Journal of Medical Engineering and Technology最新文献

In recent years, the science of ophthalmology has seen substantial developments, with an increasing demand for precise and comprehensive diagnosis. Traditional detection approaches sometimes fail to reflect the complicated interactions between diverse eye disorders. To tackle this issue, we present "Visionary Neural Networks," an innovative approach that uses the power of graph-based architectures to give a comprehensive solution for ocular disease analysis, such as cataract, diabetic retinopathy, glaucoma, and normal eyes. The proposed approach incorporates detailed features essential for disease identification, employing the intrinsic spatial relationships in retinal images. For each retinal image, the model builds a graph representation in which the pixels are the nodes and the edges are the spatial relationships between them. Through node-to-node communication and the use of graph-like neural networks in our design, the model dynamically learns context-aware features. In order to allow the architecture to carefully emphasise pertinent characteristics during feature transmission, the model is equipped to adapt the graph-based representation of the image. The proposed architecture outperforms traditional approaches, achieving an accuracy of over 76% and an AUC-ROC of 0.99. This research enhances medical diagnosis and patient care in ophthalmology by offering a meaningful methodological contribution to the field.

This study evaluated whether routine daily quality control (QC) tests can anticipate X-ray tube degradation in a digital mammography system. Over 30 months, tube loading, contrast-to-noise ratio (CNR), and image quality scores were extracted from daily QC tests and analysed alongside a documented tube failure in September 2024. These results were compared with annual tube output measurements and environmental data. A progressive rise in tube loading was observed prior to tube failure, increasing from 84.45 to 94.58 mAs (12.0%). The Mann-Kendall test confirmed a significant upward trend (τ = 0.461; p < 0.001), with Sen's slope indicating +0.19 mAs/month. Despite this, CNR values remained stable (mean = 14.76 ± 0.20), and image quality showed no degradation. However, tube output declined consistently (from 30.95 ± 3.08 to 28.76 ± 0.36 μGy/mAs), confirming reduced efficiency. Temperature and humidity remained within recommended limits. These findings suggest that gradual increases in tube loading can indicate early tube degradation, even when image quality remains unaffected. Beyond confirming equipment performance, daily QC may serve as a strategic tool to detect early signs of degradation, extend device lifespan, reduce unplanned downtime, and promote patient safety without added costs.

Laparoscopic surgery has become the standard for many surgical procedures globally, including laparoscopic Appendicectomy. However, access to adequate training, particularly in lower and middle-income countries (LMICs) remains challenging, impacting surgical proficiency and patient outcomes. This paper presents the development of a Laparoscopic Appendicectomy Simulation (LaApSi) model, a silicone-based, low-cost solution designed to enhance laparoscopic skills training, particularly in resource-constrained settings. The LaApSi model was developed by creating a negative mould based on anatomically accurate dimensions. High-grade silicone was used to replicate the texture of human tissue. The model integrates strategically placed assessment electronics with an Automated Progress Tracking System (APTS), offering real-time feedback on surgical manoeuvres and tracking trainee progress during simulations. The model is cost-effective, with initial production costs of £12.69 for the docking system and negative mould. This system can be used with multiple LaApSi models, each costing only £0.66 to produce. Equipped with the APTS, the LaApSi model provides a realistic and affordable option for simulation-based training in laparoscopic appendicectomy. Its cost-effectiveness and automated progress tracking make it a viable option for widespread adoption, particularly in LMICs. Additionally, the production methodology is adaptable to other laparoscopic procedures, presenting opportunities for broader applications in surgical education and training.

Diabetic retinopathy is a chronic and progressive eye disease in which the human retina is affected by an increase in the amount of insulin in the blood. Diabetic retinopathy, if not detected and treated in time, threatens the patient's vision and eventually causes complete blindness. Among various clinical symptoms, microaneurysm appears as the first sign of diabetic retinopathy. Accurate and reliable detection of microaneurysms is a challenging problem due to its small size and low contrast. The successful detection of microaneurysms will be more useful for the proper treatment of the disease in its early stages. In this paper, we present a method for classifying medical images of the retina to accurately detect the level of development of diabetic retinopathy. Our proposed method has six main steps. In steps one to four, the input image is pre-processed. In the first step; the detection and segmentation of blood vessels using the morphological closing operation is done. The second step; performs circular edge detection using gradient morphological operation. The third step; optical disc detection using the circular Hough transform edge detection method is done. The fourth step; the detection and segmentation of microaneurysms is done by removing blood vessels, circular edges, and optical discs and we use circular Hough transformation. In the fifth step, feature extraction is performed by considering two features, blood vessel area and microaneurysm area, and four features obtained from the gray level co-occurrence matrix. Finally, the sixth step is classification using the SVM classifier (Gaussian kernel function). We evaluated the performance of the model using EyePacs retinal fundus image database and obtained 95.20% and 97% accuracy and specificity, respectively. Experimental results show that our proposed model performs better in terms of evaluated measures compared to other methods.

Brain tumour detection and classification are critical for improving patient prognosis and treatment planning. However, manual identification from magnetic resonance imaging (MRI) scans is time-consuming, error-prone, and reliant on expert interpretation. The increasing complexity of tumour characteristics necessitates automated solutions to enhance accuracy and efficiency. This study introduces a novel ensemble deep learning model, boosted deformable and residual convolutional network with bi-directional convolutional long short-term memory (BDefRCNLSTM), for the classification and segmentation of brain tumours. The proposed framework integrates entropy-based local binary pattern (ELBP) for extracting spatial semantic features and employs the enhanced sooty tern optimisation (ESTO) algorithm for optimal feature selection. Additionally, an improved X-Net model is utilised for precise segmentation of tumour regions. The model is trained and evaluated on Figshare, Brain MRI, and Kaggle datasets using multiple performance metrics. Experimental results demonstrate that the proposed BDefRCNLSTM model achieves over 99% accuracy in both classification and segmentation, outperforming existing state-of-the-art approaches. The findings establish the proposed approach as a clinically viable solution for automated brain tumour diagnosis. The integration of optimised feature selection and advanced segmentation techniques improves diagnostic accuracy, potentially assisting radiologists in making faster and more reliable decisions.

The present study performs the fabrication and experimental performance analysis of a novel flexible keel prosthetic foot made up of functionally graded materials (FGMs). To obtain the desired mechanical properties, the keel of the prosthetic foot was fabricated by combining wood and carbon fibre-reinforced polymer (CFRP). The FGM in the foot was introduced by creating layers of a blend of burnt rice husk particles and silica rubber with different weight percentages (2%, 4%, 6%, 8%, 10%, and 12%). Moreover, a casting was performed to develop a keel prosthetic foot using a metal mould. Experimental tests were conducted to investigate the foot performance and compared with Solid Ankle Cushion Heel (SACH) foot designs by carrying out tests such as tensile test, hardness test, dorsiflexion analysis, and fatigue testing. Fatigue life and dorsiflexion angle values are found to be higher for the proposed design in comparison to the SACH foot. This study presents important insights into the potential benefits and practicality of using flexible keel prosthetic feet made from functionally graded materials. The enhanced mechanical characteristics and tailored design of these prosthetic feet can significantly improve user comfort, stability, and overall biomechanical performance.

The purpose of this paper is to describe the development of a low-cost insulin infusion pump software simulator. The simulator was built using Java programming language and replicates the interface and functions of a real low-cost insulin infusion pump currently under development. Potential users participated in a remote session, and assessment was conducted using a standard usability scale (SUS). With a sample size of 34 participants possessing different levels of knowledge regarding diabetes and infusion pumps, the mean SUS score obtained was 67.43. While the insulin infusion pump is a specialised device and its system may not be immediately familiar to all users, the results suggest that usability can improve with appropriate training and clinical support. Employing a simulation model during the development of the physical prototype may provide advantages for system design, safety, and effectiveness in health care technology delivery.

The purpose of this paper is to present a straightforward framework for Heart Rate (HR) estimation from a Phonocardiogram (PCG) record and study the association between murmur severity levels and HR. The system focuses primarily on data processing procedure, which is based on signal preprocessing using Maximal Overlap Discrete Wavelet Transform (MODWT) to delineate murmurs from heart sounds. We exploit the characteristics of Logistic function to derive an enhanced PCG envelope that serves as prerequisite for HR algorithm detection. In fact, the PCG envelope presents a cyclostationarity that can be easily detected throughout a cross-covariance autocorrelation function to calculate the Heart Rate (HR). In addition, the effect of minor and pronounced murmurs is gauged by the Energetic Ratio (ER) that provides a comprehensive idea about the superimposed murmur energy on first and second Heart sounds. The study was conducted on PASCAL dataset with 335 real clinical records. Results shows that mild murmurs with low ER are not associated with HR (p-value = 0.846). However, moderate-severe murmurs indicate a significant association with HR (p-value = 0.0002). Furthermore, HR decreases as Energetic Ratio increases. These findings could be valuable to medical professionals operating in the emergency departments.

The determination of pressure and contact area distributions in the coxofemoral joint during activities of daily living is essential to predict joint degeneration and prosthesis wear. This can also provide biomechanical justifications for preoperative planning and postoperative rehabilitation. To study the temporal evolution of pressure fields and contact areas in a person's coxofemoral joint during different activities, a parametric finite element model of the joint is developed. Eight activities of daily living are studied. Two different laws of cartilage behaviour are used: elastic and hyperelastic. The results obtained focused on a single subject are compared with those of other studies using classical hypotheses: no labrum, synovial fluid and bone deformability neglected, ideal spherical geometry of the articular surfaces and frictionless contact. The results show that activities related to sitting in and getting up from a chair are the least burdensome activities for the hip joint. Alternation between the bipodal station and the monopodal station is the most restrictive activity. For most activities, the highest pressures are in the anterolateral upper region of the femoral head and in the antero-superior region of the cotyloid. For the activities studied, considering the hyperelasticity of cartilage does not generate a significant difference compared to a simple elastic behaviour. The results are globally in agreement with numerical and analytical models using a spherical model of the joint and quantitatively enrich the knowledge of this field.