DrugMGR: a deep bioactive molecule binding method to identify compounds targeting proteins.

MOTIVATION: Understanding the intermolecular interactions of ligand-target pairs is key to guiding the optimization of drug research on cancers, which can greatly mitigate overburden workloads for wet labs. Several improved computational methods have been introduced and exhibit promising performance for these identification tasks, but some pitfalls restrict their practical applications: (i) first, existing methods do not sufficiently consider how multigranular molecule representations influence interaction patterns between proteins and compounds; and (ii) second, existing methods seldom explicitly model the binding sites when an interaction occurs to enable better prediction and interpretation, which may lead to unexpected obstacles to biological researchers. RESULTS: To address these issues, we here present DrugMGR, a deep multigranular drug representation model capable of predicting binding affinities and regions for each ligand-target pair. We conduct consistent experiments on three benchmark datasets using existing methods and introduce a new specific dataset to better validate the prediction of binding sites. For practical application, target-specific compound identification tasks are also carried out to validate the capability of real-world compound screen. Moreover, the visualization of some practical interaction scenarios provides interpretable insights from the results of the predictions. The proposed DrugMGR achieves excellent overall performance in these datasets, exhibiting its advantages and merits against state-of-the-art methods. Thus, the downstream task of DrugMGR can be fine-tuned for identifying the potential compounds that target proteins for clinical treatment. AVAILABILITY AND IMPLEMENTATION: https://github.com/lixiaokun2020/DrugMGR.

Boosting knowledge diversity, accuracy, and stability via tri-enhanced distillation for domain continual medical image segmentation.

Domain continual medical image segmentation plays a crucial role in clinical settings. This approach enables segmentation models to continually learn from a sequential data stream across multiple domains. However, it faces the challenge of catastrophic forgetting. Existing methods based on knowledge distillation show potential to address this challenge via a three-stage process: distillation, transfer, and fusion. Yet, each stage presents its unique issues that, collectively, amplify the problem of catastrophic forgetting. To address these issues at each stage, we propose a tri-enhanced distillation framework. (1) Stochastic Knowledge Augmentation reduces redundancy in knowledge, thereby increasing both the diversity and volume of knowledge derived from the old network. (2) Adaptive Knowledge Transfer selectively captures critical information from the old knowledge, facilitating a more accurate knowledge transfer. (3) Global Uncertainty-Guided Fusion introduces a global uncertainty view of the dataset to fuse the old and new knowledge with reduced bias, promoting a more stable knowledge fusion. Our experimental results not only validate the feasibility of our approach, but also demonstrate its superior performance compared to state-of-the-art methods. We suggest that our innovative tri-enhanced distillation framework may establish a robust benchmark for domain continual medical image segmentation.

TTN: Topological Transformer Network for Automated Coronary Artery Branch Labeling in Cardiac CT Angiography.

OBJECTIVE: Existing methods for automated coronary artery branch labeling in cardiac CT angiography face two limitations: 1) inability to model overall correlation of branches, since differences between branches cannot be captured directly. 2) a serious class imbalance between main and side branches. METHODS AND PROCEDURES: Inspired by the application of Transformer in sequence data, we propose a topological Transformer network (TTN), which solves the vessel branch labeling from a novel perspective of sequence labeling learning. TTN detects differences between branches by establishing their overall correlation. A topological encoding that represents the positions of vessel segments in the artery tree, is proposed to assist the model in classifying branches. Also, a segment-depth loss is introduced to solve the class imbalance between main and side branches. RESULTS: On a dataset with 325 CCTA, our method obtains the best overall result on all branches, the best result on side branches, and a competitive result on main branches. CONCLUSION: TTN solves two limitations in existing methods perfectly, thus achieving the best result in coronary artery branch labeling task. It is the first Transformer based vessel branch labeling method and is notably different from previous methods. CLINICAL IMPACT: This Pre-Clinical Research can be integrated into a computer-aided diagnosis system to generate cardiovascular disease diagnosis report, assisting clinicians in locating the atherosclerotic plaques.

An improved contrastive learning network for semi-supervised multi-structure segmentation in echocardiography.

Cardiac diseases have high mortality rates and are a significant threat to human health. Echocardiography is a commonly used imaging technique to diagnose cardiac diseases because of its portability, non-invasiveness and low cost. Precise segmentation of basic cardiac structures is crucial for cardiologists to efficiently diagnose cardiac diseases, but this task is challenging due to several reasons, such as: (1) low image contrast, (2) incomplete structures of cardiac, and (3) unclear border between the ventricle and the atrium in some echocardiographic images. In this paper, we applied contrastive learning strategy and proposed a semi-supervised method for echocardiographic images segmentation. This proposed method solved the above challenges effectively and made use of unlabeled data to achieve a great performance, which could help doctors improve the accuracy of CVD diagnosis and screening. We evaluated this method on a public dataset (CAMUS), achieving mean Dice Similarity Coefficient (DSC) of 0.898, 0.911, 0.916 with 1/4, 1/2 and full labeled data on two-chamber (2CH) echocardiography images, and of 0.903, 0.921, 0.928 with 1/4, 1/2 and full labeled data on four-chamber (4CH) echocardiography images. Compared with other existing methods, the proposed method had fewer parameters and better performance. The code and models are available at https://github.com/gpgzy/CL-Cardiac-segmentation.

MAE-TransRNet: An improved transformer-ConvNet architecture with masked autoencoder for cardiac MRI registration.

The heart is a relatively complex non-rigid motion organ in the human body. Quantitative motion analysis of the heart takes on a critical significance to help doctors with accurate diagnosis and treatment. Moreover, cardiovascular magnetic resonance imaging (CMRI) can be used to perform a more detailed quantitative analysis evaluation for cardiac diagnosis. Deformable image registration (DIR) has become a vital task in biomedical image analysis since tissue structures have variability in medical images. Recently, the model based on masked autoencoder (MAE) has recently been shown to be effective in computer vision tasks. Vision Transformer has the context aggregation ability to restore the semantic information in the original image regions by using a low proportion of visible image patches to predict the masked image patches. A novel Transformer-ConvNet architecture is proposed in this study based on MAE for medical image registration. The core of the Transformer is designed as a masked autoencoder (MAE) and a lightweight decoder structure, and feature extraction before the downstream registration task is transformed into the self-supervised learning task. This study also rethinks the calculation method of the multi-head self-attention mechanism in the Transformer encoder. We improve the query-key-value-based dot product attention by introducing both depthwise separable convolution (DWSC) and squeeze and excitation (SE) modules into the self-attention module to reduce the amount of parameter computation to highlight image details and maintain high spatial resolution image features. In addition, concurrent spatial and channel squeeze and excitation (scSE) module is embedded into the CNN structure, which also proves to be effective for extracting robust feature representations. The proposed method, called MAE-TransRNet, has better generalization. The proposed model is evaluated on the cardiac short-axis public dataset (with images and labels) at the 2017 Automated Cardiac Diagnosis Challenge (ACDC). The relevant qualitative and quantitative results (e.g., dice performance and Hausdorff distance) suggest that the proposed model can achieve superior results over those achieved by the state-of-the-art methods, thus proving that MAE and improved self-attention are more effective and promising for medical image registration tasks. Codes and models are available at https://github.com/XinXiao101/MAE-TransRNet.

SAGDTI: self-attention and graph neural network with multiple information representations for the prediction of drug-target interactions.

Motivation: Accurate identification of target proteins that interact with drugs is a vital step in silico, which can significantly foster the development of drug repurposing and drug discovery. In recent years, numerous deep learning-based methods have been introduced to treat drug-target interaction (DTI) prediction as a classification task. The output of this task is binary identification suggesting the absence or presence of interactions. However, existing studies often (i) neglect the unique molecular attributes when embedding drugs and proteins, and (ii) determine the interaction of drug-target pairs without considering biological interaction information. Results: In this study, we propose an end-to-end attention-derived method based on the self-attention mechanism and graph neural network, termed SAGDTI. The aim of this method is to overcome the aforementioned drawbacks in the identification of DTI. SAGDTI is the first method to sufficiently consider the unique molecular attribute representations for both drugs and targets in the input form of the SMILES sequences and three-dimensional structure graphs. In addition, our method aggregates the feature attributes of biological information between drugs and targets through multi-scale topologies and diverse connections. Experimental results illustrate that SAGDTI outperforms existing prediction models, which benefit from the unique molecular attributes embedded by atom-level attention and biological interaction information representation aggregated by node-level attention. Moreover, a case study on severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) shows that our model is a powerful tool for identifying DTIs in real life. Availability and implementation: The data and codes underlying this article are available in Github at https://github.com/lixiaokun2020/SAGDTI.

Feasibility Study of Intelligent Three-Dimensional Accurate Liver Reconstruction Technology Based on MRI Data.

Intelligent three-dimensional (3D) reconstruction technology plays an important role in the diagnosis and treatment of diseases. It has been widely used in assisted liver surgery. At present, the 3D reconstruction information of liver is mainly obtained based on CT enhancement data. It has also been commercialized. However, there are few reports on the display of 3D reconstruction information of the liver based on MRI. The purpose of this study is to propose a new idea of intelligent 3D liver reconstruction based on MRI technology and verify its feasibility. Two different liver scanning data (CT and MRI) were selected from the same batch of patients at the same time (patients with a time interval of no more than two weeks and without surgery). The results of liver volume, segmentation, tumor, and simulated surgery based on MRI volume data were compared with those based on CT data. The results show that the results of 3D reconstruction based on MRI data are highly consistent with those based on CT 3D reconstruction. At the same time, in addition to providing the information provided by CT 3D reconstruction, it also has its irreplaceable advantages. For example, multi-phase (early, middle and late arterial, hepatobiliary, etc.) scanning of MRI technology can provide more disease information and display of biliary diseases. In a word, MRI technology can be used for 3D reconstruction of the liver. Hence, a new feasible and effective method to show the liver itself and its disease characteristics is proposed.

Deep Atlas Network for Efficient 3D Left Ventricle Segmentation on Echocardiography.

We proposed a novel efficient method for 3D left ventricle (LV) segmentation on echocardiography, which is important for cardiac disease diagnosis. The proposed method effectively overcame the 3D echocardiography's challenges: high dimensional data, complex anatomical environments, and limited annotation data. First, we proposed a deep atlas network, which integrated LV atlas into the deep learning framework to address the 3D LV segmentation problem on echocardiography for the first time, and improved the performance based on limited annotation data. Second, we proposed a novel information consistency constraint to enhance the model's performance from different levels simultaneously, and finally achieved effective optimization for 3D LV segmentation on complex anatomical environments. Finally, the proposed method was optimized in an end-to-end back propagation manner and it achieved high inference efficiency even with high dimensional data, which satisfies the efficiency requirement of clinical practice. The experiments proved that the proposed method achieved better segmentation results and a higher inference speed compared with state-of-the-art methods. The mean surface distance, mean hausdorff surface distance, and mean dice index were 1.52 mm, 5.6 mm and 0.97 respectively. What's more, the method is efficient and its inference time is 0.02s. The experimental results proved that the proposed method has a potential clinical application for 3D LV segmentation on echocardiography.

Commensal correlation network between segmentation and direct area estimation for bi-ventricle quantification.

Accurate and automated cardiac bi-ventricle quantification based on cardiac magnetic resonance (CMR) image is a very crucial procedure for clinical cardiac disease diagnosis. Two traditional and commensal tasks, i.e., bi-ventricle segmentation and direct ventricle function index estimation, are always independently devoting to address ventricle quantification problem. However, because of inherent difficulties from the variable CMR imaging conditions, these two tasks are still open challenging. In this paper, we proposed a unified bi-ventricle quantification framework based on commensal correlation between the bi-ventricle segmentation and direct area estimation. Firstly, we proposed the area commensal correlation between the two traditional cardiac quantification tasks for the first time, and designed a novel deep commensal network (DCN) to join these two commensal tasks into a unified framework based on the proposed commensal correlation loss. Secondly, we proposed an differentiable area operator to model the proposed area commensal correlation and made the proposed model continuously differentiable. Thirdly, we proposed a high-efficiency and novel uncertainty estimation method through one-time inference based on cross-task output variability. And finally DCN achieved end-to-end optimization and fast convergence as well as uncertainty estimation with one-time inference. Experiments on the four open accessible short-axis CMR benchmark datasets (i.e., Sunnybrook, STACOM 2011, RVSC, and ACDC) showed that the proposed method achieves best bi-ventricle quantification accuracy and optimization performance. Hence, the proposed method has big potential to be extended to other medical image analysis tasks and has clinical application value.

A Combined Fully Convolutional Networks and Deformable Model for Automatic Left Ventricle Segmentation Based on 3D Echocardiography.

Segmentation of the left ventricle (LV) from three-dimensional echocardiography (3DE) plays a key role in the clinical diagnosis of the LV function. In this work, we proposed a new automatic method for the segmentation of LV, based on the fully convolutional networks (FCN) and deformable model. This method implemented a coarse-to-fine framework. Firstly, a new deep fusion network based on feature fusion and transfer learning, combining the residual modules, was proposed to achieve coarse segmentation of LV on 3DE. Secondly, we proposed a method of geometrical model initialization for a deformable model based on the results of coarse segmentation. Thirdly, the deformable model was implemented to further optimize the segmentation results with a regularization item to avoid the leakage between left atria and left ventricle to achieve the goal of fine segmentation of LV. Numerical experiments have demonstrated that the proposed method outperforms the state-of-the-art methods on the challenging CETUS benchmark in the segmentation accuracy and has a potential for practical applications.

Multi-Views Fusion CNN for Left Ventricular Volumes Estimation on Cardiac MR Images.

OBJECTIVE: Left ventricular (LV) volume estimation is a critical procedure for cardiac disease diagnosis. The objective of this paper is to address a direct LV volume prediction task. METHODS: In this paper, we propose a direct volume prediction method based on the end-to-end deep convolutional neural networks. We study the end-to-end LV volume prediction method in items of the data preprocessing, network structure, and multiview fusion strategy. The main contributions of this paper are the following aspects. First, we propose a new data preprocessing method on cardiac magnetic resonance (CMR). Second, we propose a new network structure for end-to-end LV volume estimation. Third, we explore the representational capacity of different slices and propose a fusion strategy to improve the prediction accuracy. RESULTS: The evaluation results show that the proposed method outperforms other state-of-the-art LV volume estimation methods on the open accessible benchmark datasets. The clinical indexes derived from the predicted volumes agree well with the ground truth ( ${\rm{EDV:R}}^{{\rm 2}}={\text{0.974}}$, ${\rm{RMSE\,}}= {\text{9.6}}{\rm{\,ml}}$; ${\rm{ESV:R}}^{{\rm 2}}={\text{0.976}}$, ${\rm{RMSE}}= {\text{7.1}}\,{\text{ml}}$; ${\rm{EF:R}}^{{\rm 2}} ={\text{0.828}}$, ${\rm{RMSE}}= {\text{4.71}}\% $). CONCLUSION: Experimental results prove that the proposed method may be useful for the LV volume prediction task. SIGNIFICANCE: The proposed method not only has application potential for cardiac diseases screening for large-scale CMR data, but also can be extended to other medical image research fields.

Computational Cardiac Modeling Reveals Mechanisms of Ventricular Arrhythmogenesis in Long QT Syndrome Type 8: CACNA1C R858H Mutation Linked to Ventricular Fibrillation.

Functional analysis of the L-type calcium channel has shown that the CACNA1C R858H mutation associated with severe QT interval prolongation may lead to ventricular fibrillation (VF). This study investigated multiple potential mechanisms by which the CACNA1C R858H mutation facilitates and perpetuates VF. The Ten Tusscher-Panfilov (TP06) human ventricular cell models incorporating the experimental data on the kinetic properties of L-type calcium channels were integrated into one-dimensional (1D) fiber, 2D sheet, and 3D ventricular models to investigate the pro-arrhythmic effects of CACNA1C mutations by quantifying changes in intracellular calcium handling, action potential profiles, action potential duration restitution (APDR) curves, dispersion of repolarization (DOR), QT interval and spiral wave dynamics. R858H "mutant" L-type calcium current (ICaL ) augmented sarcoplasmic reticulum calcium content, leading to the development of afterdepolarizations at the single cell level and focal activities at the tissue level. It also produced inhomogeneous APD prolongation, causing QT prolongation and repolarization dispersion amplification, rendering R858H "mutant" tissue more vulnerable to the induction of reentry compared with other conditions. In conclusion, altered ICaL due to the CACNA1C R858H mutation increases arrhythmia risk due to afterdepolarizations and increased tissue vulnerability to unidirectional conduction block. However, the observed reentry is not due to afterdepolarizations (not present in our model), but rather to a novel blocking mechanism.