VANet: a medical image fusion model based on attention mechanism to assist disease diagnosis.

BACKGROUND: Today's biomedical imaging technology has been able to present the morphological structure or functional metabolic information of organisms at different scale levels, such as organ, tissue, cell, molecule and gene. However, different imaging modes have different application scope, advantages and disadvantages. In order to improve the role of medical image in disease diagnosis, the fusion of biomedical image information at different imaging modes and scales has become an important research direction in medical image. Traditional medical image fusion methods are all designed to measure the activity level and fusion rules. They are lack of mining the context features of different modes of image, which leads to the obstruction of improving the quality of fused images. METHOD: In this paper, an attention-multiscale network medical image fusion model based on contextual features is proposed. The model selects five backbone modules in the VGG-16 network to build encoders to obtain the contextual features of medical images. It builds the attention mechanism branch to complete the fusion of global contextual features and designs the residual multiscale detail processing branch to complete the fusion of local contextual features. Finally, it completes the cascade reconstruction of features by the decoder to obtain the fused image. RESULTS: Ten sets of images related to five diseases are selected from the AANLIB database to validate the VANet model. Structural images are derived from MR images with high resolution and functional images are derived from SPECT and PET images that are good at describing organ blood flow levels and tissue metabolism. Fusion experiments are performed on twelve fusion algorithms including the VANet model. The model selects eight metrics from different aspects to build a fusion quality evaluation system to complete the performance evaluation of the fused images. Friedman's test and the post-hoc Nemenyi test are introduced to conduct professional statistical tests to demonstrate the superiority of VANet model. CONCLUSIONS: The VANet model completely captures and fuses the texture details and color information of the source images. From the fusion results, the metabolism and structural information of the model are well expressed and there is no interference of color information on the structure and texture; in terms of the objective evaluation system, the metric value of the VANet model is generally higher than that of other methods.; in terms of efficiency, the time consumption of the model is acceptable; in terms of scalability, the model is not affected by the input order of source images and can be extended to tri-modal fusion.

Hahn-PCNN-CNN: an end-to-end multi-modal brain medical image fusion framework useful for clinical diagnosis.

BACKGROUND: In medical diagnosis of brain, the role of multi-modal medical image fusion is becoming more prominent. Among them, there is no lack of filtering layered fusion and newly emerging deep learning algorithms. The former has a fast fusion speed but the fusion image texture is blurred; the latter has a better fusion effect but requires higher machine computing capabilities. Therefore, how to find a balanced algorithm in terms of image quality, speed and computing power is still the focus of all scholars. METHODS: We built an end-to-end Hahn-PCNN-CNN. The network is composed of feature extraction module, feature fusion module and image reconstruction module. We selected 8000 multi-modal brain medical images downloaded from the Harvard Medical School website to train the feature extraction layer and image reconstruction layer to enhance the network's ability to reconstruct brain medical images. In the feature fusion module, we use the moments of the feature map combined with the pulse-coupled neural network to reduce the information loss caused by convolution in the previous fusion module and save time. RESULTS: We choose eight sets of registered multi-modal brain medical images in four diease to verify our model. The anatomical structure images are from MRI and the functional metabolism images are SPECT and 18F-FDG. At the same time, we also selected eight representative fusion models as comparative experiments. In terms of objective quality evaluation, we select six evaluation metrics in five categories to evaluate our model. CONCLUSIONS: The fusion image obtained by our model can retain the effective information in source images to the greatest extent. In terms of image fusion evaluation metrics, our model is superior to other comparison algorithms. In terms of time computational efficiency, our model also performs well. In terms of robustness, our model is very stable and can be generalized to multi-modal image fusion of other organs.

Multi-Modal Medical Image Fusion Based on FusionNet in YIQ Color Space.

In order to obtain the physiological information and key features of source images to the maximum extent, improve the visual effect and clarity of the fused image, and reduce the computation, a multi-modal medical image fusion framework based on feature reuse is proposed. The framework consists of intuitive fuzzy processing (IFP), capture image details network (CIDN), fusion, and decoding. First, the membership function of the image is redefined to remove redundant features and obtain the image with complete features. Then, inspired by DenseNet, we proposed a new encoder to capture all the medical information features in the source image. In the fusion layer, we calculate the weight of each feature graph in the required fusion coefficient according to the trajectory of the feature graph. Finally, the filtered medical information is spliced and decoded to reproduce the required fusion image. In the encoding and image reconstruction networks, the mixed loss function of cross entropy and structural similarity is adopted to greatly reduce the information loss in image fusion. To assess performance, we conducted three sets of experiments on medical images of different grayscales and colors. Experimental results show that the proposed algorithm has advantages not only in detail and structure recognition but also in visual features and time complexity compared with other algorithms.

Universal Target Learning: An Efficient and Effective Technique for Semi-Naive Bayesian Learning.

To mitigate the negative effect of classification bias caused by overfitting, semi-naive Bayesian techniques seek to mine the implicit dependency relationships in unlabeled testing instances. By redefining some criteria from information theory, Target Learning (TL) proposes to build for each unlabeled testing instance P the Bayesian Network Classifier BNC P , which is independent and complementary to BNC T learned from training data T . In this paper, we extend TL to Universal Target Learning (UTL) to identify redundant correlations between attribute values and maximize the bits encoded in the Bayesian network in terms of log likelihood. We take the k-dependence Bayesian classifier as an example to investigate the effect of UTL on BNC P and BNC T . Our extensive experimental results on 40 UCI datasets show that UTL can help BNC improve the generalization performance.

Design of Meter-Scale Antenna and Signal Detection System for Underground Magnetic Resonance Sounding in Mines.

Magnetic resonance sounding (MRS) is a novel geophysical method to detect groundwater directly. By applying this method to underground projects in mines and tunnels, warning information can be provided on water bodies that are hidden in front prior to excavation and thus reduce the risk of casualties and accidents. However, unlike its application to ground surfaces, the application of MRS to underground environments is constrained by the narrow space, quite weak MRS signal, and complex electromagnetic interferences with high intensities in mines. Focusing on the special requirements of underground MRS (UMRS) detection, this study proposes the use of an antenna with different turn numbers, which employs a separated transmitter and receiver. We designed a stationary coil with stable performance parameters and with a side length of 2 m, a matching circuit based on a Q-switch and a multi-stage broad/narrowband mixed filter that can cancel out most electromagnetic noise. In addition, noises in the pass-band are further eliminated by adopting statistical criteria and harmonic modeling and stacking, all of which together allow weak UMRS signals to be reliably detected. Finally, we conducted a field case study of the UMRS measurement in the Wujiagou Mine in Shanxi Province, China, with known water bodies. Our results show that the method proposed in this study can be used to obtain UMRS signals in narrow mine environments, and the inverted hydrological information generally agrees with the actual situation. Thus, we conclude that the UMRS method proposed in this study can be used for predicting hazardous water bodies at a distance of 7-9 m in front of the wall for underground mining projects.

Removal of a series of spikes from magnetic resonance sounding signal by combining empirical mode decomposition and wavelet thresholding.

The magnetic resonance sounding (MRS) signal typically suffers from low signal-to-noise ratio (SNR). The MRS signal is severely distorted by noise, primarily harmonic and spiky noise. In terms of despiking, wavelet thresholding (WT) reconstructs the distorted content of the MRS signal, following isolation and elimination of the spiky sequence. However, WT cannot restore the MRS signal content completely when a series of spikes occurs within a given period of time. To solve this problem, a combined method of empirical mode decomposition (EMD) and WT for the removal of a series of spikes is proposed. EMD is first applied to decompose the noisy signal into several different components. The spikes that occur within a period of time are separated, the components without spikes are retained, and the components containing spiky events are selected and further processed by WT. After successively computing the wavelet coefficients of the selected components, the coefficients related to the spikes are isolated by threshold processing, and the subsequent wavelet reconstruction yields the sequence with the spikes removed. Finally, the denoised signal is obtained by adding the processed and retained components. The simulations on synthetic signals corrupted by artificial and real noise show that the proposed method improves the SNR with an accompanying improvement in the retrieval of the signal parameters. Moreover, the comparison results of the proposed and the WT methods suggest that the combined method efficiently removes a series of spikes.