Medical & Biological Engineering & Computing最新文献

Melanoma is an uncommon and dangerous type of skin cancer. Dermoscopic imaging aids skilled dermatologists in detection, yet the nuances between melanoma and non-melanoma conditions complicate diagnosis. Early identification of melanoma is vital for successful treatment, but manual diagnosis is time-consuming and requires a dermatologist with training. To overcome this issue, this article proposes an Optimized Attention-Induced Multihead Convolutional Neural Network with EfficientNetV2-fostered melanoma classification using dermoscopic images (AIMCNN-ENetV2-MC). The input pictures are extracted from the dermoscopic images dataset. Adaptive Distorted Gaussian Matched Filter (ADGMF) is used to remove the noise and maximize the superiority of skin dermoscopic images. These pre-processed images are fed to AIMCNN. The AIMCNN-ENetV2 classifies acral melanoma and benign nevus. Boosted Chimp Optimization Algorithm (BCOA) optimizes the AIMCNN-ENetV2 classifier for accurate classification. The proposed AIMCNN-ENetV2-MC is implemented using Python. The proposed approach attains an outstanding overall accuracy of 98.75%, less computation time of 98 s compared with the existing models.

Accurate recognition and linking of oncologic entities in clinical notes is essential for extracting insights across cancer research, patient care, clinical decision-making, and treatment optimization. We present the Neuro-Symbolic System for Cancer (NSSC), a hybrid AI framework that integrates neurosymbolic methods with named entity recognition (NER) and entity linking (EL) to transform unstructured clinical notes into structured terms using medical vocabularies, with the Unified Medical Language System (UMLS) as a case study. NSSC was evaluated on a dataset of clinical notes from breast cancer patients, demonstrating significant improvements in the accuracy of both entity recognition and linking compared to state-of-the-art models. Specifically, NSSC achieved a 33% improvement over BioFalcon and a 58% improvement over scispaCy. By combining large language models (LLMs) with symbolic reasoning, NSSC improves the recognition and interoperability of oncologic entities, enabling seamless integration with existing biomedical knowledge. This approach marks a significant advancement in extracting meaningful information from clinical narratives, offering promising applications in cancer research and personalized patient care.

Anesthetic-induced brain activity study is crucial in avian cognitive-, consciousness-, and sleep-related research. However, the neurobiological mechanisms underlying the generation of brain rhythms and specific connectivity of birds during anesthesia are poorly understood. Although different kinds of anesthetics can be used to induce an anesthesia state, a comparison study of these drugs focusing on the neural pattern evolution during anesthesia is lacking. Here, we recorded local field potentials (LFPs) using a multi-channel micro-electrode array inserted into the nidopallium caudolateral (NCL) of adult pigeons (Columba livia) anesthetized with chloral hydrate, pelltobarbitalum natricum or urethane. Power spectral density (PSD) and functional connectivity analyses were used to measure the dynamic temporal neural patterns in NCL during anesthesia. Neural decoding analysis was adopted to calculate the probability of the pigeon's brain state and the kind of injected anesthetic. In the NCL during anesthesia, we found elevated power activity and functional connectivity at low-frequency bands and depressed power activity and connectivity at high-frequency bands. Decoding results based on the spectral and functional connectivity features indicated that the pigeon's brain states during anesthesia and the injected anesthetics can be effectively decoded. These findings provide an important foundation for future investigations on how different anesthetics induce the generation of specific neural patterns.

Research at the mesoscale bone trabeculae arrangement yields intriguing results that, due to their clinical resolution, can be applied in clinical field, contributing significantly to the diagnosis of bone-related diseases. While the literature offers quantitative morphometric parameters for a thorough characterization of the mesoscale bone network, there is a gap in understanding relationships among them, particularly in the context of various bone pathologies. This research aims to bridge these gaps by offering a quantitative evaluation of the interplay among morphometric parameters and mechanical response at mesoscale in osteoporotic and non-osteoporotic bones. Bone mechanical response, dependent on trabecular arrangement, is defined by apparent stiffness, computationally calculated using the Gibson-Ashby model. Key findings indicate that: (i) in addition to bone density, measured using X-ray absorptiometry, trabecular connectivity density, trabecular spacing and degree of anisotropy are crucial parameters for characterize osteoporosis state; (ii) apparent stiffness values exhibit strong correlations with bone density and connectivity density; (iii) connectivity density and degree of anisotropy result the best predictors of mechanical response. Despite the inherent heterogeneity in bone structure, suggesting the potential benefit of a larger sample size in the future, this approach presents a valuable method to enhance discrimination between osteoporotic and non-osteoporotic samples.

Cognitive disturbance in identifying, processing, and responding to salient or novel stimuli are typical attributes of schizophrenia (SCH), and P300 has been proven to serve as a reliable psychosis endophenotype. The instability of neural processing across trials, i.e., trial-to-trial variability (TTV), is getting increasing attention in uncovering how the SCH "noisy" brain organizes during cognition processes. Nevertheless, the TTV in the brain network remains unrevealed, notably how it varies in different task stages. In this study, resorting to the time-varying directed electroencephalogram (EEG) network, we investigated the time-resolved TTV of the functional organizations subserving the evoking of P300. Results revealed anomalous TTV in time-varying networks across the delta, theta, alpha, beta1, and beta2 bands of SCH. The TTV of cross-band time-varying network properties can efficiently recognize SCH (accuracy: 83.39%, sensitivity: 89.22%, and specificity: 74.55%) and evaluate the psychiatric symptoms (i.e., Hamilton's depression scale-24, r = 0.430, p = 0.022, RMSE = 4.891; Hamilton's anxiety scale-14, r = 0.377, p = 0.048, RMSE = 4.575). Our study brings new insights into probing the time-resolved functional organization of the brain, and TTV in time-varying networks may provide a powerful tool for mining the substrates accounting for SCH and diagnostic evaluation of SCH.

The work elucidates the importance of accurate Parkinson's disease classification within medical diagnostics and introduces a novel framework for achieving this goal. Specifically, the study focuses on enhancing disease identification accuracy utilizing boosting methods. A standout contribution of this work lies in the utilization of a light gradient boosting machine (LGBM) coupled with hyperparameter tuning through grid search optimization (GSO) on the Parkinson's disease dataset derived from speech recording signals. In addition, the Synthetic Minority Over-sampling Technique (SMOTE) has also been employed as a pre-processing technique to balance the dataset, enhancing the robustness and reliability of the analysis. This approach is a novel addition to the study and underscores its potential to enhance disease identification accuracy. The datasets employed in this work include both gender-specific and combined cases, utilizing several distinctive feature subsets including baseline, Mel-frequency cepstral coefficients (MFCC), time-frequency, wavelet transform (WT), vocal fold, and tunable-Q-factor wavelet transform (TQWT). Comparative analyses against state-of-the-art boosting methods, such as AdaBoost and XG-Boost, reveal the superior performance of our proposed approach across diverse datasets and metrics. Notably, on the male cohort dataset, our method achieves exceptional results, demonstrating an accuracy of 0.98, precision of 1.00, sensitivity of 0.97, F1-Score of 0.98, and specificity of 1.00 when utilizing all features with GSO-LGBM. In comparison to AdaBoost and XGBoost, the proposed framework utilizing LGBM demonstrates superior accuracy, achieving an average improvement of 5% in classification accuracy across all feature subsets and datasets. These findings underscore the potential of the proposed methodology to enhance disease identification accuracy and provide valuable insights for further advancements in medical diagnostics.

While reaction-diffusion processes are utilized in multiple scientific fields, these phenomena have seen limited practical application in the polymer industry. Although self-regulating processes driven by parallel reaction and diffusion can lead to patterned structures, most polymeric products with repeating subunits are still prepared by methods that require complex and expensive instrumentation. A notable, high-added-value example is surgical mesh, which is often manufactured by weaving or knitting. In our present work, we demonstrate how the polymer and the biomedical industry can benefit from the pattern-forming capabilities of reaction-diffusion. We would like to propose a self-regulating method that facilitates the creation of surgical meshes from biocompatible polymers. Since the control of the process assumes a thorough understanding of the underlying phenomena, the theoretical background, as well as a mathematical model that can accurately describe the empirical data, is also introduced and explained. Our method offers the benefits of conventional techniques while introducing additional advantages not attainable with them. Most importantly, the method proposed in this paper enables the rapid creation of meshes with an average pore size that can be adjusted easily and tailored to fit the intended area of application.

Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their accountability and transparency level must be improved to transfer this success into clinical practice. The reliability of machine learning decisions must be explained and interpreted, especially for supporting the medical diagnosis. For this task, the deep learning techniques' black-box nature must somehow be lightened up to clarify its promising results. Hence, we aim to investigate the impact of the ResNet-50 deep convolutional design for Barrett's esophagus and adenocarcinoma classification. For such a task, and aiming at proposing a two-step learning technique, the output of each convolutional layer that composes the ResNet-50 architecture was trained and classified for further definition of layers that would provide more impact in the architecture. We showed that local information and high-dimensional features are essential to improve the classification for our task. Besides, we observed a significant improvement when the most discriminative layers expressed more impact in the training and classification of ResNet-50 for Barrett's esophagus and adenocarcinoma classification, demonstrating that both human knowledge and computational processing may influence the correct learning of such a problem.

Total hip replacement (THR) with cemented stem is a common procedure for patients with hip osteoarthritis. When primary THR fails, removal of the cement is problematic and poses challenges during revision surgeries. The possibility of proximal partial cementing of the hip stem was explored to mitigate the problem. 3D finite element analysis was performed to investigate the feasibility of reduced cement length for effective implant fixation and load transmission. Three levels of cement reduction (40 mm, 80 mm, and 100 mm) in the femoral stem were evaluated. All models were assigned loadings of peak forces acting on the femur during walking and stair climbing. The experimental and predicted max/min principal bone strains were fitted into regression models and showed good correlations. FE results indicated stress increment in the femoral bone, stem, and cement due to cement reduction. A notable increase of bone stress was observed with large cement reduction of 80-100 mm, particularly in Gruen zones 3 and 5 during walking and Gruen zones 3 and 6 during stair climbing. The increase of cement stresses could be limited to 11% with a cement reduction of 40 mm. The findings suggested that a 40-mm cement reduction in hip stem fixation was desirable to avoid unwanted complications after cemented THR.

Cardiac catheter ablation requires an adequate contact between myocardium and catheter tip. Our aim was to quantify the relationship between the contact force (CF) and the resulting mechanical deformation induced by the catheter tip using an ex vivo model and computational modeling. The catheter tip was inserted perpendicularly into porcine heart samples. CF values ranged from 10 to 80 g. The computer model was built to simulate the same experimental conditions, and it considered a 3-parameter Mooney-Rivlin model based on hyper-elastic material. We found a strong correlation between the CF and insertion depth (ID) (R² = 0.96, P < 0.001), from 0.7 ± 0.3 mm at 10 g to 6.9 ± 0.1 mm at 80 g. Since the surface deformation was asymmetrical, two transversal diameters (minor and major) were identified. Both diameters were strongly correlated with CF (R² ≥ 0.95), from 4.0 ± 0.4 mm at 20 g to 10.3 ± 0.0 mm at 80 g (minor), and from 6.4 ± 0.7 mm at 20 g to 16.7 ± 0.1 mm at 80 g (major). An optimal fit between computer and experimental results was achieved, with a prediction error of 0.74 and 0.86 mm for insertion depth and mean surface diameter, respectively.