Medical & Biological Engineering & Computing最新文献

筛选
英文 中文
Deep learning for retinal vessel segmentation: a systematic review of techniques and applications. 视网膜血管分割的深度学习:技术和应用的系统回顾。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-02-18 DOI: 10.1007/s11517-025-03324-y
Zhihui Liu, Mohd Shahrizal Sunar, Tian Swee Tan, Wan Hazabbah Wan Hitam
{"title":"Deep learning for retinal vessel segmentation: a systematic review of techniques and applications.","authors":"Zhihui Liu, Mohd Shahrizal Sunar, Tian Swee Tan, Wan Hazabbah Wan Hitam","doi":"10.1007/s11517-025-03324-y","DOIUrl":"10.1007/s11517-025-03324-y","url":null,"abstract":"<p><p>Ophthalmic diseases are a leading cause of vision loss, with retinal damage being irreversible. Retinal blood vessels are vital for diagnosing eye conditions, as even subtle changes in their structure can signal underlying issues. Retinal vessel segmentation is key for early detection and treatment of eye diseases. Traditionally, ophthalmologists manually segmented vessels, a time-consuming process based on clinical and geometric features. However, deep learning advancements have led to automated methods with impressive results. This systematic review, following PRISMA guidelines, examines 79 studies on deep learning-based retinal vessel segmentation published between 2020 and 2024 from four databases: Web of Science, Scopus, IEEE Xplore, and PubMed. The review focuses on datasets, segmentation models, evaluation metrics, and emerging trends. U-Net and Transformer architectures have shown success, with U-Net's encoder-decoder structure preserving details and Transformers capturing global context through self-attention mechanisms. Despite their effectiveness, challenges remain, suggesting future research should explore hybrid models combining U-Net, Transformers, and GANs to improve segmentation accuracy. This review offers a comprehensive look at the current landscape and future directions in retinal vessel segmentation.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2191-2208"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143442601","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}
引用次数: 0
InspirationOnly: synthesizing expiratory CT from inspiratory CT to estimate parametric response map. InspirationOnly:从吸气CT合成呼气CT,估计参数响应图。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-02-17 DOI: 10.1007/s11517-025-03322-0
Tiande Zhang, Haowen Pang, Yanan Wu, Jiaxuan Xu, Zhenyu Liang, Shuyue Xia, Chenwang Jin, Rongchang Chen, Shouliang Qi
{"title":"InspirationOnly: synthesizing expiratory CT from inspiratory CT to estimate parametric response map.","authors":"Tiande Zhang, Haowen Pang, Yanan Wu, Jiaxuan Xu, Zhenyu Liang, Shuyue Xia, Chenwang Jin, Rongchang Chen, Shouliang Qi","doi":"10.1007/s11517-025-03322-0","DOIUrl":"10.1007/s11517-025-03322-0","url":null,"abstract":"<p><p>Chronic obstructive pulmonary disease (COPD) is a highly heterogeneous disease with various phenotypes. Registered inspiratory and expiratory CT images can generate the parametric response map (PRM) that characterizes phenotypes' spatial distribution and proportions. However, increased radiation dosage, scan time, quality control, and patient cooperation requirements limit the utility of PRM. This study aims to synthesize a PRM using only inspiratory CT scans. First, a CycleGAN with perceptual loss and a multiscale discriminator (MPCycleGAN) is proposed and trained to synthesize registered expiratory CT images from inspiratory images. Next, a strategy named InspirationOnly is introduced, where synthesized images replace actual expiratory CT images. The image synthesizer outperformed state-of-the-art models, achieving a mean absolute error of 105.66 ± 36.64 HU, a peak signal-to-noise ratio of 21.43 ± 1.87 dB, and a structural similarity of 0.84 ± 0.02. The intraclass correlation coefficients of emphysema, fSAD, and normal proportions between the InspirationOnly and ground truth were 0.995, 0.829, and 0.914, respectively. The proposed MPCycleGAN enables the InspirationOnly strategy to yield PRM using only inspiratory CT. The estimated COPD phenotypes are consistent with those from dual-phase CT and correlated with the spirometry parameters. This offers a potential tool for characterizing phenotypes of COPD, particularly when expiratory CT images are unavailable.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2277-2294"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143442606","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}
引用次数: 0
Precise dental caries segmentation in X-rays with an attention and edge dual-decoder network. 基于注意力和边缘双解码网络的x射线龋齿精确分割。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-02-17 DOI: 10.1007/s11517-025-03318-w
Feng Huang, Jiaxing Yin, Yuxin Ma, Hao Zhang, Shunv Ying
{"title":"Precise dental caries segmentation in X-rays with an attention and edge dual-decoder network.","authors":"Feng Huang, Jiaxing Yin, Yuxin Ma, Hao Zhang, Shunv Ying","doi":"10.1007/s11517-025-03318-w","DOIUrl":"10.1007/s11517-025-03318-w","url":null,"abstract":"<p><p>Caries segmentation holds significant clinical importance in medical image analysis, particularly in the early detection and treatment of dental caries. However, existing deep learning segmentation methods often struggle with accurately segmenting complex caries boundaries. To address this challenge, this paper proposes a novel network, named AEDD-Net, which combines an attention mechanism with a dual-decoder structure to enhance the performance of boundary segmentation for caries. Unlike traditional methods, AEDD-Net integrates atrous spatial pyramid pooling with cross-coordinate attention mechanisms to effectively fuse global and multi-scale features. Additionally, the network introduces a dedicated boundary generation module that precisely extracts caries boundary information. Moreover, we propose an innovative boundary loss function to further improve the learning of boundary features. Experimental results demonstrate that AEDD-Net significantly outperforms other comparison networks in terms of Dice coefficient, Jaccard similarity, precision, and sensitivity, particularly showing superior performance in boundary segmentation. This study provides an innovative approach for automated caries segmentation, with promising potential for clinical applications.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2259-2276"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143442607","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}
引用次数: 0
Drug repositioning based on mutual information for the treatment of Alzheimer's disease patients. 基于互信息的药物重新定位治疗阿尔茨海默病患者。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-02-17 DOI: 10.1007/s11517-025-03325-x
Claudia Cava, Isabella Castiglioni
{"title":"Drug repositioning based on mutual information for the treatment of Alzheimer's disease patients.","authors":"Claudia Cava, Isabella Castiglioni","doi":"10.1007/s11517-025-03325-x","DOIUrl":"10.1007/s11517-025-03325-x","url":null,"abstract":"<p><p>Computational drug repositioning approaches should be investigated for the identification of new treatments for Alzheimer's patients as a huge amount of omics data has been produced during pre-clinical and clinical studies. Here, we investigated a gene network in Alzheimer's patients to detect a proper therapeutic target. We screened the targets of different drugs (34,006 compounds) using data available in the Connectivity Map database. Then, we analyzed transcriptome profiles of Alzheimer's patients to discover a network of gene-drugs based on mutual information, representing an index of dependence among genes. This study identified a network consisting of 25 genes and compounds and interconnected biological processes using computational approaches. The results also highlight the diagnostic role of the 25 genes since we obtained good classification performances using a neural network model. We also suggest 12 repurposable drugs (like KU-60019, AM-630, CP55940, enflurane, ginkgolide B, linopirdine, apremilast, ibudilast, pentoxifylline, roflumilast, acitretin, and tamibarotene) interacting with 6 genes (ATM, CNR1, GLRB, KCNQ2, PDE4B, and RARA), that we linked to retrograde endocannabinoid signaling, synaptic vesicle cycle, morphine addiction, and homologous recombination.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2249-2257"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143442603","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}
引用次数: 0
SAMP-Net: a medical image segmentation network with split attention and multi-layer perceptron. SAMP-Net:一种分离注意和多层感知器的医学图像分割网络。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-03-11 DOI: 10.1007/s11517-025-03331-z
Xiaoxuan Ma, Sihan Shan, Dong Sui
{"title":"SAMP-Net: a medical image segmentation network with split attention and multi-layer perceptron.","authors":"Xiaoxuan Ma, Sihan Shan, Dong Sui","doi":"10.1007/s11517-025-03331-z","DOIUrl":"10.1007/s11517-025-03331-z","url":null,"abstract":"<p><p>Convolutional neural networks (CNNs) have achieved remarkable success in computer vision, particularly in medical image segmentation. U-Net, a prominent architecture, marked a major breakthrough and remains widely used in practice. However, its uniform downsampling strategy and simple stacking of convolutional layers in the encoder limit its ability to capture rich features at multiple depths, reducing its efficiency for rapid image processing. To address these limitations, this paper proposes a novel segmentation network that integrates attention mechanisms with multilayer perceptrons (MLPs). The network is designed to progressively capture and refine features at different levels. At the low-level layers, the primary feature conservation (PFC) block is introduced to preserve essential spatial details and reduce the loss of primary features during downsampling. In the mid-level layers, the compact attention block (CAB) enhances feature interaction through a multi-path attention structure, improving the network's ability to capture diverse semantic information. At the high-level layers, Shift MLP and Tokenized MLP blocks are incorporated. The Shift MLP block shifts feature channels along different axes, allowing for enhanced local feature modeling by focusing on specific regions of the convolutional features. The Tokenized MLP block converts these features into abstract tokens and leverages MLPs to model their representations in the latent space, effectively reducing the number of parameters and computational complexity while improving segmentation performance. The experiments conducted on the colorectal cancer tumor dataset CCI and the public dataset ISIC-2018 demonstrate that the proposed method significantly outperforms U-Net, U-Net++, Swin-U-Net, Attention U-Net, and RA-U-Net in terms of performance, with average improvements of 6.67%, 5.53%, 10.18%, 4.78%, and 3.55%, respectively. The code is available at the following link: https://github.com/QingTianer/SAMP-Net.git.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2437-2450"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143598161","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}
引用次数: 0
When deep learning is not enough: artificial life as a supplementary tool for segmentation of ultrasound images of breast cancer. 当深度学习还不够时:人工生命作为乳腺癌超声波图像分割的辅助工具。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2024-03-18 DOI: 10.1007/s11517-024-03026-x
Nalan Karunanayake, Stanislav S Makhanov
{"title":"When deep learning is not enough: artificial life as a supplementary tool for segmentation of ultrasound images of breast cancer.","authors":"Nalan Karunanayake, Stanislav S Makhanov","doi":"10.1007/s11517-024-03026-x","DOIUrl":"10.1007/s11517-024-03026-x","url":null,"abstract":"<p><p>Segmentation of tumors in ultrasound (US) images of the breast is a critical issue in medical imaging. Due to the poor quality of US images and the varying specifications of US machines, segmentation and classification of abnormalities present difficulties even for trained radiologists. The paper aims to introduce a novel AI-based hybrid model for US segmentation that offers high accuracy, requires relatively smaller datasets, and is capable of handling previously unseen data. The software can be used for diagnostics and the US-guided biopsies. A unique and robust hybrid approach that combines deep learning (DL) and multi-agent artificial life (AL) has been introduced. The algorithms are verified on three US datasets. The method outperforms 14 selected state-of-the-art algorithms applied to US images characterized by complex geometry and high level of noise. The paper offers an original classification of the images and tests to analyze the limits of the DL. The model has been trained and verified on 1264 ultrasound images. The images are in the JPEG and PNG formats. The age of the patients ranges from 22 to 73 years. The 14 benchmark algorithms include deformable shapes, edge linking, superpixels, machine learning, and DL methods. The tests use eight-region shape- and contour-based evaluation metrics. The proposed method (DL-AL) produces excellent results in terms of the dice coefficient (region) and the relative Hausdorff distance H<sub>3</sub> (contour-based) as follows: the easiest image complexity level, Dice = 0.96 and H<sub>3</sub> = 0.26; the medium complexity level, Dice = 0.91 and H<sub>3</sub> = 0.82; and the hardest complexity level, Dice = 0.90 and H<sub>3</sub> = 0.84. All other metrics follow the same pattern. The DL-AL outperforms the second best (Unet-based) method by 10-20%. The method has been also tested by a series of unconventional tests. The model was trained on low complexity images and applied to the entire set of images. These results are summarized below. (1) Only the low complexity images have been used for training (68% unknown images): Dice = 0.80 and H<sub>3</sub> = 2.01. (2) The low and the medium complexity images have been used for training (51% unknown images): Dice = 0.86 and H<sub>3</sub> = 1.32. (3) The low, medium, and hard complexity images have been used for training (35% unknown images): Dice = 0.92 and H<sub>3</sub> = 0.76. These tests show a significant advantage of DL-AL over 30%. A video demo illustrating the algorithm is at http://tinyurl.com/mr4ah687 .</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2497-2520"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140159417","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}
引用次数: 0
LGENet: disentangle anatomy and pathology features for late gadolinium enhancement image segmentation. LGENet:解开晚期钆增强图像分割的解剖和病理特征。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-02-24 DOI: 10.1007/s11517-025-03326-w
Mingjing Yang, Kangwen Yang, Mengjun Wu, Liqin Huang, Wangbin Ding, Lin Pan, Lei Yin
{"title":"LGENet: disentangle anatomy and pathology features for late gadolinium enhancement image segmentation.","authors":"Mingjing Yang, Kangwen Yang, Mengjun Wu, Liqin Huang, Wangbin Ding, Lin Pan, Lei Yin","doi":"10.1007/s11517-025-03326-w","DOIUrl":"10.1007/s11517-025-03326-w","url":null,"abstract":"<p><p>Myocardium scar segmentation is essential for clinical diagnosis and prognosis for cardiac vascular diseases. Late gadolinium enhancement (LGE) imaging technology has been widely utilized to visualize left atrial and ventricular scars. However, automatic scar segmentation remains challenging due to the imbalance between scar and background and the variation in scar sizes. To address these challenges, we introduce an innovative network, i.e., LGENet, for scar segmentation. LGENet disentangles anatomy and pathology features from LGE images. Note that inherent spatial relationships exist between the myocardium and scarring regions. We proposed a boundary attention module to allow the scar segmentation conditioned on anatomical boundary features, which could mitigate the imbalance problem. Meanwhile, LGENet can predict scar regions across multiple scales with a multi-depth decision module, addressing the scar size variation issue. In our experiments, we thoroughly evaluated the performance of LGENet using LAScarQS 2022 and EMIDEC datasets. The results demonstrate that LGENet achieved promising performance for cardiac scar segmentation.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2311-2323"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143484491","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}
引用次数: 0
Comprehensive comparison of different BITA graft configurations: a computational study integrating TTFM and hemodynamic predictors. 综合比较不同的BITA移植物构型:一项整合TTFM和血流动力学预测因子的计算研究。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-03-05 DOI: 10.1007/s11517-025-03336-8
Ahmad Masoudi, Hossein Ali Pakravan
{"title":"Comprehensive comparison of different BITA graft configurations: a computational study integrating TTFM and hemodynamic predictors.","authors":"Ahmad Masoudi, Hossein Ali Pakravan","doi":"10.1007/s11517-025-03336-8","DOIUrl":"10.1007/s11517-025-03336-8","url":null,"abstract":"<p><p>Bilateral internal thoracic artery (BITA) grafting utilizes both the left (LITA) and right (RITA) internal thoracic arteries simultaneously and is recommended in the literature. However, the optimal configuration for BITA grafting remains uncertain. In this study, three-dimensional numerical simulations of different BITA configurations were conducted to identify the optimal configuration and assess their performance using the fractional flow reserve (FFR), transit time flow meter (TTFM), and hemodynamic parameters. The vessel geometry of a patient who underwent a BITA grafting with a Y-graft configuration was extracted from CT angiography images, and three other configurations (pedicle, LITA as free graft, and RITA as free graft) with different degrees of stenosis were reconstructed. Results showed that, in mild to moderate stenosis (FFR > 0.7), the Y-graft configuration was less favorable for graft quality, as it had higher pulsatility index (PI) and systolic reverse flow (SRF) values, leading to increased competitive flow. However, as stenosis severity increased, these differences decreased, and for severe stenosis, the results were similar for all BITA configurations. Furthermore, the results showed that the Y-graft configuration was less effective in reducing TAWSS compared to other configurations. Oscillatory shear index (OSI) and relative residence time (RRT) did not show significant differences.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2393-2406"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143558460","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}
引用次数: 0
Correction to: Optimization of CBCT data with image processing methods and production with fused deposition modeling 3D printing. 校正:用图像处理方法优化CBCT数据,用熔融沉积建模3D打印生产。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 DOI: 10.1007/s11517-024-03277-8
Hamdi Sayin, Bekir Aksoy, Koray Özsoy, Derya Yildirim
{"title":"Correction to: Optimization of CBCT data with image processing methods and production with fused deposition modeling 3D printing.","authors":"Hamdi Sayin, Bekir Aksoy, Koray Özsoy, Derya Yildirim","doi":"10.1007/s11517-024-03277-8","DOIUrl":"10.1007/s11517-024-03277-8","url":null,"abstract":"","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2247"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142933400","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}
引用次数: 0
New AI explained and validated deep learning approaches to accurately predict diabetes. 新的人工智能解释并验证了深度学习方法,以准确预测糖尿病。
IF 2.6 4区 医学
Medical & Biological Engineering & Computing Pub Date : 2025-08-01 Epub Date: 2025-03-04 DOI: 10.1007/s11517-025-03338-6
Ifra Shaheen, Nadeem Javaid, Nabil Alrajeh, Yousra Asim, Syed Muhammad Abrar Akber
{"title":"New AI explained and validated deep learning approaches to accurately predict diabetes.","authors":"Ifra Shaheen, Nadeem Javaid, Nabil Alrajeh, Yousra Asim, Syed Muhammad Abrar Akber","doi":"10.1007/s11517-025-03338-6","DOIUrl":"10.1007/s11517-025-03338-6","url":null,"abstract":"<p><p>Diabetes is a metabolic condition that can lead to chronic illness and organ failure if it remains untreated. Accurate detection is essential to reduce these risks at an early stage. Recent advancements in predictive models show promising results. However, these models exhibit inadequate accuracy, struggle with class imbalance, and lack interpretability of the decision-making process. To overcome these issues, we propose two novel deep models for early and accurate diabetes prediction: LeDNet (inspired by LeNet and the Dual Attention Network) and HiDenNet (influenced by the highway network and DenseNet). The models are trained using the Diabetes Health Indicators dataset, which has an inherent class imbalance problem and results in biased predictions. This imbalance is mitigated by employing the majority-weighted minority over-sampling technique. Experimental findings demonstrate that LeDNet achieves an F1-score of 85%, recall of 84%, accuracy of 85%, and precision of 86%. Similarly, HiDenNet achieves accuracy, F1-score, recall, and precision of 85%, 86%, 86%, and 86%, respectively. Both proposed models outperform the state-of-the-art deep learning (DL) models. K-fold cross-validation is applied to ensure models' stability at different data splits. Local interpretable model-agnostic explanations and Shapley additive explanations techniques are utilized to enhance interpretability and overcome the traditional black-box nature of DL models. By providing both local and global insights into feature contributions, these explainable artificial intelligence techniques provide transparency to LeDNet and HiDenNet in diabetes prediction. LeDNet and HiDenNet not only improve decision-making transparency but also enhance diabetes prediction accuracy, making them reliable tools for clinical decision-making and early diagnosis.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2373-2392"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143544183","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}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信