Computer Methods in Biomechanics and Biomedical Engineering最新文献

Cardiovascular disease (CVD) is the one of the most fatal diseases in the world we have seen in last two decades. For heart disease detection, imprecision in clinical parameters may occur due to error in taking readings or in measuring devices or environmental conditions etc. Hence, introducing fuzzy set theory in feature engineering may give better results as it deals with uncertainty. But in fuzzy set theory, only one uncertainty is considered, which is membership degree or degree of belongingness. Intuitionistic fuzzy set (IFS) considers two uncertainties - membership degree and non-membership degree and so IFS may provide efficient results. To reduce the risk of heart disease, an advanced deep learning algorithm will play a significant role in heart disease prediction that will help physicians to diagnose early. In this paper, we have established a transformation of patient features using i) intuitionistic fuzzy parameters, where Sugeno-type fuzzy complement is used and ii) fuzzy parameters, where gamma membership function is used. These transformed attributes are applied on Deep Learning prediction algorithm as Multi-layer Perceptron (MLP). The novelty of the paper lies from feature transformation to deep learning. It is observed that intuitionistic fuzzy transformation approach, keeping model parameters intact, significantly outperforms non-fuzzy method and gammy fuzzy Transformation, which is reflected in evaluation mechanisms.

The significant research carried out on medical healthcare networks is giving computing innovations lots of space to produce the most recent innovations. Pancreatic cancer, which ranks among of the most common tumors that are thought to be fatal and unsuspected since it is positioned in the region of the abdomen beyond the stomach and can't be adequately treated once diagnosed. In radiological imaging, such as MRI and CT, computer-aided diagnosis (CAD), quantitative evaluations, and automated pancreatic cancer classification approaches are routinely provided. This study provides a dynamic weighted ensemble framework for pancreatic cancer classification inspired by game theory. Grey Level Co-occurrence Matrix (GLCM) is utilized for feature extraction, together with Gaussian kernel-based fuzzy rough sets theory (GKFRST) for feature reduction and the Random Forest (RF) classifier for categorization. The ResNet50 and VGG16 are used in the transfer learning (TL) paradigm. The combination of the outcomes from the TL paradigm and the RF classifier paradigm is suggested using an innovative ensemble classifier that relies on the game theory method. When compared with the current models, the ensemble technique considerably increases the pancreatic cancer classification accuracy and yields exceptional performance. The study improves the categorization of pancreatic cancer by using game theory, a mathematical paradigm that simulates strategic interactions. Because game theory has been not frequently used in the discipline of cancer categorization, this research is distinctive in its methodology.

The conversion of a person's intentions into device commands through the use of brain-computer interface (BCI) is a feasible communication method for individuals with nervous system disorders. While common spatial pattern (CSP) is commonly used for feature extraction in BCIs, it has limitations. It is known for its susceptibility to noise and tendency to overfit. Moreover, high-dimensional, and irrelevant features can make it harder for a classifier to learn effectively. To address these challenges, exploring potential solutions is crucial. This paper introduces Regularized CSP with diagonal loading (DL-CSP) and Pearson correlation coefficient (PCC) based feature selection to extract the most discriminative motor imagery EEG (MI-EEG) features. Three classifiers in an ensemble are considered; bidirectional long short-term memory (Bi-LSTM), K-nearest neighbors (KNN) and naïve Bayes (NB). Decision level fusion through majority voting is exploited to leverage diverse perspectives and increase the overall system robustness. Experiments have been implemented using three publicly available datasets for MI classification; BCI competition IV-IIA (data-1), BCI Competition III-IVa (data-2), and a stroke patients' dataset (data-3). The accuracy achieved, according to the results, is 86.96% for data-1, 91.70% for data-2, and 85.75% for data-3. These percentages outperform the accuracy achieved by any state-of-the-art techniques.

As humans age, they experience deformity and a decrease in their bone strength, such brittleness in the bones ultimately lead to bone fracture. Magnetic field exposure combined with physical exercise may be useful in mitigating age-related bone loss by improving the canalicular fluid motion within the bone's lacuno-canalicular system (LCS). Nevertheless, an adequate amount of fluid induced shear stress is necessary for the bone mechano-transduction and solute transport in the case of brittle bone diseases. The underlying mechanisms of how magnetic fields, in combination to mechanical loading, affects the canalicular fluid motion still need to be explored. Accordingly, this study aims to develop a computer model to investigate the role of magnetic fields on loading-induced canalicular fluid flow in a curvy lacunar canalicular space with irregular osteocyte cell processes and walls. Moreover, this study considers canalicular fluid as non-Newtonian fluid, i.e., Jeffery fluid. In addition, a machine learning model was further employed for the estimation of parameters which significantly influence the canalicular fluid flow in response to loading and magnetic field. The results show that static magnetic field modulates the loading-induced canalicular fluid flow. Additionally, present study accelerates the fluid induced wall shear stress in case of osteoporosis.

Cardiac arrhythmias are major global health concern and their early detection is critical for diagnosis. This study comprehensively evaluates the effectiveness of CNNs and LSTMs for the classification of cardiac arrhythmias, considering three PhysioNet datasets. ECG records are segmented to accommodate around ∼10s of ECG data. Followed by transformation to scalograms using DWT for training VGG-16; and WTS for feature extraction and dimensionality reduction for training LSTM network. VGG-16 achieved 96.44% test accuracy while LSTM achieved 92%. Results also highlight the effectiveness of VGG-16 for short-duration ECG analysis, while LSTM excels in long-term monitoring on edge devices for personalized healthcare.

Slow eye movements (SEMs) are a reliable physiological marker of drivers' sleep onset, often accompanied by EEG alpha wave attenuation. A parallel multimodal 1D convolutional neural network (PM-1D-CNN) model is proposed to classify SEMs. The model uses two parallel 1D-CNN blocks to extract features from EOG and EEG signals, which are then fused and fed into fully connected layers for classification. Results show that the PM-1D-CNN outperforms the SGL-1D-CNN and Bimodal-LSTM networks in both subject-to-subject and cross-subject evaluations, confirming its effectiveness in detecting sleep onset.

Biomechanical properties of a mandibular first molar with different cavity designs [traditional access cavities (TEC-I & TEC-II), ninja access cavity (NEC), conservative access cavity (CEC), truss access cavity (Tr-EC), caries-driven access cavity (Cd-EC), caries-driven truss access cavity (Cd-TrEC)] were compared using finite element (FE) analysis. Models were subjected to three different loads. The highest stress distribution was observed on the enamel surface of the Cd-EC design and the dentin surface of the TEC-II. The stress was mainly concentrated on the lingual root surfaces and in the pericervical area. Enlarging the access cavity significantly increased stress distribution on enamel and dentin.

Breast cancer (BC) is a malignant tumor that occurs in breast tissue. This project aims to predict the prognosis of BC patients using genes related to hypoxia and endoplasmic reticulum stress (ERS). RNA-seq and clinical data for BC were downloaded from TCGA and GEO databases. Hypoxia and ERS-related genes were collected from the Genecards database. Univariate/multivariate Cox regression and Lasso regression analyses were used to screen genes and construct prognostic models. Patients were divided into high-risk (HR) and low-risk (LR) groups based on risk scores. The CIBERSORT algorithm was used to analyze differences in immune infiltration between the two groups. The mutations of the two groups were analyzed statistically. The CellMiner database was used for drug prediction and the TISCH database for single-cell sequencing analysis. We screened 8 feature genes to construct a prognostic model. Patients in the HR group had a remarkably worse prognosis. TP53 exhibited a higher mutation frequency in the HR group. CIBERSORT analysis uncovered a remarkable increase in the infiltration levels of Macrophages M0 and Tregs in cancer patients and HR patients. Drug sensitivity prediction demonstrated that the expression of IVL was greatly negatively linked with the sensitivity of COLCHICINE. PTGS2 had a remarkably negative correlation with the Vincristine sensitivity. The prognostic model based on 8 hypoxia and ERS-related genes can predict the survival, immune status, and potential drugs of BC patients, bringing a new perspective on individualized treatment.

This study introduces a novel anchor-type proximal femoral nail (AT-PFN) to improve the bone-fixation integrity over the standard screw-type nail (SST-PFN). Quasi-static incremental cyclic load test was performed to investigate load-displacement, cumulative deformation energy, time-strain, and backbone curves. The finite element analysis (FEA) was implemented to identify the stress and strain distributions. Additionally, non-destructive dynamic tests were conducted, and the measurements were processed using statistical pattern recognition, based on vector autoregression and principal component analysis to investigate the nonlinearity due to bone-fixation interface. The results demonstrate that the AT-PFN significantly improves the bone-fixation integrity compared to SST-PFN.

Repetitive mechanical stresses on the knee joint during daily activities accumulate fatigue damage in the articular cartilage (AC), leading to wear and knee osteoarthritis (KOA). Effective treatments remain limited, underscoring the need for predictive approaches to identify KOA early. This study proposes a mathematical model to estimate AC degradation under cyclic loading from walking. By integrating Miner's rule, Monte Carlo simulations, and Weibull distributions, the model predicts remaining cycles before AC loses 60% of its volume, a critical KOA threshold. Verified against a well-established model, it offers potential for personalized strategies to prevent or delay KOA progression.