High radiation dose in computed tomography (CT) scans increases the lifetime risk of cancer and has become a major clinical concern. Recently, iterative reconstruction algorithms with total variation (TV) regularization have been developed to reconstruct CT images from highly undersampled data acquired at low mAs levels order reduce imaging dose. Nonetheless, low-contrast structures tend be smoothed out by TV regularization, posing great challenge for method. To solve this problem, work we...

Purpose: To develop a novel algorithm that incorporates prior treatment knowledge into intensity modulated radiation therapy optimization to facilitate automatic planning and adaptive radiotherapy (ART) replanning. Methods: The automatically creates plan guided by the DVH curves of reference contains information on clinician-approved dose-volume trade-offs among different targets/organs portions curve for an organ. In ART, is initial same patient, while selected from library clinically...

Accurate segmentation of the prostate on computed tomography (CT) for treatment planning is challenging due to CT's poor soft tissue contrast. Magnetic resonance imaging (MRI) has been used aid delineation, but its final accuracy limited by MRI-CT registration errors. We developed a deep attention-based strategy CT-based synthetic MRI (sMRI) deal with CT delineation challenge without acquisition.We which employs an sMRI-aided attention network accurately segment CT. Our method consists three...

Abstract Motivation In recent years, a large number of biological experiments have strongly shown that miRNAs play an important role in understanding disease pathogenesis. The discovery miRNA–disease associations is beneficial for diagnosis and treatment. Since inferring these through time-consuming expensive, researchers sought to identify the utilizing computational approaches. Graph Convolutional Networks (GCNs), which exhibit excellent performance link prediction problems, been...

This study proposed a deep learning-based tracking method for ultrasound (US) image-guided radiation therapy. The cascade learning model is composed of an attention network, mask region-based convolutional neural network (mask R-CNN), and long short-term memory (LSTM) network. learns mapping from US image to suspected area landmark motion in order reduce the search region. R-CNN then produces multiple region-of-interest (ROI) proposals reduced region identifies via three heads: bounding box...

Predicting the associations between human microbes and drugs (MDAs) is one critical step in drug development precision medicine areas. Since discovering these through wet experiments time-consuming labor-intensive, computational methods have already been an effective way to tackle this problem. Recently, graph contrastive learning (GCL) approaches shown great advantages embeddings of nodes from heterogeneous biological graphs (HBGs). However, most GCL-based don't fully capture rich structure...

Abstract Motivation Accurately identifying the drug–target interactions (DTIs) is one of crucial steps in drug discovery and repositioning process. Currently, many computational-based models have already been proposed for DTI prediction achieved some significant improvement. However, these approaches pay little attention to fuse multi-view similarity networks related drugs targets an appropriate way. Besides, how fully incorporate known interaction relationships accurately represent not well...

Four-dimensional Cone Beam Computed Tomography (4D-CBCT) has been developed to provide respiratory phase resolved volumetric imaging in image guided radiation therapy (IGRT). Inadequate number of projections each bin results low quality 4D-CBCT images with obvious streaking artifacts. In this work, we propose two novel algorithms: an iterative reconstruction algorithm and enhancement algorithm, utilizing a temporal nonlocal means (TNLM) method. We define TNLM energy term for given set...

Purpose: Four‐dimensional computed tomography (4DCT) has been widely used in cancer radiotherapy for accurate target delineation and motion measurement tumors the thorax upper abdomen areas. However, its prolonged scanning duration causes a considerable increase of radiation dose compared to conventional CT, which is major concern clinical application. This work develop new algorithm reconstruct 4DCT images from undersampled projections acquired at low mA s levels order reduce imaging dose....

Accurate and automatic multi-needle detection in three-dimensional (3D) ultrasound (US) is a key step of treatment planning for US-guided brachytherapy. However, most current studies are concentrated on single-needle by only using small number images with needle, regardless the massive database US without needles. In this paper, we propose workflow considering needles as auxiliary. Concretely, train position-specific dictionaries 3D overlapping patches auxiliary images, where develop an...

Registration and fusion of magnetic resonance imaging (MRI) transrectal ultrasound (TRUS) the prostate can provide guidance for brachytherapy. However, accurate registration remains a challenging task due to lack ground truth regarding voxel-level spatial correspondence, limited field view, low contrast-to-noise ratio, signal-to-noise ratio in TRUS. In this study, we proposed fully automated deep learning approach based on weakly supervised method address these issues. We employed techniques...

Abstract Identifying disease-associated microRNAs (miRNAs) could help understand the deep mechanism of diseases, which promotes development new medicine. Recently, network-based approaches have been widely proposed for inferring potential associations between miRNAs and diseases. However, these ignore importance different relations in meta-paths when learning embeddings Besides, they pay little attention to screening out reliable negative samples is crucial improving prediction accuracy. In...

Monte Carlo (MC) simulation is commonly considered as the most accurate dose calculation method for proton therapy. Aiming at achieving fast MC calculations clinical applications, we have previously developed a graphics-processing unit (GPU)-based tool, gPMC. In this paper, report our recent updates on gPMC in terms of its accuracy, portability, and functionality, well comprehensive tests tool. The new version, v2.0, was under OpenCL environment to enable portability across different...

Low-dose computed tomography (CT) is desirable for treatment planning and simulation in radiation therapy. Multiple rescanning replanning during the course with a smaller amount of dose than single conventional full-dose CT crucial step adaptive We developed machine learning-based method to improve image quality low-dose therapy simulation. used residual block concept self-attention strategy cycle-consistent adversarial network framework. A fully convolution neural blocks attention gates...

Abstract Due to the inter- and intra- variation of respiratory motion, it is highly desired provide real-time volumetric images during treatment delivery lung stereotactic body radiation therapy (SBRT) for accurate active motion management. In this proof-of-concept study, we propose a novel generative adversarial network integrated with perceptual supervision derive instantaneous from single 2D projection. Our proposed network, named TransNet, consists three modules, i.e. encoding,...

Myopia, projected to affect 50% population globally by 2050, is a leading cause of vision loss. Eyes with pathological myopia exhibit distinctive shape distributions, which are closely linked the progression vision-threatening complications. Recent understanding eye-shape-based biomarkers requires magnetic resonance imaging (MRI), however, it costly and unrealistic in routine ophthalmology clinics. We present Fundus2Globe, first AI framework that synthesizes patient-specific 3D eye globes...

Protein-protein interactions play a fundamental role in biological systems. Accurate detection of protein-protein interaction sites (PPIs) remains challenge. And, the methods PPIs prediction based on experiments are expensive. Recently, lot computation-based have been developed and made great progress. However, current computational only focus one form protein, using protein spatial conformation or primary sequence. protein's natural hierarchical structure is ignored. In this study, we...

Spatial transcriptomics (ST) technology provides gene expression profiles with spatial context, offering critical insights into cellular interactions and tissue architecture. A core task in ST is domain identification, which involves detecting coherent regions similar patterns. However, existing methods often fail to fully exploit information, leading limited representational capacity suboptimal clustering accuracy. Here, we introduce MAEST, a novel graph neural network model designed...

In the sphere of Cone Beam Computed Tomography (CBCT), acquiring X-ray projections from sufficient angles is indispensable for traditional image reconstruction methods to accurately reconstruct 3D anatomical intricacies. However, this acquisition procedure linear accelerator-mounted CBCT systems in radiotherapy takes approximately one minute, impeding its use ultra-fast intra-fractional motion monitoring during treatment delivery. To address challenge, we introduce Patient-specific...

Adaptive radiation therapy (ART) can reduce normal tissue toxicity and/or improve tumor control through treatment adaptations based on the current patient anatomy. Developing an efficient and effective re-planning algorithm is important step toward clinical realization of ART. For process, manual trial-and-error approach to fine-tune planning parameters time-consuming usually considered unpractical, especially for online It desirable automate this yield a plan acceptable quality with minimal...

Multi-needle localization in ultrasound (US) images is a crucial step of treatment planning for US-guided prostate brachytherapy. However, current computer-aided technologies are mostly focused on single-needle digitization, while manual digitization labor intensive and time consuming. In this paper, we proposed deep learning-based workflow fast automatic multi-needle including needle shaft detection tip detection. The major composed two components: large margin mask R-CNN model (LMMask...