainty of the estimates with a reduced number of diffusion gradients. Then, multiple deep networks are trained and their results are fused for the final prediction of tissue microstructure and uncertainty quantification. The deep network was evaluated on the MGH Connectome Diffusion Microstructure Dataset. Results indicate that our approach outperforms the traditional methods in terms of estimation accuracy.

In order to improve the registration accuracy of brain magnetic resonance images (MRI), some deep learning registration methods use segmentation images for training model. However, the segmentation values are constant for each label, which leads to the gradient variation concentrating on the boundary. Thus, the dense deformation field (DDF) is gathered on the boundary and there even appears folding phenomenon. In order to fully leverage the label information, the morphological opening and closing information maps are introduced to enlarge the non-zero gradient regions and improve the accuracy of DDF estimation. The opening information maps supervise the registration model to focus on smaller, narrow brain regions. The closing information maps supervise the registration model to pay more attention to the complex boundary region. Then, opening and closing morphology networks (OC_Net) are designed to automatically generate opening and closing information maps to realize the end-to-end training process. Finally, a new registration architecture, VM_seg+oc, is proposed by combining OC_Net and VoxelMorph. Experimental results show that the registration accuracy of VM_seg+oc is significantly improved on LPBA40 and OASIS1 datasets. Especially, VM_seg+oc can well improve registration accuracy in smaller brain regions and narrow regions.

To obtain excellent regression results under the condition of small sample hyperspectral data, a deep neural network with simulated annealing (SA-DNN) is proposed. According to the characteristics of data, the attention mechanism was applied to make the network pay more attention to effective features, thereby improving the operating efficiency. By introducing an improved activation function, the data correlation was reduced based on increasing the operation rate, and the problem of over-fitting was alleviated. By introducing simulated annealing, the network chose the optimal learning rate by itself, which avoided falling into the local optimum to the greatest extent. To evaluate the performance of the SA-DNN, the coefficient of determination (R²), root mean square error (RMSE), and other metrics were used to evaluate the model. The results show that the performance of the SA-DNN is significantly better than other traditional methods.

Accurate localization of cranial nerves and responsible blood vessels is important for diagnosing trigeminal neuralgia (TN) and hemifacial spasm (HFS). Manual delineation of the nerves and vessels on medical images is time-consuming and labor-intensive. Due to the development of convolutional neural networks (CNNs), the performance of medical image segmentation has been improved. In this work, we investigate the plans for automated segmentation of cranial nerves and responsible vessels for TN and HFS, which has not been comprehensively studied before. Different inputs are given to the CNN to find the best training configuration of segmenting trigeminal nerves, facial nerves, responsible vessels and brainstem, including the image modality and the number of segmentation targets. According to multiple experiments with seven training plans, we suggest training with the combination of three-dimensional fast imaging employing steady-state acquisition (3D-FIESTA) and three-dimensional time-of-flight magnetic resonance angiography (3D-TOF-MRA), and separate segmentation of cranial nerves and vessels.

In this paper, the main researches are focused on the horizontal nonlinear vibration characteristics of roll systems for rolling mill, mainly including the study of forced vibration and free vibration of the roller. Firstly, the nonlinear damping parameters and nonlinear stiffness parameters within interface of the rolling mill are both considered, and a fractional-order differential term is also introduced to model the horizontal nonlinear vibration. Secondly, the averaging method is introduced to solve the forced vibration system of the mill roll system, and the amplitude-frequency characteristic curves of the system are obtained for different orders, external excitation amplitudes, stiffness and fractional order coefficients. Thirdly, the amplitude-frequency and phase-frequency characteristics of the free vibration of the mill roll system are investigated at different fractional orders. Then, the accuracy of the averaging method for solving the dynamic characteristics of the system is verified by numerical analysis, and the effect of the fractional differential term coefficients and order on the dynamic characteristics of the roll system are investigated. Finally, the time-frequency characteristics and phase-frequency characteristics of free vibration systems at different fractional orders are studied. The validity of the theoretical study is also verified through experiments.