Medical & Biological Engineering & Computing最新文献

{"title":"Inferring concussion history in athletes using pose and ground reaction force estimation and stability analysis of plyometric exercise videos.","authors":"William Alves, Athanasios Babouras, Paul A Martineau, Danielle Schutt, Shawn Robbins, Thomas Fevens","doi":"10.1007/s11517-025-03411-0","DOIUrl":"https://doi.org/10.1007/s11517-025-03411-0","url":null,"abstract":"<p><p>Concussions present a significant risk to athletes, with females exhibiting higher rates and prolonged recovery times than males. Current sideline concussion detection methods, such as the King-Devick test commonly used as a rapid screening tool designed to evaluate eye movement, attention, language, and cognitive processing abilities suffer from validity issues. This is especially true among young athletes highlighting the need for more accurate and objective assessment tools. This study investigates the ability of the Microsoft Kinect V2 pose estimation depth sensor to reliably measure subtle postural stability differences between athletes with a history of concussion and healthy controls. Traditional methods make use of expensive force plates which require trained personnel and controlled environments, limiting their use in resource-limited settings. Inspired by previous research utilizing force plates, our study analyzes video recordings of athletes performing specific exercises to detect dynamic balance deficits. A machine learning approach is employed to predict ground reaction forces from pose estimation video recordings, which are then analyzed to measure time to stabilization. Results reveal significant differences in movement mechanics between concussed and control groups, with the drop vertical jump (DVJ) exercise demonstrating the highest discriminatory power. Notably, concussed individuals exhibit longer time to stabilization (mean difference <math><mo>=</mo></math> 0.089 s, p = 0.046) during DVJ, indicating potential lingering balance impairments. While single-leg squat (SLS) and single-leg hop (SLH) exercises showed fewer discriminatory metrics than DVJ, they still provide valuable insights into balance capabilities. The DVJ yielded the largest statistical difference between injured and healthy male athletes, while the SLH was more effective for females and the SLS, while effective for ACL rehab progress assessment, was equally ineffective for both males and females.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144592706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Brain region localization: a rapid Parkinson's disease detection method based on EEG signals.","authors":"Mingliang Zhang, Hang Liu, Zhenghao Guo, Cui Wang, Timo Hamalainen, Fengyu Cong","doi":"10.1007/s11517-025-03388-w","DOIUrl":"https://doi.org/10.1007/s11517-025-03388-w","url":null,"abstract":"<p><p>Parkinson's disease (PD) is a prevalent neurodegenerative disorder worldwide, often progressing to mild cognitive impairment (MCI) and dementia. Clinical diagnosis of PD mainly depends on characteristic motor symptoms, which can lead to misdiagnosis, underscoring the need for reliable biomarkers. Early detection of PD and effective monitoring of disease progression are crucial for enhancing patient outcomes. Electroencephalogram (EEG) signals, as non-invasive neural recordings, show great promise as diagnostic biomarkers. In this study, we present a novel approach for PD diagnosis through the analysis of EEG signals from distinct brain regions. We used two publicly available EEG datasets and constructed three-dimensional (3D) time-frequency spectrograms for each brain region using the continuous wavelet transform (CWT). To improve feature representation, these spectrograms were encoded in the red-green-blue (RGB) color space. A ResNet18 model was trained separately on the spectrograms of each brain region, and its performance was assessed using the leave-one-subject-out cross-validation (LOSOCV) method. The proposed method achieved classification accuracies of 92.86% and 90.32% on the two datasets, respectively. The experimental results confirm the efficacy of our approach, highlighting its potential as a valuable tool to aid clinical diagnosis of PD.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144592756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-pseudo-sensor fusion approach to estimating the lower limb joint moments based on deep neural network.","authors":"Xisheng Yu, Zeguang Pei","doi":"10.1007/s11517-025-03406-x","DOIUrl":"https://doi.org/10.1007/s11517-025-03406-x","url":null,"abstract":"<p><p>Reliable feedback of gait variables, such as joint moments, is critical for designing controllers of intelligent assistive devices that can assist the wearer outdoors. To estimate lower extremity joint moments quickly and accurately outside the laboratory, a novel multimodal motion intent recognition system by fusing traditional deep learning models is proposed in this paper. The developed estimation method uses the joint kinematics data and individual feature parameters to estimate lower limb joint moments in the sagittal plane under different motion conditions: walking, running, and stair ascent and descent. Specifically, seven deep learning models that use combination of convolutional neural network, recurrent neural networks and attention mechanisms as the unit models of the framework are designed. To improve the performance of the unit models, a data augmentation module is designed in the system. Using those unit models, a novel framework, DeepMPSF-Net, which treats the output of each unit model as a pseudo-sensor observation and utilizes variable weight fusion methods to improve classification accuracy and kinetics estimation performance, is proposed. The results show that the augmented DeepMPSF-Net can accurately identify the locomotion, and the estimation performance (PCC) of joint moments is improved to 0.952 (walking), 0.988 (running), 0.925 (stair ascent), and 0.921 (stair descent), respectively. It also suggests that the estimation system is expected to contribute to the development of intelligent assistive devices for the lower limbs.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144592755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MTMedFormer: multi-task vision transformer for medical imaging with federated learning.","authors":"Anirban Nath, Sneha Shukla, Puneet Gupta","doi":"10.1007/s11517-025-03404-z","DOIUrl":"https://doi.org/10.1007/s11517-025-03404-z","url":null,"abstract":"<p><p>Deep learning has revolutionized medical imaging, improving tasks like image segmentation, detection, and classification, often surpassing human accuracy. However, the training of effective diagnostic models is hindered by two major challenges: the need for large datasets for each task and privacy laws restricting the sharing of medical data. Multi-task learning (MTL) addresses the first challenge by enabling a single model to perform multiple tasks, though convolution-based MTL models struggle with contextualizing global features. Federated learning (FL) helps overcome the second challenge by allowing models to train collaboratively without sharing data, but traditional methods struggle to aggregate stable feature maps due to the permutation-invariant nature of neural networks. To tackle these issues, we propose MTMedFormer, a transformer-based multi-task medical imaging model. We leverage the transformers' ability to learn task-agnostic features using a shared encoder and utilize task-specific decoders for robust feature extraction. By combining MTL with a hybrid loss function, MTMedFormer learns distinct diagnostic tasks in a synergistic manner. Additionally, we introduce a novel Bayesian federation method for aggregating multi-task imaging models. Our results show that MTMedFormer outperforms traditional single-task and MTL models on mammogram and pneumonia datasets, while our Bayesian federation method surpasses traditional methods in image segmentation.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144585496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Non-specific neck pain evaluation using functional linear models with the limma correction.","authors":"Elisa Aragón-Basanta, Guillermo Ayala, Álvaro Page, Pilar Serra-Añó","doi":"10.1007/s11517-025-03400-3","DOIUrl":"https://doi.org/10.1007/s11517-025-03400-3","url":null,"abstract":"<p><p>We have analyzed the relationship between disability and neck flexion-extension kinematics in non-specific neck pain subjects. A functional approach is used considering the angle, velocity, and acceleration curves. Different regression models have been fitted for each time in order to obtain these curves using scalar predictors such as the Neck Disability Index (NDI), age, sex, and neck length. In addition to classical regression, a limma (Linear Models for Microarray Data) model has been used, which improves the fit by modifying the estimation of the variances of the different fits using an empirical Bayes approach. As point-by-point adjustments are performed, this introduces a multiple comparison problem, and the corresponding p-values have to be adjusted in order to control the false discovery rate (FDR). In particular, a Benjamini-Hochberg method was used. The results show significant differences between raw and adjusted p-values for all variables, so spurious results were detected, e.g., the effect of neck length on velocity and acceleration curves. Differences between usual multiple linear regressions and the modified fits using the limma method (limma models) are minor, with a slight decrement of p-values in limma models. Once the p-values are adjusted, none of the variables analyzed significantly affects the angular curves. In contrast, NDI and age affect velocity and acceleration curves. Furthermore, the study of p-values throughout the movement shows that velocity and acceleration curves provide complementary information, so they should be used together in neck kinematics studies.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144585497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantifying features from X-ray images to assess early stage knee osteoarthritis.","authors":"Tanjina Helaly, Tanvir R Faisal, Ahmed Suparno Bahar Moni, Mahmuda Naznin","doi":"10.1007/s11517-025-03405-y","DOIUrl":"https://doi.org/10.1007/s11517-025-03405-y","url":null,"abstract":"<p><p>Knee osteoarthritis (KOA) is a progressive degenerative joint disease and a leading cause of disability worldwide. Manual diagnosis of KOA from X-ray images is subjective and prone to inter- and intra-observer variability, making early detection challenging. While deep learning (DL)-based models offer automation, they often require large labeled datasets, lack interpretability, and do not provide quantitative feature measurements. Our study presents an automated KOA severity assessment system that integrates a pretrained DL model with image processing techniques to extract and quantify key KOA imaging biomarkers. The pipeline includes contrast limited adaptive histogram equalization (CLAHE) for contrast enhancement, DexiNed-based edge extraction, and thresholding for noise reduction. We design customized algorithms that automatically detect and quantify joint space narrowing (JSN) and osteophytes from the extracted edges. The proposed model quantitatively assesses JSN and finds the number of intercondylar osteophytes, contributing to severity classification. The system achieves accuracies of 88% for JSN detection, 80% for osteophyte identification, and 73% for KOA classification. Its key strength lies in eliminating the need for any expensive training process and, consequently, the dependency on labeled data except for validation. Additionally, it provides quantitative data that can support classification in other OA grading frameworks.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144568052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Model-based analysis of sEMG signals using Stockwell transform features under varied muscle fiber composition and conduction velocity.","authors":"Venugopal G, Sidharth N, P A Karthick","doi":"10.1007/s11517-025-03403-0","DOIUrl":"https://doi.org/10.1007/s11517-025-03403-0","url":null,"abstract":"<p><p>In this study, sEMG signals of adductor pollicis (AP) and triceps brachii (TB) muscles that vary in fiber type proportion are generated at different levels of maximum voluntary contraction (MVC) by integrating various model components reported in existing studies. The current distribution function of the existing sEMG model is modified with time-varying action potential conduction velocity values for type I and II motor units of the muscles. To validate the model, sEMG signals were recorded from both muscles at 30%, 50%, and 70% of maximum voluntary contraction (MVC) until fatigue; AP using a pulley-rope setup and TB during isometric contractions with dumbbells. Stockwell transform (S transform) is used to compute the time-frequency (TF) spectrum of the initial and final 2 s segments of the signals. From the obtained singular values (SVs), features such as maximum SV, SV energy, and SV entropy are computed. The statistical analysis performed using the Mann-Whitney U test showed significant differences (p < 0.05) in the extracted features of AP and TB for most of the aspects. The Bland-Altman analysis demonstrated a high degree of agreement between simulated and experimental features, with the mean difference falling within the 95% confidence interval in most cases. The TF spectrum generated using the S transform shows a shift in frequency components towards lower frequencies during the final segment of simulated and recorded signals at the selected levels of MVCs. The proposed model helps to study the fiber-type characteristics of other skeletal muscles under different neuromuscular conditions.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144561783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Finite element stress analysis of the hindfoot after medial displacement calcaneal osteotomy with different translation distances.","authors":"Jinyang Lyu, Jian Xu, Jiazhang Huang, Chao Zhang, Xu Wang, Jian Yu, Xin Ma","doi":"10.1007/s11517-025-03309-x","DOIUrl":"10.1007/s11517-025-03309-x","url":null,"abstract":"<p><p>The medial displacement calcaneal osteotomy (MDCO) is one of commonly used procedures to restore the hindfoot alignment of the flatfoot deformity. However, the selection of the amount of translation for MDCO and its biomechanical effect on the hindfoot was rarely reported. This study employs finite element analysis to investigate stress distribution in the hindfoot following MDCO across varying translation distances. An adult-acquired flatfoot deformity (AAFD) finite element (FE) model consisting of 16 bones, 56 ligaments, and soft tissues was used. MDCO procedure was simulated with the translation distance of 0 mm, 2 mm, 4 mm, 6 mm, 8 mm, 10 mm, 12 mm, and 14 mm. Contact pressure on the plantar surface, the articular surface of the tibiotalar joint and the subtalar joint, and von Mises stress on the resection surface of the calcaneus under different translation distances were analyzed and compared. Results showed the MDCO reduces 12.46 to 33.32% peak contact pressure on the plantar surface, the tibiotalar joint, and the posterior facet of the subtalar joint, and shifts pressure from lateral to medial. But the difference in peak pressure for different translation distances larger than 4 mm was small. The MDCO also reduces the stress on the distal calcaneal resected surface. The study highlights the use of patient-specific computational modeling for preoperative plans.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"1955-1964"},"PeriodicalIF":2.6,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143081553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Masked autoencoders with generalizable self-distillation for skin lesion segmentation.","authors":"Yichen Zhi, Hongxia Bie, Jiali Wang, Lihan Ren","doi":"10.1007/s11517-024-03086-z","DOIUrl":"10.1007/s11517-024-03086-z","url":null,"abstract":"<p><p>In the field of skin lesion image segmentation, accurate identification and partitioning of diseased regions is of vital importance for in-depth analysis of skin cancer. Self-supervised learning, i.e., MAE, has emerged as a potent force in the medical imaging domain, which autonomously learns and extracts latent features from unlabeled data, thereby yielding pre-trained models that greatly assist downstream tasks. To encourage pre-trained models to more comprehensively learn the global structural and local detail information inherent in dermoscopy images, we introduce a Teacher-Student architecture, named TEDMAE, by incorporating a self-distillation mechanism, it learns holistic image feature information to improve the generalizable global knowledge learning of the student MAE model. To make the image features learned by the model suitable for unknown test images, two optimization strategies are, Exterior Conversion Augmentation (EC) utilizes random convolutional kernels and linear interpolation to effectively transform the input image into one with the same shape but altered intensities and textures, while Dynamic Feature Generation (DF) employs a nonlinear attention mechanism for feature merging, enhancing the expressive power of the features, are proposed to enhance the generalizability of global features learned by the teacher model, thereby improving the overall generalization capability of the pre-trained models. Experimental results from the three public skin disease datasets, ISIC2019, ISIC2017, and PH <math><msup><mrow></mrow> <mn>2</mn></msup> </math> indicate that our proposed TEDMAE method outperforms several similar approaches. Specifically, TEDMAE demonstrated optimal segmentation and generalization performance on the ISIC2017 and PH <math><msup><mrow></mrow> <mn>2</mn></msup> </math> datasets, with Dice scores reaching 82.1% and 91.2%, respectively. The best Jaccard values were 72.6% and 84.5%, while the optimal HD95% values were 13.0% and 8.9%, respectively.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"1915-1929"},"PeriodicalIF":2.6,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140873290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using principal component analysis to determine which vestibular stimuli provide best biomarkers for separating Alzheimer's from mixed Alzheimer's disease.","authors":"S Marzban, Z Dastgheib, B Lithgow, Z Moussavi","doi":"10.1007/s11517-024-03110-2","DOIUrl":"10.1007/s11517-024-03110-2","url":null,"abstract":"<p><p>Alzheimer's disease (AD) is often mixed with cerebrovascular disease (AD-CVD). Heterogeneity of dementia etiology and the overlapping of neuropathological features of AD and AD-CVD make feature identification of the two challenging. Separation of AD from AD-CVD is important as the optimized treatment for each group may differ. Recent studies using vestibular responses recorded from electrovestibulography (EVestG™) have offered promising results for separating these two pathologies. An EVestG measurement records responses to several different physical stimuli (called tilts). In previous research, the number of EVestG features from different tilts was selected based on physiological intuition to classify AD from AD-CVD. As the number of potential characteristic features from all tilts can be very large, in this study, we used an algorithm based on principal component analysis (PCA) to rank the most effective vestibular stimuli for differentiating AD from AD-CVD. Analyses were performed on the EVestG signals of 28 individuals with AD and 24 with AD-CVD. The results of this study showed that tilts simulating the otolithic organs (utricle and saccule) generated the most characteristic features for separating AD from AD-CVD.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"1943-1953"},"PeriodicalIF":2.6,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140913154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}