Medical & Biological Engineering & Computing最新文献

Predicting a patient's future health state through the analysis of their clinical records is an emerging area in the field of intelligent medicine. It has the potential to assist healthcare professionals in prescribing treatments safely, making more accurate diagnoses, and improving patient care. However, clinical notes have been underutilized due to their complexity, high dimensionality, and sparsity. Nevertheless, these clinical records hold significant promise for enhancing clinical decision. To tackle these problems, a novel feedback attention-based bidirectional long short-term memory (FABiLSTM) model has been proposed for more effective diagnosis using clinical records. This model incorporates PubMedBERT for filtering irrelevant information, enhances global vector word embeddings for numerical representations and K-means clustering, and performs to explore term frequency and inverse document frequency intricacies. The proposed approach excels in capturing information, aiding accurate disease prediction. The predictive capability is further enhanced with the help of a billiards-inspired optimization algorithm. The effectiveness of the FABiLSTM method has been assessed with the MIMIC-III dataset, yielding impressive results in accuracy, precision, F1 score, and recall score of 98.52%, 98%, 98.2%, and 98.2% individually. These results reveal ways in which the proposed technique excels in comparison with current practices.

Deep learning (DL) requires a large amount of training data to improve performance and prevent overfitting. To overcome these difficulties, we need to increase the size of the training dataset. This can be done by augmentation on a small dataset. The augmentation approaches must enhance the model's performance during the learning period. There are several types of transformations that can be applied to medical images. These transformations can be applied to the entire dataset or to a subset of the data, depending on the desired outcome. In this study, we categorize data augmentation methods into four groups: Absent augmentation, where no modifications are made; basic augmentation, which includes brightness and contrast adjustments; intermediate augmentation, encompassing a wider array of transformations like rotation, flipping, and shifting in addition to brightness and contrast adjustments; and advanced augmentation, where all transformation layers are employed. We plan to conduct a comprehensive analysis to determine which group performs best when applied to brain CT images. This evaluation aims to identify the augmentation group that produces the most favorable results in terms of improving model accuracy, minimizing diagnostic errors, and ensuring the robustness of the model in the context of brain CT image analysis.

Motor imagery (MI) based brain-computer interfaces (BCIs) decode the users' intentions from electroencephalography (EEG) to achieve information control and interaction between the brain and external devices. In this paper, firstly, we apply Riemannian geometry to the covariance matrix extracted by spatial filtering to obtain robust and distinct features. Then, a multiscale temporal-spectral segmentation scheme is developed to enrich the feature dimensionality. In order to determine the optimal feature configurations, we utilize a linear learning-based temporal window and spectral band (TWSB) selection method to evaluate the feature contributions, which efficiently reduces the redundant features and improves the decoding efficiency without excessive loss of accuracy. Finally, support vector machines are used to predict the classification labels based on the selected MI features. To evaluate the performance of our model, we test it on the publicly available BCI Competition IV dataset 2a and 2b. The results show that the method has an average accuracy of 79.1% and 83.1%, which outperforms other existing methods. Using TWSB feature selection instead of selecting all features improves the accuracy by up to about 6%. Moreover, the TWSB selection method can effectively reduce the computational burden. We believe that the framework reveals more interpretable feature information of motor imagery EEG signals, provides neural responses discriminative with high accuracy, and facilitates the performance of real-time MI-BCI.

We aimed to investigate the electrocardiogram (ECG) features in persons with chronic disorders of consciousness (DOC, ≥ 29 days since injury, DSI) resulted from the most severe brain damages. The ECG data from 30 patients with chronic DOC and 18 healthy controls (HCs) were recorded during resting wakefulness state for about five minutes. The patients were classified into vegetative state (VS) and minimally conscious state (MCS). Eight ECG metrics were extracted for comparisons between the subject subgroups, and regression analysis of the metrics were conducted on the DSI (29-593 days). The DOC patients exhibit a significantly higher heart rate (HR, p = 0.009) and lower values for SDNN (p = 0.001), CVRR (p = 0.009), and T-wave amplitude (p < 0.001) compared to the HCs. However, there're no significant differences in QRS, QT, QTc, or ST amplitude between the two groups (p > 0.05). Three ECG metrics of the DOC patients-HR, SDNN, and CVRR-are significantly correlated with the DSI. The ECG abnormalities persist in chronic DOC patients. The abnormalities are mainly manifested in the rhythm features HR, SDNN and CVRR, but not the waveform features such as QRS width, QT, QTc, ST and T-wave amplitudes.

The impact of drug delivery and particulate matter exposure on the human respiratory tract is influenced by various anatomical and physiological factors, particularly the structure of the respiratory tract and its fluid dynamics. This study employs computational fluid dynamics (CFD) to investigate airflow in two 3D models of the human air conducting zone. The first model uses a combination of CT-scan images and geometrical data from human cadaver to extract the upper and central airways down to the ninth generation, while the second model develops the lung airways from the first Carina to the end of the ninth generation using Kitaoka's deterministic algorithm. The study examines the differences in geometrical characteristics, airflow rates, velocity, Reynolds number, and pressure drops of both models in the inhalation and exhalation phases for different lobes and generations of the airways. From trachea to the ninth generation, the average air flowrates and Reynolds numbers exponentially decay in both models during inhalation and exhalation. The steady drop is the case for the average air velocity in Kitaoka's model, while that experiences a maximum in the 3rd or 4th generation in the quasi-realistic model. Besides, it is shown that the flow field remains laminar in the upper and central airways up to the total flow rate of 15 l/min. The results of this work can contribute to the understanding of flow behavior in upper respiratory tract.

In clinical practice, the morphology of the left atrial appendage (LAA) plays an important role in the selection of LAA closure devices for LAA closure procedures. The morphology determination is influenced by the segmentation results. The LAA occupies only a small part of the entire 3D medical image, and the segmentation results are more likely to be biased towards the background region, making the segmentation of the LAA challenging. In this paper, we propose a lightweight attention mechanism called fusion attention, which imitates human visual behavior. We process the 3D image of the LAA using a method that involves overview observation followed by detailed observation. In the overview observation stage, the image features are pooled along the three dimensions of length, width, and height. The obtained features from the three dimensions are then separately input into the spatial attention and channel attention modules to learn the regions of interest. In the detailed observation stage, the attention results from the previous stage are fused using element-wise multiplication and combined with the original feature map to enhance feature learning. The fusion attention mechanism was evaluated on a left atrial appendage dataset provided by Liaoning Provincial People's Hospital, resulting in an average Dice coefficient of 0.8855. The results indicate that the fusion attention mechanism achieves better segmentation results on 3D images compared to existing lightweight attention mechanisms.

Accurate brain tumor segmentation with multi-modal MRI images is crucial, but missing modalities in clinical practice often reduce accuracy. The aim of this study is to propose a mixture-of-experts and semantic-guided network to tackle the issue of missing modalities in brain tumor segmentation. We introduce a transformer-based encoder with novel mixture-of-experts blocks. In each block, four modality experts aim for modality-specific feature learning. Learnable modality embeddings are employed to alleviate the negative effect of missing modalities. We also introduce a decoder guided by semantic information, designed to pay higher attention to various tumor regions. Finally, we conduct extensive comparison experiments with other models as well as ablation experiments to validate the performance of the proposed model on the BraTS2018 dataset. The proposed model can accurately segment brain tumor sub-regions even with missing modalities. It achieves an average Dice score of 0.81 for the whole tumor, 0.66 for the tumor core, and 0.52 for the enhanced tumor across the 15 modality combinations, achieving top or near-top results in most cases, while also exhibiting a lower computational cost. Our mixture-of-experts and sematic-guided network achieves accurate and reliable brain tumor segmentation results with missing modalities, indicating its significant potential for clinical applications. Our source code is already available at https://github.com/MaggieLSY/MESG-Net .

The current diagnosis of diabetic retinopathy is based on fundus images and clinical experience. However, considering the ineffectiveness and non-portability of medical devices, we aimed to develop a diagnostic model for diabetic retinopathy based on glucose series data from the wearable continuous glucose monitoring system. Therefore, this study developed a novel method, i.e., double deep latent autoencoder, for exploring glycemic variability influence from multi-day glucose data for diabetic retinopathy. Specifically, the model proposed in this research could encode continuous glucose sensor data with non-continuous and variable length via the integration of a data reorganization module and a novel encoding module with fragmented-missing-wise objective function. Additionally, the model implements a double deep autoencoder, which integrated convolutional neural network, long short-term memory, to jointly capturing the inter-day and intra-day glucose latent features from glucose series. The effectiveness of the proposed model is evaluated through a cross-validation method to clinical datasets of 765 type 2 diabetes patients. The proposed method achieves the highest accuracy value (0.89), precision value (0.88), and F1 score (0.73). The results suggest that our model can be used to remotely diagnose and screen for diabetic retinopathy by learning potential features of glucose series data collected by wearable continuous glucose monitoring systems.

The gastrointestinal (GI) peristalsis is an involuntary wave-like contraction of the GI wall that helps to propagate food along the tract. Many GI diseases, e.g., gastroparesis, are known to cause motility disorders in which the physiological contractile patterns of the wall get disrupted. Therefore, to understand the pathophysiology of these diseases, it is necessary to understand the mechanism of GI motility. We present a coupled electromechanical model to describe the mechanism of GI motility and the transduction pathway of cellular electrical activities into mechanical deformation and the generation of intraluminal pressure (IP) waves in the GI tract. The proposed model consolidates a smooth muscle cell (SMC) model, an actin-myosin interaction model, a hyperelastic constitutive model, and a Windkessel model to construct a coupled model that can describe the origin of peristaltic contractions in the intestine. The key input to the model is external electrical stimuli, which are converted into mechanical contractile waves in the wall. The model recreated experimental observations efficiently and was able to establish a relationship between change in luminal volume and pressure with the compliance of the GI wall and the peripheral resistance to bolus flow. The proposed model will help us understand the GI tract's function in physiological and pathophysiological conditions.

Medical image denoising has been a subject of extensive research, with various techniques employed to enhance image quality and facilitate more accurate diagnostics. The evolution of denoising methods has highlighted impressive results but struggled to strike equilibrium between noise reduction and edge preservation which limits its applicability in various domains. This paper manifests the novel methodology that integrates an adaptive masking strategy, transformer-based U-Net Prior generator, edge enhancement module, and modified non-local block (MNLB) for denoising brain MRI clinical images. The adaptive masking strategy maintains the vital information through dynamic mask generation while the prior generator by capturing hierarchical features regenerates the high-quality prior MRI images. Finally, these images are fed to the edge enhancement module to boost structural information by maintaining crucial edge details, and the MNLB produces the denoised output by deriving non-local contextual information. The comprehensive experimental assessment is performed by employing two datasets namely the brain tumor MRI dataset and Alzheimer's dataset for diverse metrics and compared with conventional denoising approaches. The proposed denoising methodology achieves a PSNR of 40.965 and SSIM of 0.938 on the Alzheimer's dataset and also achieves a PSNR of 40.002 and SSIM of 0.926 on the brain tumor MRI dataset at a noise level of 50% revealing its supremacy in noise minimization. Furthermore, the impact of different masking ratios on denoising performance is analyzed which reveals that the proposed method showed PSNR of 40.965, SSIM of 0.938, MAE of 5.847, and MSE of 3.672 at the masking ratio of 60%. Moreover, the findings pave the way for the advancement of clinical image processing, facilitating precise detection of tumors in clinical MRI images.