Biocybernetics and Biomedical Engineering最新文献

Background

Intraventricular hemorrhage (IVH) is a common and significant complication in premature infants. While cranial ultrasound is the golden standard for IVH detection, it may not identify lesions until hours or days after occurring, which limits early intervention. Predicting IVH in premature infants would be highly advantageous. Recent studies have shown that EEG data’s amplitude and frequency modulation features could offer predictive insights for neurological diseases in adults.

Methods

To investigate the association between IVH and EEG monitoring, a retrospective case-control study was conducted in preterm infants. All infants underwent amplitude integrated EEG monitoring for at least 3 days after birth. The study included 20 cases who had an IVH diagnosed on cranial ultrasound and had a negative ultrasound 24 h earlier, and 20 matched controls without IVH. Amplitude and frequency modulation features were extracted from single-channel EEG data, and various machine learning algorithms were evaluated to create a predictive model.

Results

Cases had an average gestational age and birth weight of 26.4 weeks and 965 g, respectively. The best-performing algorithm was adaptive boosting. EEG data from 24 h before IVH detection proved predictive with an area under the receiver operating characteristic curve of 93 %, an accuracy of 91 %, and a Kappa value of 0.85. The most informative features were the slow varying instantaneous frequency and amplitude in the Delta frequency band.

Conclusion

Amplitude and frequency modulation features obtained from single-channel EEG signals in extremely preterm infants show promise for predicting IVH occurrence within 24 h before detection on cranial ultrasound.

Chronic kidney disease (CKD) affects the morphological structure and causes significant degradation in kidney function, leading to renal replacement treatment in affected individuals. Vascular rarefaction is thought to be an important factor in accelerating kidney damage in CKD patients, therefore, the assessment of renal morphology and vasculature is crucial in nephrology. The objective of this study was to evaluate the morphological and vascular changes caused by CKD in mice kidneys. In this study, dual photoacoustic microscopy (PAM) and optical coherence microscopy (OCM) oriented wide-field high-resolution imaging modalities were employed for diseased renal imaging. The unilateral ureteral obstruction (UUO) model was used to prepare renal samples with CKD, and the developed wide-field dual imaging system was used to image both control and CKD-affected kidneys for assessing vascular and morphological changes during CKD progression. The obtained results reveal a gradual alteration in vascular intensity and pelvis space with the progress of UUO disease. Furthermore, a quantitative micro-vessel analysis was performed based on the node, junction, and mesh of the vessel, which provides details on the increasing microvascular-related characteristics in the peripheral area as the disease progresses. Thus, by concurrently employing the advantages of each optical imaging technique, the proposed method of assessing the OCM-based morphological and PAM-based vascular properties of the renal sample using a wide-field multimodal imaging system can be an efficient technique for whole volume analysis without any exogenous contrast agents in kidney histopathology.

Congestive heart failure (CHF) is a serious threat to human health. Electrocardiogram (ECG) signals have been proven to be useful in the detection of CHF. However, the low amplitude and short duration of the ECG signals, as well as the superimposed noise during the real-time acquisition of the signal, seriously affect the CHF detection. To improve the detection rate of CHF, this paper proposes a congestive heart failure detection method based on Gramian angular field (GAF) and two-dimensional symbolic phase permutation entropy (SPPE2D). The significant advantage of this method is that it reduces the sensitivity to noise, and good performance can be obtained without denoising using raw ECG signals. We segment the original ECG signals into 2 s non-overlapping segments and convert them into images using the GAF method. Then, the SPPE2D algorithm is proposed to measure the complexity between normal sinus rhythm (NSR) and CHF, and analyze the anti-noise performance of the algorithm. Finally, the SPPE2D features of GAF images are computed and input into a support vector machine (SVM) for CHF detection. Classification accuracy on the Massachusetts Institute of Technology − Beth Israel Hospital Normal Sinus Rhythm Database and Beth Israel Deaconess Medical Center Congestive Heart Failure Database is 99.59%, sensitivity is 99.42%, specificity is 99.80%, and F1-score is 99.62%. The accuracy of detecting CHF reach more than 97.75% in the other five CHF databases. The experimental results show that the method based on GAF and SPPE2D can effectively detect CHF by images of ECG signals and has good robustness. CHF can be detected using the 2 s sample lengths of ECG signals recording with high sensitivity, giving clinicians ample time to treat patients with CHF.

Attention Deficit Hyperactivity Disorder (ADHD) is a neurological condition, typically manifesting in childhood. Behavioral studies are used to treat the illness, but there is no conclusive way to diagnose it. To comprehend changes in the brain, electroencephalography (EEG) signals of ADHD patients are frequently examined. In the proposed study, we introduce EEG feature map (EEG-FM)-based image construction to input deep learning architectures for classifying ADHD. To demonstrate the effectiveness of the proposed method, EEG data of 15 ADHD patients and 18 control subjects are analyzed and detection performance is presented. EEG-FM-based images are obtained using both traditional time domain features used in EEG analysis, such as Hjorth parameters (activity, mobility, complexity), skewness, kurtosis, and peak-to-peak, and nonlinear features such as the largest Lyapunov Exponent, correlation dimension, Hurst exponent, Katz fractal dimension, Higuchi fractal dimension, and approximation entropy. EEG-FM-based images are used to train DarkNet19 architecture and deep features are extracted for each image dataset. Fewer deep features are chosen for each image dataset using the Minimum Redundancy Maximum Relevance (mRMR) feature selection method, and the concatenated deep feature set is created by merging the selected features. Finally, various machine learning methods are used to classify the concatenated deep features. Our EEG-FM and DarkNet19-based approach yields classification accuracies for ADHD between 96.6% and 99.9%. Experimental results indicate that the use of EEG-FM-based images as input to DarkNet19 architecture gives significant advantages in the detection of ADHD.

This study aimed to establish a blood vessel model within a hollow fiber bioreactor to evaluate the impact of high and fluctuating glucose levels on human umbilical vein endothelial cells (HUVECs) under laminar shear stress (LSS). HUVECs were cultured for 48 h in normal (5 mM), high (20 mM), and variable (20 mM / 5 mM alternating every 24 h) glucose concentrations under LSS of 0.66 Pa. An automated medium replacement system was developed. The control cultures remained static. The analysis included cell viability via cytometric analysis, glucose consumption, lactate production via electroenzymatic methods, and the expression of 21 genes via qPCR. The percentage of apoptotic cells did not significantly differ across glucose concentrations under LSS. HUVECs favor glycolysis for energy regardless of LSS. Under LSS, the IL1B, CCL2, and SELE genes were upregulated under high-glucose conditions and downregulated under variable-glucose conditions. A few other genes related to inflammation, oxidative stress, cell adhesion and apoptosis were upregulated under high-glucose conditions. In conclusion, using the blood vessel model we effectively examined the impact of glucose profiles on HUVECs under LSS in a device replicating the cylindrical geometry of blood vessels. LSS and tubular cell arrangement might mitigate the adverse effects of variable glucose on endothelial cells.

Significance

The biomechanical properties of the cornea are important for vision and ocular health. Optical coherence elastography (OCE) has the potential to improve our capacity to measure these properties.

Aim

This study tested a static compression OCE method utilising a commercially available optical coherence tomography (OCT) device, to estimate the Young’s modulus of ex-vivo porcine corneal tissue.

Approach: OCT was used to image corneal tissue samples before and during loading by static compression. The compressive force was measured with a piezoresistive force sensor, and tissue deformation was quantified through automated image analysis. Ten ex-vivo porcine corneas were assessed and the corneal thickness was also measured to assess the impact of corneal swelling.

Results

An average (standard deviation) Young’s modulus of 0.271 (+/- 0.091) MPa was determined across the 10 corneas assessed. There was a mean decrease of 1.78 % in corneal thickness at the end of the compression series. These results showed that there was a moderate association between corneal thickness and the Young’s modulus recording (R² = 0.274).

Conclusions

Optical coherence elastography utilising clinical instrumentation, can reliably characterise the mechanical properties of the cornea. These results support the further investigation of the technique for in-vivo measurement of the mechanical properties of the human cornea.

The aim of this study was to introduce and evaluate the baroreflex entropy index (BEI), a novel tool derived from standard lead II electrocardiograph (EKG) for autonomic function (AF) assessment in type 2 diabetes mellitus (T2DM). Researchers with distinct roles (analysis and data preparation) analyzed anonymized EKG data from healthy controls and two patient groups with T2DM (well controlled and poorly controlled). BEI was compared between groups, and correlations with glycemic markers (HbA1c, fasting glucose) were investigated. Logistic regression was used to assess the association between BEI and T2DM risk. BEI showed good repeatability and differentiation between groups. Notably, it required only single-lead EKG. BEI was inversely correlated with glycemic markers, suggesting improved baroreflex regulation with better glycemic control. BEI also outperformed small-scale multiscale entropy in group discrimination. Logistic regression identified BEI as a protective factor for T2DM. BEI represents a promising tool for monitoring AF, assessing glycemic control, and potentially stratifying T2DM risk. Further validation in larger longitudinal studies and an exploration of the applicability of BEI to other diseases are warranted.

“Y-shaped” airway stents have been widely used in the treatment of airway diseases, especially airway stenosis, due to their excellent flexibility. However, the current research on the flexibility of “Y-shaped” airway stents is still blank, limiting the possibility of improving the performance of stents in complex clinical disease. This study aimed to establish multi-dimensional evaluation of the flexibility of a novel segmented “Y-shaped” airway stent and two kinds of conventional stents. We evaluated the flexibility of the segmented stent, wholly knitted stent, and silicone stent by in vitro mechanical testing and finite element analysis methods. That is, the bending force and spring-back force of three kinds of stent were measured in left–right, anterior-posterior and longitudinal directions. The torque of the stents in torsion-recovery test of branches of stent was also executed. Finite element analysis was performed to evaluate the change of diameter. According to the detection, the bending force and spring-back force of the branch of the segmented stent during left–right and anterior-posterior compression, and the torque during torsion and recovery were lower than those of the other two stents. In finite element analysis, the diameter change of the segmented stent was minimal among the three stents. The flexibility of the segmented “Y-shaped” airway stent was better than that of the conventional “Y-shaped” airway stents, indicating that it has better adaptability and resistance to compression when implanted in the body.

Parkinson’s disease (PD) is a neurodegenerative disorder that influence brain’s neurological, behavioral, and physiological functions and includes motor and nonmotor manifestations. Although there have been several PD diagnosis systems with supervised machine learning techniques, there are more efforts that need to enhance the accurate detection of PD in its early stage. The current paper developed a novel approach by integrating Least Squares Support Vector Regression (LS-SVR) and Fuzzy Clustering for Unified Parkinson’s Disease Rating Scale (UPDRS) diagnosis. This paper used feature selection and Principal Component Analysis (PCA) to overcome the multicollinearity issues in data. This paper used a large medical dataset including Motor- and Total-UPDRS to demonstrate how the proposed method can improve prediction performance via extensive evaluations and comparisons with existing methods. Compared to other prediction methods, the experimental results demonstrate that the proposed method provided the best accuracy for Total-UPDRS (Root Mean Squared Error = 0.7348; R² = 0.9169) and Motor-UPDRS (Root Mean Squared Error = 0.8321; R² = 0.8756) predictions.

Convolutional neural networks (CNN) have been increasingly popular in image categorization in recent years. Hyperparameter optimization is a critical stage in enhancing the effectiveness of CNNs and achieving better results. Properly tuning hyperparameters allows the model to exhibit improved performance and facilitates faster learning. Misconfigured hyperparameters can prolong the training time or lead to the model not learning at all. Manually tuning hyperparameters is a time-consuming and challenging process. Automatically adjusting hyperparameters helps save time and resources. This study aims to propose an approach that shows higher classification performance than unoptimized convolutional neural network models, even at low epoch values, by automatically optimizing the hyperparameters of AlexNet and DarkNet19 with equilibrium optimization, the newest metaheuristic algorithm. In this respect, the proposed approach optimizes the number and size of filters in the first five convolutional layers in AlexNet and DarkNet19 using an equilibrium optimization algorithm. To evaluate the efficacy of the suggested method, experimental analyses were conducted on the pneumonia and COVID-19 datasets. An important advantage of this approach is its ability to accurately classify medical images. The testing process suggests that utilizing the proposed approach to optimize hyperparameters for AlexNet and DarkNet19 led to a 7% and 4.07% improvement, respectively, in image classification accuracy compared to non-optimized versions of the same networks. Furthermore, the approach displayed superior classification performance even in a few epochs compared to AlexNet, ShuffleNet, DarkNet19, GoogleNet, MobileNet-V2, VGG-16, VGG-19, ResNet18, and Inceptionv3. As a result, automatic tuning of the hyperparameters of AlexNet and DarkNet-19 with EO enabled the performance of these two models to increase significantly.