Biomedical Signal Processing and Control最新文献

{"title":"White patchy skin lesion classification using feature enhancement and interaction transformer module","authors":"Zhiming Li ,&nbsp;Shuying Jiang ,&nbsp;Fan Xiang ,&nbsp;Chunying Li ,&nbsp;Shuli Li ,&nbsp;Tianwen Gao ,&nbsp;Kaiqiao He ,&nbsp;Jianru Chen ,&nbsp;Junpeng Zhang ,&nbsp;Junran Zhang","doi":"10.1016/j.bspc.2025.107819","DOIUrl":"10.1016/j.bspc.2025.107819","url":null,"abstract":"<div><div>White patchy skin lesions have always been difficult to distinguish, yet precise identification of specific types can enable targeted treatment and alleviate patient anxiety. Deep convolutional neural networks (DCNNs) show great potential in this regard. However, DCNNs still exhibit limitations such as incapacity to capture global correlations, inability to discern invisible data distributions, and difficulty in handling imbalanced datasets. To address these challenges, we propose a feature enhancement and interaction transformer module for accurately identifying common white patches. Our approach begins with the development of a dual-position encoding attention and convolution hybrid submodule, which aims to model the global information of the feature domain and enhance feature representation. Subsequently, we construct a feature interaction submodule on the batch dimension to enable the DCNN to explore sample relationships, learn about invisible distribution from the dataset, and reduce the imbalance problem. Based on the dataset comprising four types of white patchy skin lesions, the proposed approach achieved an accuracy, precision, recall, F1 score, and AUC of 92.65 %, 92.83 %, 92.65 %, 92.74 %, and 0.98, respectively. These results demonstrate the superior performance of our approach compared to other state-of-the-art models, underscoring its potential to enhance the classification of white patchy skin lesions and expand their clinical applications without compromising the integrity of the DCNN structure.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"107 ","pages":"Article 107819"},"PeriodicalIF":4.9,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143601260","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":"Multi-task coordinate attention gating network for speech emotion recognition under noisy circumstances","authors":"Linhui Sun,&nbsp;Yunlong Lei,&nbsp;Zixiao Zhang,&nbsp;Yi Tang,&nbsp;Jing Wang,&nbsp;Lei Ye,&nbsp;Pingan Li","doi":"10.1016/j.bspc.2025.107811","DOIUrl":"10.1016/j.bspc.2025.107811","url":null,"abstract":"<div><div>Speech emotion recognition (SER) has recently made great progress in ideal environments, but their performance deteriorates dramatically when applied in complex real-world environments, mainly due to poor model robustness and generalization ability. To this end, we propose a multi-task coordinate attention gated network (MTCAGN) framework. For the SER main task, we propose a multi-scale gated convolutional neural network model with the coordinate attention mechanism, which captures a wide range of emotional features at different scales and the key global information, accurately focusing on salient emotional features in speech signals. Speech enhancement is used as an auxiliary task during the training phase, and the overall robustness of the system is strengthened through shared representation learning, allowing it to withstand complex interferences in noisy scenarios. In the inference phase, the speech enhancement branch is removed and only the SER task is retained. Therefore, our proposed method improves the robustness of the SER system without increasing inference complexity. To simulate the noise scenario, we construct three noisy speech datasets by randomly mixing clean audio from IEMOCAP or EMODB dataset with noise from the MUSAN dataset. The empirical findings evince that our proposed model exhibits superior performance in challenging low signal-to-noise ratio environments compared to the present state-of-the-art techniques, as indicated by weighted and unweighted accuracy metrics.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"107 ","pages":"Article 107811"},"PeriodicalIF":4.9,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592441","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":"Optimization of blood glucose prediction with LSTM-XGBoost fusion and integration of statistical features for enhanced accuracy","authors":"Loubna Mazgouti ,&nbsp;Nacira Laamiri ,&nbsp;Jaouher Ben Ali ,&nbsp;Najiba EL Amrani El Idrissi ,&nbsp;Véronique Di Costanzo ,&nbsp;Roomila Naeck ,&nbsp;Jean-Mark Ginoux","doi":"10.1016/j.bspc.2025.107814","DOIUrl":"10.1016/j.bspc.2025.107814","url":null,"abstract":"<div><div>Technological developments, most notably Continuous Glucose Monitoring (CGM) devices, have made it possible to manage diabetes mellitus by accessing trustworthy data, a chronic disease requiring constant monitoring of Blood Glucose (BG) levels to keep them as close to normal as possible, aiming to prevent potentially serious complications. Technological advancements have fuelled research in artificial intelligence to develop accurate methods for predicting future BG levels. These initiatives, supported by numerous studies, aim to improve the quality of life of people with diabetes by anticipating and avoiding dangerous fluctuations in BG levels.</div><div>Diabetes, a long-term health condition, requires checking BG levels to maintain them within a safe range and reduce the risk of severe complications. Modern advancements in technology, such as CGM systems, have made it easier to obtain data. This has led to increased interest in intelligence research to create methods for predicting future BG levels. These efforts, backed by studies, strive to improve the well-being of individuals with diabetes by predicting and preventing fluctuations in blood sugar levels.</div><div>In this study, our approach focuses on predicting future BG levels in diabetic patients using only CGM data. This approach stands out from many existing methods that require details about patients’ daily activities, such as meals, insulin therapy, and emotional factors. To this end, we explored different approaches for predicting glucose levels in diabetic patients, particularly emphasizing the use of the Long Short-Term Memory (LSTM) model. We examined various scenarios, including LSTM alone, LSTM with the integration of temporal features, and the combination of LSTM with the XGBoost model, incorporating additional features as inputs for the LSTM model.</div><div>Experimental results show that the proposed method performs exceptionally well in terms of Root Mean Square Error (RMSE) across both short and long terms. Compared to previous works, the proposed method is considered as effective and accurate even when considering long horizon. Considering horizons of 15, 30, 45, and 60 min, the combination of LSTM and XGBoost, with the integration of additional features for the LSTM model as input, achieved the lowest average values of RMSE, namely respectively 7.97 mg/dl, 9.63 mg/dl, 10.72 mg/dl, and 10.93 mg/dl. This result was validated using CGM data of 12 patients. This approach distinguishes itself by its ability to provide accurate results compared to other methods, emphasizing its potential in improving the management of diabetes mellitus.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"107 ","pages":"Article 107814"},"PeriodicalIF":4.9,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592552","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":"Challenges and artificial intelligence solutions for clinically optimal hepatic venous vessel segmentation","authors":"Håvard Bjørke Jenssen ,&nbsp;Varatharajan Nainamalai ,&nbsp;Egidijus Pelanis ,&nbsp;Rahul P. Kumar ,&nbsp;Andreas Abildgaard ,&nbsp;Finn Kristian Kolrud ,&nbsp;Bjørn Edwin ,&nbsp;Jingfeng Jiang ,&nbsp;Joseph Vettukattil ,&nbsp;Ole Jakob Elle ,&nbsp;Å smund Avdem Fretland","doi":"10.1016/j.bspc.2025.107822","DOIUrl":"10.1016/j.bspc.2025.107822","url":null,"abstract":"<div><h3>Background</h3><div>: Liver vessel identification is crucial for clinical disease assessment and treatment planning, especially concerning local treatment of liver tumors. As artificial intelligence (AI) develops in radiology, opportunities arise to craft models adept at hepatic venous vessel segmentation, opening possibilities for creating patient-specific models of the liver anatomy quickly, despite the diverse features of CT images encountered in clinical settings.</div></div><div><h3>Objective:</h3><div>This research evaluates the performance of AI models combined with various pre-processing filters for liver vessel segmentation, emphasizing clinically relevant results. A novel evaluation method was introduced to offer more anatomically accurate assessments, moving beyond traditional metrics like the Dice score.</div></div><div><h3>Methods:</h3><div>Using open-source and proprietary datasets, we implemented residual UNet and Dense UNet in combination with smoothness and vesselness filters. We used a clinical evaluation approach focused on major and minor liver vessels, thereby underscoring the precision of AI outcomes.</div></div><div><h3>Results:</h3><div>The Dense UNet model with a specific pre-processing filter produced an average Dice score of 0.8144 in our internal dataset. For the public test dataset, the score was 0.7859. Both scores were higher than those not using pre-processing filters, 0.8052 and 0.7765. Clinical assessments showed 85% of AI predictions accurately identified all wanted vessel structures, though segmentation beyond the vessel borders did occur in half the predictions.</div></div><div><h3>Conclusion:</h3><div>This study highlights the effectiveness of AI in liver vessel segmentation, with the Dense UNet model combined with pre-processing filters showing high Dice scores and clinical accuracy.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107822"},"PeriodicalIF":4.9,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592297","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}

{"title":"LiteNeXt: A novel lightweight ConvMixer-based model with Self-embedding Representation Parallel for medical image segmentation","authors":"Ngoc-Du Tran,&nbsp;Thi-Thao Tran,&nbsp;Quang-Huy Nguyen,&nbsp;Manh-Hung Vu,&nbsp;Van-Truong Pham","doi":"10.1016/j.bspc.2025.107773","DOIUrl":"10.1016/j.bspc.2025.107773","url":null,"abstract":"<div><div>The emergence of deep learning techniques has advanced the image segmentation task, especially for medical images. Many neural network models have been introduced in the last decade bringing the automated segmentation accuracy close to manual segmentation. However, cutting-edge models like Transformer-based architectures rely on large scale annotated training data, and are generally designed with densely consecutive layers in the encoder, decoder, and skip connections resulting in large number of parameters. Additionally, for better performance, they often be pretrained on a larger data, thus requiring large memory size and increasing resource expenses. In this study, we propose a new lightweight but efficient model, namely LiteNeXt, based on convolutions and mixing modules with simplified decoder, for medical image segmentation. The model is trained from scratch with small amount of parameters (0.71M) and Giga Floating Point Operations Per Second (0.42). To handle boundary fuzzy as well as occlusion or clutter in objects especially in medical image regions, we propose the Marginal Weight Loss that can help effectively determine the marginal boundary between object and background. Additionally, the Self-embedding Representation Parallel technique is proposed as an innovative data augmentation strategy that utilizes the network architecture itself for self-learning augmentation, enhancing feature extraction robustness without external data. Experiments on public datasets including Data Science Bowls, GlaS, ISIC2018, PH2, Sunnybrook, and Lung X-ray data show promising results compared to other state-of-the-art CNN-based and Transformer-based architectures. Our code is released at: <span><span>https://github.com/tranngocduvnvp/LiteNeXt</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"107 ","pages":"Article 107773"},"PeriodicalIF":4.9,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592553","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":"REBSA: Enhanced backtracking search for multi-threshold segmentation of breast cancer images","authors":"Shiqi Xu ,&nbsp;Wei Jiang ,&nbsp;Yi Chen ,&nbsp;Ali Asghar Heidari ,&nbsp;Lei Liu ,&nbsp;Huiling Chen ,&nbsp;Guoxi Liang","doi":"10.1016/j.bspc.2025.107733","DOIUrl":"10.1016/j.bspc.2025.107733","url":null,"abstract":"<div><div>Breast cancer has become one of the most common cancers among women globally. Early diagnosis and intervention play a crucial role in breast cancer management. Automatic segmentation of histological images of breast cancer utilizing Multi-Threshold Image Segmentation (MTIS) technology can assist doctors in making more accurate diagnostic decisions for patients. However, traditional methods face challenges in terms of segmentation efficiency and accuracy. This paper proposes a Renyi entropy-based MTIS to address this issue using an improved backtracking search algorithm (REBSA). The proposed method enhances the original BSA by introducing a random reselection strategy to enhance diversity of the population and enhance the algorithm’s exploration capability. Additionally, an enhanced quality mechanism is incorporated, which improves the quality of candidate solutions while maintaining a degree of randomness. The integration of these two approaches significantly enhances the performance of the BSA. In order to confirm the performance of the proposed REBSA, several tests were carried out using the CEC 2017 benchmark functions, including diversity balance analysis, parameter sensitivity analysis, and stability analysis. Additionally, REBSA was compared with various basic and advanced algorithms. The results demonstrate that REBSA achieved the top rank on most functions across different dimensions, proving its exceptional optimization performance and robustness. Finally, the proposed REBSA was applied to MTIS tasks on breast cancer histopathological images. The results verified that REBSA achieved higher segmentation accuracy and efficiency. Compared to other approaches, it can retain more pathological tissue details and rank higher than other methods in several image evaluation metrics, demonstrating its ability to handle the difficult problem of breast cancer tissue image segmentation. Moreover, this study utilized a real clinical dataset of breast cancer histopathological images, further demonstrating the suggested method’s efficacy in practical diagnostic scenarios. It provides reliable technical support for medical image analysis, assisting doctors in improving diagnostic accuracy and early screening efficiency.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107733"},"PeriodicalIF":4.9,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592296","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":"Attention-enhanced U-Net based network for cancerous tissue segmentation","authors":"Yuchen Wang ,&nbsp;Kainan Ma ,&nbsp;Yang Li ,&nbsp;Limin Cao ,&nbsp;Zhaoyuxuan Wang ,&nbsp;Yiheng Zhou ,&nbsp;Qian Sun ,&nbsp;Chaoxing You ,&nbsp;Shuang Xia ,&nbsp;Ming Liu","doi":"10.1016/j.bspc.2025.107728","DOIUrl":"10.1016/j.bspc.2025.107728","url":null,"abstract":"<div><div>Cancerous tissue segmentation is a key step in further refining the identification of cancer cell aggregation regions after cell segmentation, which is crucial for early diagnosis, precise staging and personalized treatment strategies for cancer. However, there are relatively few researchers in this field, highlighting the need for further exploration. This paper has proposed an automated cancer segmentation method based on Attention-Enhanced U-Net, fusing two key features, color and density. The method is mainly divided into two steps: cell segmentation and cancer segmentation. On both Multi-Organ Nuclei Segmentation and Triple-Negative Breast Cancer datasets, the cell segmentation results achieved 73.91% and 77.51% F1-Score, Mean Intersection over Union scores of 63.55% and 67.30%, and Dice Similarity Coefficient of 72.41% and 77.52%, and which are better than other deep learning models. We also tested cancer segmentation on images from the pathology library, achieving a Dice Similarity Coefficient of 87.54%, which is also better than end-to-end deep learning models. This method achieves accurate automated cancer segmentation without relying on cancer labels, reduces the cost of acquiring labeled data, and has very high practical feasibility.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107728"},"PeriodicalIF":4.9,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143578238","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":"Computer-aided diagnosis of spinal deformities based on keypoints detection in human back depth images","authors":"Malong Tan,&nbsp;Renchao Jin,&nbsp;Dun Liu,&nbsp;Shan Jiang,&nbsp;Xiangyang Xu,&nbsp;Enmin Song","doi":"10.1016/j.bspc.2025.107764","DOIUrl":"10.1016/j.bspc.2025.107764","url":null,"abstract":"<div><div>Physicians often need to manually measure physiological indicators through examinee’s X-ray for spinal deformities. However, this process is time-consuming and harmful to the human body because of radiation. Currently, computer-aided diagnosis technology has been gradually applied in the screening of spinal deformities. Due to the increased pressure on the spine and lower back during pregnancy, individuals may experience varying degrees of spinal deformities. We propose a novel method to help doctors non-invasively, accurately, and quickly diagnose spine deformity for postpartum women. First, this proposed method captures depth images of the human back by a Kinect DK camera. Second, it employs a region-growing method to extract the human back region and converts the back depth image into a point cloud image through mapping. After that, due to the issues of noise and voids in the depth image, joint bilateral filtering algorithm is used for repair and smoothing. Third, the OpenPose network is utilized to extract spinal keypoints (<em>i.e.,</em> C7, T12, L5 and S5). Then, we use Delaunay triangulation and linear interpolation to transform the point cloud image into a three-dimensional rectangular mesh. Subsequently, based on the concavity and convexity of the back surface, interpolation fitting of spine point set is performed using horizontal adjustment and B-spline fitting methods to reconstruct the spinal curve. Last, the experimental results indicate that the prediction accuracies for the six physiological indicators are 0.83, 0.43, 0.81, 0.82, 0.80, and 0.76, respectively. Additionally, the two PCK indicators are 0.87 and 0.90 for spinal point detection.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107764"},"PeriodicalIF":4.9,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563165","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":"Gaussian regressed generative adversarial network based hermitian extreme gradient boosting for plant leaf disease detection","authors":"S. Prakadeswaran ,&nbsp;A.Bazila Banu","doi":"10.1016/j.bspc.2025.107761","DOIUrl":"10.1016/j.bspc.2025.107761","url":null,"abstract":"<div><div>Identifying diseases from images of plant leaves is one of the most important research topics in the field of agriculture. Cassava is a rich plant in protein and vitamins, particularly in the leaves, and it is also employed as a substitute for rice. Since, the leaf is the most susceptible portion of a plant, it is said to be affected easily in comparison with the other parts. Therefore, the detection of plant leaf diseases may decrease the possibility that the plant will undergo additional damage. Many research works have been designed for Cassava plant leaf disease diagnosis, but the accuracy of leaf disease detection was not improved with the minimum amount of error and time. In this paper, a novel technique called Gaussian Regressed Generative Adversarial Networks based Hermitian Extreme Gradient Boosting (GRGAN-HEGB) is introduced for accurate cassava plant disease detection. The GRGAN-HEGB technique is composed of three parts: preprocessing, feature extraction, and classification. At first, the numbers of real cassava leaf images are collected from the Cassava Disease Classification dataset. Then, the collected cassava leaf images are preprocessed using the Gaussian Process Regressed Generative Adversarial Network (GPR-GAN) model. Feature extraction is performed by employing Continuous Hermitian Contingency Correlation (CH-CC) model to extract the most influential features such as texture, shape, and colour for disease pattern detection. Lastly, the Cophenetic Optimized Extreme Gradient Boosting classification process is performed to classify the input cassava leaf images with extracted features. As a result, the accuracy of leaf disease detection is improved with a short processing time. Experimental evaluation is carried out using the cassava disease classification dataset by considering different metrics such as disease detection accuracy, disease detection time, precision, and recall. The statistical results confirm that the proposed technique achieves higher accuracy by 12 %, precision by 2 %, and recall by 2 % with minimum disease detection processing time by 10 % than the conventional classification methods.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107761"},"PeriodicalIF":4.9,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563164","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":"Altered visual network modularity and communication in ADHD subtypes: Classification via source-localized EEG modules","authors":"Amir Hossein Ghaderi ,&nbsp;Shiva Taghizadeh ,&nbsp;Mohammad Ali Nazari","doi":"10.1016/j.bspc.2025.107755","DOIUrl":"10.1016/j.bspc.2025.107755","url":null,"abstract":"<div><div>The neurobiological basis of ADHD subtypes remains unclear. This study investigates how ADHD subtypes affect modularity of functional brain networks across different oscillatory bands. We analyzed EEG data from across three groups: normally developing, ADHD-Inattentive, and ADHD-Combined. EEG source-localized current densities were estimated and lagged coherence between the ROIs was calculated across seven frequency bands. We evaluated the modularity of five functional brain networks (default mode, central control, salience, visual, and sensorimotor) and assessed edge betweenness centrality to examine network communications. Nonparametric tests revealed significantly lower visual network modularity in ADHD groups in the alpha1 band (8–10 Hz). Communication between the visual and other networks (excluding the salience) was also significantly reduced in ADHD groups. No significant modularity or inter-network communication differences were observed between the ADHD subtypes. A supervised classification algorithm using subnetwork modularity as input achieved high accuracy (88.9 %) in classifying normally developing and ADHD groups based on alpha1 band data. The modularity of the sensorimotor, visual, and default mode networks emerged as key predictors. However, classification accuracy declined when distinguishing between the two ADHD subtypes. Results suggest impairment in early sensory (visual) processing in ADHD. Additionally, the combined modularity of the sensorimotor, visual, and default mode networks may serve as a potential biomarker for ADHD. Our results support a shared neural basis for ADHD subtypes, reinforcing the view that they are likely subtypes of the same disorder rather than distinct conditions.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"106 ","pages":"Article 107755"},"PeriodicalIF":4.9,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563261","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}