Bioengineering最新文献

Falls are a major health hazard for older adults; therefore, in the context of an aging population, predicting the risk of a patient suffering falls in the near future is of great impact for health care systems. Currently, the standard prospective fall risk assessment instrument relies on a set of clinical and functional mobility assessment tools, one of them being the Timed Up and Go (TUG) test. Recently, wearable inertial measurement units (IMUs) have been proposed to capture motion data that would allow for the building of estimates of fall risk. The hypothesis of this study is that the data gathered from IMU readings while the patient is performing the TUG test can be used to build a predictive model that would provide an estimate of the probability of suffering a fall in the near future, i.e., assessing prospective fall risk. This study applies deep learning convolutional neural networks (CNN) and recurrent neural networks (RNN) to build such predictive models based on features extracted from IMU data acquired during TUG test realizations. Data were obtained from a cohort of 106 older adults wearing wireless IMU sensors with sampling frequencies of 100 Hz while performing the TUG test. The dependent variable is a binary variable that is true if the patient suffered a fall in the six-month follow-up period. This variable was used as the output variable for the supervised training and validations of the deep learning architectures and competing machine learning approaches. A hold-out validation process using 75 subjects for training and 31 subjects for testing was repeated one hundred times to obtain robust estimations of model performances At each repetition, 5-fold cross-validation was carried out to select the best model over the training subset. Best results were achieved by a bidirectional long short-term memory (BLSTM), obtaining an accuracy of 0.83 and AUC of 0.73 with good sensitivity and specificity values.

In this in vitro study, we aimed to evaluate and compare the antibacterial efficacy of a novel erbium-doped yttrium aluminum garnet laser modality, shock wave enhanced emission of photoacoustic streaming (SWEEPS), ultrasonically activated irrigation (UAI), and single needle irrigation (SNI) against old bacterial biofilm. A two-week-old Enterococcus faecalis biofilm was cultivated on transversal dentinal discs made from the middle third of the roots of single-rooted, single-canal premolars. Biofilm growth was confirmed using scanning electron microscopy (SEM) and confocal laser scanning microscopy (CLSM). The dentine samples were randomly distributed into three experimental groups and one control group based on the irrigation protocol used: Group 1, SWEEPS; Group 2, UAI; and Group 3, SNI. The root canals were irrigated with a 3% sodium hypochlorite solution. Antibacterial efficacy was evaluated quantitatively through bacterial culture and qualitatively through CLSM and SEM. Both SWEEPS and UAI demonstrated a statistically significant reduction in Enterococcus faecalis colony-forming units (CFUs) (p < 0.001), while SNI did not show a statistically significant reduction (p = 0.553). No significant difference was observed between the efficacy of SWEEPS and UAI (p > 0.05). The SWEEPS and UAI techniques were equally effective in eliminating mature E. faecalis biofilm.

Computer tomography (CT) scans' capabilities in detecting lesions have been increasing remarkably in the past decades. In this paper, we propose a multi-organ lesion detection (MOLD) approach to better address real-life chest-related clinical needs. MOLD is a challenging task, especially within a large, high resolution image volume, due to various types of background information interference and large differences in lesion sizes. Furthermore, the appearance similarity between lesions and other normal tissues demands more discriminative features. In order to overcome these challenges, we introduce depth-aware (DA) and skipped-layer hierarchical training (SHT) mechanisms with the novel Dense 3D context enhanced (Dense 3DCE) lesion detection model. The novel Dense 3DCE framework considers the shallow, medium, and deep-level features together comprehensively. In addition, equipped with our SHT scheme, the backpropagation process can now be supervised under precise control, while the DA scheme can effectively incorporate depth domain knowledge into the scheme. Extensive experiments have been carried out on a publicly available, widely used DeepLesion dataset, and the results prove the effectiveness of our DA-SHT Dense 3DCE network in the MOLD task.

Multimodal emotion recognition has emerged as a promising approach to capture the complex nature of human emotions by integrating information from various sources such as physiological signals, visual behavioral cues, and audio-visual content. However, current methods often struggle with effectively processing redundant or conflicting information across modalities and may overlook implicit inter-modal correlations. To address these challenges, this paper presents a novel multimodal emotion recognition framework which integrates audio-visual features with viewers' EEG data to enhance emotion classification accuracy. The proposed approach employs modality-specific encoders to extract spatiotemporal features, which are then aligned through contrastive learning to capture inter-modal relationships. Additionally, cross-modal attention mechanisms are incorporated for effective feature fusion across modalities. The framework, comprising pre-training, fine-tuning, and testing phases, is evaluated on multiple datasets of emotional responses. The experimental results demonstrate that the proposed multimodal approach, which combines audio-visual features with EEG data, is highly effective in recognizing emotions, highlighting its potential for advancing emotion recognition systems.

The mechanical properties of sclera play an important role in ocular functions, protection, and disease. Modulating the sclera's properties by exogenous crosslinking offers a way to expand the tissue's range of properties for study of the possible influences on the eye's behavior and diseases such as glaucoma and myopia. The focus of this work was to evaluate the effects of genipin crosslinking targeting the porcine perilimbal sclera (PLS) since the stiffness of this tissue was previously found in a number of studies to influence the eye's intraocular pressure (IOP). The work includes experiments on tensile test specimens and whole globes. The specimen tests showed decreased strain-rate dependence and increased relaxation stress due to the cross-linker. Whole globe perfusion experiments demonstrated that eyes treated with genipin in the perilimbal region had increased IOPs compared to the control globes. Migration of the cross-linker from the target tissue to other tissues is a confounding factor in whole globe, biomechanical measurements, with crosslinking. A novel quantitative genipin assay of the trabecular meshwork (TM) was developed to exclude globes where the TM was inadvertently crosslinked. The perfusion study, therefore, suggests that elevated stiffness of the PLS can significantly increase IOP apart from effects of the TM in the porcine eye. These results demonstrate the importance of PLS biomechanics in aqueous outflow regulation and support additional investigations into the distal outflow pathways as a key source of outflow resistance.

Maxillofacial defects, located in a region characterized by a complex interplay of soft and hard tissues, along with a sophisticated capillary and neural network, have long posed significant challenges in both clinical practice and research [...].

Cell nuclei interpretation is crucial in pathological diagnostics, especially in tumor specimens. A critical step in computational pathology is to detect and analyze individual nuclear properties using segmentation algorithms. Conventionally, a semantic segmentation network is used, where individual nuclear properties are derived after post-processing a segmentation mask. In this study, we focus on showing that an object-detection-based instance segmentation network, the Mask R-CNN, after integrating it with a Feature Pyramidal Network (FPN), gives mature and reliable results for nuclei detection without the need for additional post-processing. The results were analyzed using the Kumar dataset, a public dataset with over 20,000 nuclei annotations from various organs. The dice score of the baseline Mask R-CNN improved from 76% to 83% after integration with an FPN. This was comparable with the 82.6% dice score achieved by modern semantic-segmentation-based networks. Thus, evidence is provided that an end-to-end trainable detection-based instance segmentation algorithm with minimal post-processing steps can reliably be used for the detection and analysis of individual nuclear properties. This represents a relevant task for research and diagnosis in digital pathology, which can improve the automated analysis of histopathological images.

The detection of Cerebral Microbleeds (CMBs) is crucial for diagnosing cerebral small vessel disease. However, due to the small size and subtle appearance of CMBs in susceptibility-weighted imaging (SWI), manual detection is both time-consuming and labor-intensive. Meanwhile, the presence of similar-looking features in SWI images demands significant expertise from clinicians, further complicating this process. Recently, there has been a significant advancement in automated detection of CMBs using a Convolutional Neural Network (CNN) structure, aiming at enhancing diagnostic efficiency for neurologists. However, existing methods still show discrepancies when compared to the actual clinical diagnostic process. To bridge this gap, we introduce a novel multimodal detection and classification framework for CMBs' diagnosis, termed MM-UniCMBs. This framework includes a light-weight detection model and a multi-modal classification network. Specifically, we proposed a new CMBs detection network, CMBs-YOLO, designed to capture the salient features of CMBs in SWI images. Additionally, we design an innovative language-vision classification network, CMBsFormer (CF), which integrates patient textual descriptions-such as gender, age, and medical history-with image data. The MM-UniCMBs framework is designed to closely align with the diagnostic workflow of clinicians, offering greater interpretability and flexibility compared to existing methods. Extensive experimental results show that MM-UniCMBs achieves a sensitivity of 94% in CMBs' classification and can process a patient's data within 5 s.

Changes in biomechanical properties such as elasticity modulus, viscosity, and poroelastic features are linked to the health status of biological tissues. Ultrasound elastography is a non-invasive imaging tool that quantitatively maps these biomechanical characteristics for diagnostic and treatment monitoring purposes. Mathematical models are essential in ultrasound elastography as they convert the raw data obtained from tissue displacement caused by ultrasound waves into the images observed by clinicians. This article reviews the available mathematical frameworks of continuum mechanics for extracting the biomechanical characteristics of biological tissues in ultrasound elastography. Continuum-mechanics-based approaches such as classical viscoelasticity, elasticity, and poroelasticity models, as well as nonlocal continuum-based models, are described. The accuracy of ultrasound elastography can be increased with the recent advancements in continuum modelling techniques including hyperelasticity, biphasic theory, nonlocal viscoelasticity, inversion-based elasticity, and incorporating scale effects. However, the time taken to convert the data into clinical images increases with more complex models, and this is a major challenge for expanding the clinical utility of ultrasound elastography. As we strive to provide the most accurate imaging for patients, further research is needed to refine mathematical models for incorporation into the clinical workflow.

The topic of phonation onset gestures is of great interest to singers, acousticians, and voice physiologists alike. The vocal pedagogue and voice researcher Manuel Garcia, in the mid-19th century, first coined the term "coup de la glotte". Given that Garcia defined the process as "a precise articulation of the glottis that leads to a precise and clean tone attack", the term can certainly be linked to the concept of "vocal onset" as we understand it today. However, Garcia did not, by any means, have the technical measures at his disposal to investigate this phenomenon. In order to better understand modern ways of investigating vocal onset-and the limitations that still exist-it seems worthwhile to approach the subject from a historical perspective. High-speed video laryngoscopy (HSV) can be regarded as the gold standard among today's examination methods. Nonetheless, it still does not allow the three-dimensionality of vocal fold vibrations to be examined as it relates to vocal onset. Clearly, measuring methods in voice physiology have developed fundamentally since Garcia's time. This offers grounds for hope that the still unanswered questions around the phenomenon of vocal onset will be resolved in the near future. One promising approach could be to develop ultra-fast three-dimensional MRI further.