Computer methods and programs in biomedicine最新文献

{"title":"Autocorrelation maps for optimal setting in cardiac resynchronization therapy.","authors":"Anna Přibilová, Jana Švehlíková, Michal Šašov, Ján Zelinka, Beáta Ondrušová, Róbert Hatala, Milan Tyšler","doi":"10.1016/j.cmpb.2024.108519","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108519","url":null,"abstract":"<p><strong>Background and objective: </strong>Patients with chronic heart failure are treated with implanted devices artificially stimulating the ventricular myocardium to support the ventricular activation propagation dynamics. The criterion for stimulation/pacing timing is a shortening of the QRS duration in the ECG signal. The study suggests additional ECG parameters that could be helpful in cardiac resynchronization therapy (CRT) device pacing settings.</p><p><strong>Methods: </strong>This issue was approached by computing and evaluating autocorrelation maps derived from body surface potential maps during the QRS complex. The autocorrelation maps were calculated from the body surface potential maps of seventeen patients, fourteen of whom were diagnosed with the left bundle branch block (LBBB) and three with the right bundle branch block (RBBB). Eleven of the LBBB patients were responders, and all three RBBB patients and three LBBB patient were non-responders. The body surface potential maps were measured during their spontaneous heart rhythm and optimal CRT setting. The patients' autocorrelation maps were compared with the autocorrelation maps of a control group of 33 healthy persons using two-sample Kolmogorov-Smirnov and Wilcoxon rank-sum statistical tests.</p><p><strong>Results: </strong>The autocorrelation maps from spontaneous rhythm were significantly different (p < 0.00008) in healthy and LBBB groups, which was shown on 19 parameters extracted from the autocorrelation maps by both the statistical tests of equality. In the optimal CRT setting in the LBBB responders, four of the studied parameters (Shannon entropy of the histogram of the autocorrelation map's values, and mean, standard deviation, and geometrical mean of the autocorrelation map's positive values) were not significantly different from the parameters of the healthy subjects (p > 0.19).</p><p><strong>Conclusions: </strong>Selected parameters of autocorrelation maps can be used as additional parameters for optimal CRT pacing settings, leading to patients' positive responses to the treatment.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108519"},"PeriodicalIF":4.9,"publicationDate":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142778840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A model-based quantification of nonlinear expiratory resistance in Plethysmographic data of COPD patients.","authors":"Theodore Lerios, Jennifer L Knopp, Camilla Zilianti, Matteo Pecchiari, J Geoffrey Chase","doi":"10.1016/j.cmpb.2024.108520","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108520","url":null,"abstract":"<p><strong>Background: </strong>Chronic obstructive pulmonary disease (COPD) is characterised by airway obstruction with an increase in airway resistance (R) to airflow in the lungs. An extreme case of expiratory airway resistance is expiratory flow limitation, a common feature of severe COPD. Current analyses quantify expiratory R with linear model-based methods, which do not capture non-linearity's noted in COPD literature. This analysis utilises a simple nonlinear model to describe patient-specific nonlinear expiratory resistance dynamics typical of COPD and assesses its ability to both fit measured data and also to discriminate between severity levels of COPD.</p><p><strong>Methods: </strong>Plethysmographic data, including alveolar pressure and airway flow, was collected from n=100 subjects (40 healthy, 60 COPD) in a previous study. Healthy cohorts included Young (20-32 years) and Elderly (64-85 years) patients. COPD patients were divided into those with expiratory flow limitation (FL) and those without (NFL). Inspiratory R was treated as linear (R₁_,insp). Expiratory R was modelled with two separate models for a comparison: linear with constant resistance (R₁_,exp), and nonlinear time-varying resistance (R₂_,exp(t)) using b-splines.</p><p><strong>Results: </strong>Model fit to PQ loops show inspiration is typically linear. Linear R₁_,exp captured expiratory dynamics in healthy cohorts (RMSE 0.3 [0.2 - 0.4] cmH₂O), but did not capture nonlinearity in COPD patients. COPD cohorts showed PQ-loop ballooning during expiration, which was better captured by non-linear R₂_,exp(t) (RMSE 1.7[1.3-2.8] vs. 0.3[0.2-0.4] cmH₂O in FL patients). Airway resistance is higher in COPD than healthy cohorts (mean R_2,exp(t) for Young (1.9 [1.6-2.8]), Elderly (2.4 [1.4-3.5]), NFL (4.9 [3.9-6.6]) and FL (13.5 [10.4-21.9]) cmH₂O/L/s, with p ≤ 0.0001 between aggregated measures for Young and Elderly healthy subjects and NFL and FL COPD subjects). FL patients showed non-linear R₂_,exp(t) dynamics during flow deceleration, differentiating them from NFL COPD patients.</p><p><strong>Conclusions: </strong>Linear model metrics describe expiration dynamics well in healthy subjects, but fail to capture nonlinear dynamics in COPD patients. Overall, the model-based method presented shows promise in detecting expiratory flow limitation, as well as describing different dynamics in healthy, COPD, and FL COPD patients. This method may thus be clinically useful in the diagnosis or monitoring of COPD patients using Plethysmography data, without the need for additional expiratory flow limitation confirmation procedures.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108520"},"PeriodicalIF":4.9,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142791268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using machine learning models for predicting monthly iPTH levels in hemodialysis patients.","authors":"Chih-Chieh Hsieh, Chin-Wen Hsieh, Mohy Uddin, Li-Ping Hsu, Hao-Huan Hu, Shabbir Syed-Abdul","doi":"10.1016/j.cmpb.2024.108541","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108541","url":null,"abstract":"<p><strong>Background and objective: </strong>Intact parathyroid hormone (iPTH), also known as active parathyroid hormone, is an important indicator commonly for monitoring secondary hyperparathyroidism (SHPT) in patients undergoing hemodialysis. The aim of this study was to use machine learning (ML) models to predict monthly iPTH levels in patients undergoing hemodialysis.</p><p><strong>Methods: </strong>We conducted a retrospective study on patients undergoing regular hemodialysis. Patients' blood examinations data was collected from Taiwan Society of Nephrology - Kidney Dialysis, Transplantation (TSN-KiDiT) registration system, and patients' medications data was collected from Pingtung Christian Hospital (PTCH), Taiwan. We used five different ML models to classify patients into three distinct categories based on their iPTH levels: iPTH < 150, iPTH ≥ 150 & iPTH < 600, and iPTH ≥ 600(pg/ml).</p><p><strong>Results: </strong>We ultimately included 1,351 patients in our study and processed the data in four different ways. These methods varied based on the duration of the data (either using data from just one month or continuously over three months) and the number of features used (either all 52 features or only 20 most important features identified by SHapley Additive exPlanations (SHAP) analysis). The XGBoost model, using data from a continuous three-month period and all available features, yielded the best Weighted AUROC (0.922).</p><p><strong>Conclusions: </strong>ML is highly effective in predicting iPTH levels in hemodialysis patients, notably accurate for those with iPTH over 600 pg/ml. This method enables early identification of high-risk patients, reducing reliance on retrospective blood test assessments. Future research should focus on advancing explainable AI methods to foster clinicians' trust, and developing adaptable ML frameworks that could seamlessly integrate with existing healthcare systems.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108541"},"PeriodicalIF":4.9,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142784424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Laplacian-guided hierarchical transformer: A network for medical image segmentation.","authors":"Yuxiao Chen, Diwei Su, Jianxu Luo","doi":"10.1016/j.cmpb.2024.108526","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108526","url":null,"abstract":"<p><strong>Background and objective: </strong>Accurate medical image segmentation is crucial for diagnosis and treatment planning, particularly in tumor localization and organ measurement. Despite the success of Transformer models in various domains, they still struggle to capture high-frequency features, limiting their performance in medical image segmentation, especially in edge texture extraction. To overcome this limitation and improve segmentation accuracy, this study proposes a novel model architecture aimed at enhancing the Transformer's ability to capture and integrate both high-frequency and low-frequency features.</p><p><strong>Methods: </strong>Our model combines the extraction of high-frequency features using a Laplacian pyramid with the capture of low-frequency features through a Local-Global Feature Aggregation Module. A Feature Interaction Fusion module is employed to integrate these features, focusing on target areas. Additionally, a new bridging module facilitates the transfer of spatial information between the encoder and decoder via layer-wise attention mechanisms. The model's performance was evaluated using the Synapse dataset with statistical measures such as the Dice Similarity Coefficient and Hausdorff Distance. The code is available at https://github.com/chenyuxiao123/LGHF.</p><p><strong>Results: </strong>The proposed model demonstrated state-of-the-art performance in 2D medical image segmentation, achieving a Dice Similarity Coefficient of 84.10% and a Hausdorff Distance of 12.78. The evaluation metrics indicate significant improvements compared to existing methods.</p><p><strong>Conclusion: </strong>This novel model architecture, with its enhanced capability to capture and integrate both high-frequency and low-frequency features, shows significant potential for advancing medical image segmentation. The results on the Synapse dataset demonstrate its effectiveness and suggest its application could improve diagnosis and treatment planning in clinical settings.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108526"},"PeriodicalIF":4.9,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142789409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-resolution modeling of aqueous humor dynamics in the conventional outflow pathway of a normal human donor eye.","authors":"Alireza Karimi, Reza Razaghi, Ansel Stanik, Siddharth Daniel D'costa, Iman Mirafzal, Mary J Kelley, Ted S Acott, Haiyan Gong","doi":"10.1016/j.cmpb.2024.108538","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108538","url":null,"abstract":"<p><strong>Background and objective: </strong>The conventional aqueous outflow pathway, which includes the trabecular meshwork (TM), juxtacanalicular tissue (JCT), and inner wall endothelium of Schlemm's canal (SC) and its basement membrane, plays a significant role in regulating intraocular pressure (IOP) by controlling aqueous humor outflow resistance. Despite its significance, the biomechanical and hydrodynamic properties of this region remain inadequately understood. Fluid-structure interaction (FSI) and computational fluid dynamics (CFD) modeling using high-resolution microstructural images of the outflow pathway provides a comprehensive method to estimate these properties under varying conditions, offering valuable understandings beyond the capabilities of current imaging techniques.</p><p><strong>Methods: </strong>In this study, we utilized high-resolution 3D serial block-face scanning electron microscopy (SBF-SEM) to image the TM/JCT/SC complex of a normal human donor eye perfusion-fixed at an IOP of 7 mm Hg. We developed a detailed 3D finite element (FE) model of the pathway using SBF-SEM images to simulate the biomechanical environment. The model included the TM/JCT/SC complex (structure) with interspersed aqueous humor (fluid). We employed a 3D, inverse FE algorithm to calculate the unloaded geometry of the TM/JCT/SC complex and utilized FSI to simulate the pressurization of the complex from 0 to 15 mm Hg.</p><p><strong>Results: </strong>Our simulations revealed that the resultant velocity distribution in the aqueous humor across the TM/JCT/SC complex is heterogeneous. The JCT and its deepest regions, specifically the basement membrane of the inner wall of SC, exhibited a volumetric average velocity of ∼0.011 mm/s, which is higher than the TM regions, with a volumetric average velocity of ∼0.007 mm/s. Shear stress analysis indicated that the maximum shear stress, based on our FE code criteria, was 0.5 Pa starting from 10 µm into the TM from the anterior chamber and increased to 0.95 Pa in the JCT and its adjacent SC inner wall basement membrane. Also, the tensile stress and strain distributions showed significant variations, with the first principal stress reaching up to 57 Pa (compressive volumetric average) and the first principal strain reaching up to 3.5 % in areas of high mechanical loading. The resultant stresses, strains, and velocities exhibited relatively similar average values across the TM, JCT, and SC regions, primarily due to the uniform elastic moduli assigned to these components. Our computational fluid dynamics (CFD) analysis revealed that while the velocity of the aqueous humor remained consistent, the maximum shear stress was reduced by a factor of thirty.</p><p><strong>Conclusion: </strong>The uneven distribution of shear stress and velocity within the TM/JCT/SC complex highlights the complex biomechanical environment that regulates aqueous humor outflow.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108538"},"PeriodicalIF":4.9,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142791277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-time ultrasound shear wave elastography using a local phase gradient.","authors":"Enrique González-Mateo, Francisco Camarena, Noé Jiménez","doi":"10.1016/j.cmpb.2024.108529","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108529","url":null,"abstract":"<p><strong>Background and objective: </strong>Current approaches for ultrasound spectral elastography make use of block processing, resulting in long computational times. This work describes a real-time, robust, and quantitative imaging modality to map the elastic and viscoelastic properties of soft tissues using ultrasound.</p><p><strong>Methods: </strong>This elastographic technique relies on the spectral estimation of the shear-wave phase speed by combining a local phase-gradient method and angular filtering. We first apply directional filtering in the spatio-temporal frequency domain for providing one-way, smooth, and harmonic displacement maps in the frequency range of interest. Thanks to this, we can apply a simple, fast, and local phase gradient approach to obtain the axial and lateral components of the wavevector, which are linked to phase velocity and soft-tissue elasticity and viscoelasticity. The technique is validated numerically and experimentally using a 7.6 MHz ultrasound probe, tested in calibrated soft-tissue phantoms and ex vivo liver tissues. The method is compared with state-of-the-art spectral methods.</p><p><strong>Results: </strong>The technique significantly reduces the computation time, e.g., the reconstruction time for a 155 × 315-pixel phase-velocity map was 0.16 s, while local-phase velocity-imaging techniques was 156.73 s for 2D implementation and 13.56 s for the 1D version, a reduction between two and three orders of magnitude, while showing a similar accuracy and resolution than standard methods.</p><p><strong>Conclusions: </strong>This approach eliminates the need for block processing that may limit the spatial resolution and computational time of the velocity map. In this way, the phase gradient elastography method is revealed as an efficient and robust approach for real-time spectral elastography.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108529"},"PeriodicalIF":4.9,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142789411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Raw photoplethysmogram waveforms versus peak-to-peak intervals for machine learning detection of atrial fibrillation: Does waveform matter?","authors":"Jonas L Isaksen, Bolette Arildsen, Cathrine Lind, Malene Nørregaard, Kevin Vernooy, Ulrich Schotten, Thomas Jespersen, Konstanze Betz, Astrid N L Hermans, Jørgen K Kanters, Dominik Linz","doi":"10.1016/j.cmpb.2024.108537","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108537","url":null,"abstract":"<p><strong>Background: </strong>Machine learning-based analysis can accurately detect atrial fibrillation (AF) from photoplethysmograms (PPGs), however the computational requirements for analyzing raw PPG waveforms can be significant. The analysis of PPG-derived peak-to-peak intervals may offer a more feasible solution for smartphone deployment, provided the diagnostic utility is comparable.</p><p><strong>Aims: </strong>To compare raw PPG waveforms and PPG-derived peak-to-peak intervals as input signals for machine learning detection of AF.</p><p><strong>Methods: </strong>We developed specialized neural networks for raw waveform and peak-to-peak interval analyses and trained them on 7,704 PPGs from 106 patients from the TeleCheck-AF project. We evaluated the neural networks on 48,912 PPGs from 416 patients from the VIRTUAL-SAFARI project. We recorded computational requirements, sensitivity, positive predictive value (PPV), and F1 score.</p><p><strong>Results: </strong>With 1.6 million trainable parameters, the waveform model was more than 100 times as complex as the interval model (15,513 parameters) and required 19 times more computational power. In external validation, metrics were comparable between the interval and waveform models. For the interval model vs. the waveform model, sensitivity was 91.7 % vs. 81.9 % (p=0.4), PPV was 80.5 % vs. 84.5 % (p=0.3), and F1 score was 85.6 % vs. 81.3 % (p=0.5), respectively.</p><p><strong>Conclusion: </strong>PPG-derived peak-to-peak intervals and PPG waveforms were equivalent as input signals to neural networks in terms of accurate AF detection. The reduced computational requirements of the interval model make it a more suitable option for deployment on digital end-user devices such as smartphones.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108537"},"PeriodicalIF":4.9,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142791283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SIAM: Spatial and Intensity Awareness Module for cerebrovascular segmentation","authors":"Yunqing Chen ,&nbsp;Cheng Chen ,&nbsp;Xiaoheng Li ,&nbsp;Ruoxiu Xiao","doi":"10.1016/j.cmpb.2024.108511","DOIUrl":"10.1016/j.cmpb.2024.108511","url":null,"abstract":"<div><h3>Background and objectives:</h3><div>Cerebrovascular segmentation plays a crucial role in guiding the diagnosis and treatment of cerebrovascular diseases. With the rapid advancements in deep learning models, significant progress has been made in 3D cerebrovascular segmentation. However, they often rely on massive images and annotations, which is still challenging in cerebrovascular segmentation.</div></div><div><h3>Methods:</h3><div>Considering the unique pixel and spatial features inherent to vascular structures, such as vessel shape, location, and high pixel intensity characteristics, we propose a novel Spatial and Intensity Awareness Module (SIAM) for limited cerebrovascular segmentation. This module introduces spatial and pixel intensity perturbations to construct new matching data for model learning. Using collaborative training and shared features, SIAM gains the awareness of spatial and pixel intensity, thereby endowing the model with cerebrovascular semantics. Owing to the awareness learning belonging to an independent training module, SIAM satisfies the attribute of plug-and-play.</div></div><div><h3>Results:</h3><div>To validate SIAM, we carried out experiments on three cerebrovascular datasets with different modalities. The results demonstrate that SIAM enables the models to perform remarkably in normal and limited cerebrovascular segmentation. It can be seamlessly integrated into existing segmentation models without disrupting structural integrity.</div></div><div><h3>Conclusion:</h3><div>SIAM effectively learns and adapts to the unique spatial and pixel intensity features of vascular structures through collaborative training and shared features. Our experiments on three different cerebrovascular datasets confirm its robustness and efficacy even with limited data. Furthermore, its plug-and-play nature allows for seamless integration into existing models, preserving their structural integrity. Our code is available at <span><span>https://github.com/QingYunA/SIAM-Spatial-and-Intensity-Awareness-Module-for-3D-Cerebrovascular-Segmentation</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"Article 108511"},"PeriodicalIF":4.9,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142758957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dependence of acoustophoretic aggregation on the impedance of microchannel's walls.","authors":"Yiming Li, Dongfang Liang, Alexandre Kabla, Yuning Zhang, Jun Ma, Xin Yang","doi":"10.1016/j.cmpb.2024.108530","DOIUrl":"https://doi.org/10.1016/j.cmpb.2024.108530","url":null,"abstract":"<p><strong>Background and objectives: </strong>Acoustofluidic manipulation of particles and biological cells has been widely applied in various biomedical and engineering applications, including effective separation of cancer cell, point-of-care diagnosis, and cell patterning for tissue engineering. It is often implemented within a polydimethylsiloxane (PDMS) microchannel, where standing surface acoustic waves (SSAW) are generated by sending two counter-propagating ultrasonic waves on a piezoelectric substrate.</p><p><strong>Methods: </strong>In this paper, we develop a full cross-sectional model of the acoustofluidic device using finite element method, simulating the wave excitation on the substrate and wave propagation in both the fluid and the microchannel wall. This model allows us to carry out extensive parametric analyses concerning the acoustic properties of the fluid and the microchannel wall, as well as the dimensions of the channel, to explore their influences on the acoustic field, fluid flow and microparticle aggregation.</p><p><strong>Results: </strong>Our findings demonstrate an order-of-magnitude enhancement in acoustic pressure amplitude and aggregation speed and a reduction in the particle threshold radius to submicron levels, which can be achieved through adjustments to the channel height and the difference in acoustic impedance between the channel wall and the fluid. The optimum channel heights are determined, which depend on the acoustic properties of the channel wall. The particle trajectories, movements along pressure nodal planes, and terminal positions are identified, with relative strength between the radiation force and the streaming force compared in different combinations of parameters.</p><p><strong>Conclusions: </strong>This work demonstrates that finetuning the dimensions and acoustic properties of the fluid and microchannel wall in acoustofluidic device can greatly enhance particle aggregation throughput and reduce constraints on particle size. Our findings offer valuable insights into device design and optimization.</p>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"260 ","pages":"108530"},"PeriodicalIF":4.9,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142789408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing chest X-ray diagnostics: A novel CycleGAN-based preprocessing approach for enhanced lung disease classification in ChestX-Ray14","authors":"Aya Hage Chehade ,&nbsp;Nassib Abdallah ,&nbsp;Jean-Marie Marion ,&nbsp;Mathieu Hatt ,&nbsp;Mohamad Oueidat ,&nbsp;Pierre Chauvet","doi":"10.1016/j.cmpb.2024.108518","DOIUrl":"10.1016/j.cmpb.2024.108518","url":null,"abstract":"<div><h3>Background and objective:</h3><div>Chest radiography is a medical imaging technique widely used to diagnose thoracic diseases. However, X-ray images may contain artifacts such as irrelevant objects, medical devices, wires and electrodes that can introduce unnecessary noise, making difficult the distinction of relevant anatomical structures, and hindering accurate diagnoses. We aim in this study to address the issue of these artifacts in order to improve lung diseases classification results.</div></div><div><h3>Methods:</h3><div>In this paper we present a novel preprocessing approach which begins by detecting images that contain artifacts and then we reduce the artifacts’ noise effect by generating sharper images using a CycleGAN model. The DenseNet-121 model, used for the classification, incorporates channel and spatial attention mechanisms to specifically focus on relevant parts of the image. Additional information contained in the dataset, namely clinical characteristics, were also integrated into the model.</div></div><div><h3>Results:</h3><div>We evaluated the performance of the classification model before and after applying our proposed artifact preprocessing approach. These results clearly demonstrate that our preprocessing approach significantly improves the model’s AUC by 5.91% for pneumonia and 6.44% for consolidation classification, outperforming previous studies for the 14 diseases in the ChestX-Ray14 dataset.</div></div><div><h3>Conclusion:</h3><div>This research highlights the importance of considering the presence of artifacts when diagnosing lung diseases from radiographic images. By eliminating unwanted noise, our approach enables models to focus on relevant diagnostic features, thereby improving their performance. The results demonstrated that our approach is promising, highlighting its potential for broader applications in lung disease classification.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"259 ","pages":"Article 108518"},"PeriodicalIF":4.9,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}