Medical & Biological Engineering & Computing最新文献

{"title":"GESur_Net: attention-guided network for surgical instrument segmentation in gastrointestinal endoscopy.","authors":"Yaru Ma, Yuying Liu, Xin Chen, Zhongqing Zheng, Yufeng Wang, Siyang Zuo","doi":"10.1007/s11517-025-03440-9","DOIUrl":"https://doi.org/10.1007/s11517-025-03440-9","url":null,"abstract":"<p><p>Surgical instrument segmentation plays an important role in robotic autonomous surgical navigation systems as it can accurately locate surgical instruments and estimate their posture, which helps surgeons understand the position and orientation of the instruments. However, there are still some problems affecting segmentation accuracy, like insufficient attention to the edges and center of surgical instruments, insufficient usage of low-level feature details, etc. To address these issues, a lightweight network for surgical instrument segmentation in gastrointestinal (GI) endoscopy (GESur_Net) is proposed. The pixel data aggregation (PDA) mechanism is proposed to analyze the pixel value distribution in the feature map to obtain the importance of each feature channel. The skip connection attention (SK_A) block is proposed to enhance the attention on critical regions of the surgical instruments. The global guidance attention (GGA) block is proposed to fuse high-level semantic information with low-level detailed features, enabling the acquisition of both fine-grained resolution and global semantic information. In addition, we constructed a new dataset, the Gastrointestinal Endoscopic Instrument (GEI) dataset, hoping to provide valuable resources for future research. Extensive experiments conducted on our presented GEI dataset and the Kvasir-instrument dataset demonstrate that the proposed GESur_Net increases the segmentation accuracy and outperforms state-of-the-art segmentation models.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145024629","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":"FetalMLOps: operationalizing machine learning models for standard fetal ultrasound plane classification.","authors":"Matteo Testi, Maria Chiara Fiorentino, Matteo Ballabio, Giorgio Visani, Massimo Ciccozzi, Emanuele Frontoni, Sara Moccia, Gennaro Vessio","doi":"10.1007/s11517-025-03436-5","DOIUrl":"10.1007/s11517-025-03436-5","url":null,"abstract":"<p><p>Fetal standard plane detection is essential in prenatal care, enabling accurate assessment of fetal development and early identification of potential anomalies. Despite significant advancements in machine learning (ML) in this domain, its integration into clinical workflows remains limited-primarily due to the lack of standardized, end-to-end operational frameworks. To address this gap, we introduce FetalMLOps, the first comprehensive MLOps framework specifically designed for fetal ultrasound imaging. Our approach adopts a ten-step MLOps methodology that covers the entire ML lifecycle, with each phase meticulously adapted to clinical needs. From defining the clinical objective to curating and annotating fetal US datasets, every step ensures alignment with real-world medical practice. ETL (extract, transform, load) processes are developed to standardize, anonymize, and harmonize inputs, enhancing data quality. Model development prioritizes architectures that balance accuracy and efficiency, using clinically relevant evaluation metrics to guide selection. The best-performing model is deployed via a RESTful API, following MLOps best practices for continuous integration, delivery, and performance monitoring. Crucially, the framework embeds principles of explainability and environmental sustainability, promoting ethical, transparent, and responsible AI. By operationalizing ML models within a clinically meaningful pipeline, FetalMLOps bridges the gap between algorithmic innovation and real-world application, setting a precedent for trustworthy and scalable AI adoption in prenatal care.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145016544","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 dual-branch encoder network based on squeeze-and-excitation UNet and transformer for 3D PET-CT image tumor segmentation.","authors":"Mingrui Li, Ruiming Zhu, Minghao Li, Haoran Wang, Yueyang Teng","doi":"10.1007/s11517-025-03427-6","DOIUrl":"https://doi.org/10.1007/s11517-025-03427-6","url":null,"abstract":"<p><p>Recognition of tumors is very important in clinical practice and radiomics; however, the segmentation task currently still needs to be done manually by experts. With the development of deep learning, automatic segmentation of tumors is gradually becoming possible. This paper combines the molecular information from PET and the pathology information from CT for tumor segmentation. A dual-branch encoder is designed based on SE-UNet (Squeeze-and-Excitation Normalization UNet) and Transformer, 3D Convolutional Block Attention Module (CBAM) is added to skip-connection, and BCE loss is used in training for improving segmentation accuracy. The new model is named TASE-UNet. The proposed method was tested on the HECKTOR2022 dataset, which obtains the best segmentation accuracy compared with state-of-the-art methods. Specifically, we obtained results of 76.10 <math><mo>%</mo></math> and 3.27 for the two key evaluation metrics, DSC and HD95. Experiments demonstrate that the designed network is reasonable and effective. The full implementation is available at https://github.com/LiMingrui1/TASE-UNet .</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145001837","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":"Deep learning-based morphological analysis of human sperm.","authors":"Yiran Xu, Yuqiu Chen, Boxuan Zhang, Yimo Yan, Hongen Liao, Ran Liu","doi":"10.1007/s11517-025-03418-7","DOIUrl":"10.1007/s11517-025-03418-7","url":null,"abstract":"<p><p>Sperm head morphology has been identified as a characteristic that can be used to predict a male's semen quality. Here, harnessing the close relationship considering sperm head shape to quality and morphology, we propose a joint learning model for sperm head segmentation and morphological category prediction. In the model, the sperm category prediction and the ellipticity, calculated by using the segmented sperm head profile, are used to synthesize the morphology to which the sperm belongs. In traditional clinical testing, fertility experts analyze sperm morphology by 2D images of sperm samples, which cannot represent the whole character of their quality and morphological category. To overcome the problem that single-angle 2D images cannot accurately identify sperm morphology, we use a deep-learning-based tracking and detection system to dynamically acquire sperm images with multiple frames and angles and then use the multi-frame and multi-angle time-series images of sperm to determine sperm morphology based on the multi-task model proposed in this study. Performing better than 3D sperm reconstruction and traditional computer-assisted sperm assessment systems, this approach enables end-to-end analysis of viable spermatozoa, requiring minimal computing power and utilizing equipment already available in most embryology laboratories.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144976430","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":"Gait anomaly detection based on video-derived 3D pose estimation.","authors":"Lingling Chen, Ye Zheng, Zhuo Gong, Ding Wang","doi":"10.1007/s11517-025-03339-5","DOIUrl":"10.1007/s11517-025-03339-5","url":null,"abstract":"<p><p>With the increase of age, the lower limb strength and function of the elderly gradually decline. Timely detection of motor dysfunction is of great significance for the prevention of disability, disease intervention, and improvement of living quality. Focusing on gait monitoring of the elderly living in groups, such as nursing homes, an abnormal gait recognition network based on daily walking information is proposed. We improve a multi-view 3D pose estimation network to extract gait parameters from the TUG exercise for monitoring, and design the abnormal gait recognition network to solve the problems of late evaluation of movement ability, large subjectivity, and the balance between accuracy and speed of the elderly living in groups. At a frame rate of 21.75 fps, the pose estimation accuracy is stable above 96.53%, and the joint error is controlled within 3.63°. In gait anomaly detection, the sensitivity reaches 96.71% and the inference speed reaches 512 ms; the F1 score reaches 0.9680, which is very close to the optimal value of the participant-comparison model, and the AUROC reaches 0.9694. This humble gait monitoring technology has great potential to provide assisted care and improve the overall well-being of the elderly.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2651-2663"},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143677340","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":"HADCN: a hierarchical ascending densely connected network for enhanced medical image segmentation.","authors":"Dibin Zhou, Mingxuan Zhao, Wenhao Liu, Xirui Gu","doi":"10.1007/s11517-025-03342-w","DOIUrl":"10.1007/s11517-025-03342-w","url":null,"abstract":"<p><p>Medical image segmentation is a key component in computer-aided diagnostic technology. In the past few years, the U-shaped architecture-based hierarchical model has become the mainstream approach, which however often fails to provide accurate results due to the loss of detailed features. To address this issue, this paper proposes a hierarchical ascending densely connected network, called HADCNet, to capture both local short-range and global long-range pathological features in a hierarchically organized network for more accurate segmentation. First, HADCNet devises a cross-scale ascending densely connected structure with a multi-path attention gate (MAG) to achieve full-scale interaction of global pathological features. Then, spatial-channel reconstruction units (called SRU and CRU) are introduced to decrease redundant computation and facilitate representative feature learning. Finally, multi-scale outputs are aggregated for final imaging. Extensive experiments demonstrate that our method achieves an average DSC of 84.45% and HD95 of 17.55 mm on the Synapse dataset (for multi-organ segmentation), with a similarly impressive performance on the ACDC (for cardiac diagnosis) and ISIC2018 datasets (for lesion segmentation). Additionally, HADCNet can be flexibly incorporated into existing backbone networks for better performance, e.g., combining HADC with TransUnet and SwinUnet, respectively, leads to 3.28% and 2.53% Dice score improvements.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2585-2600"},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143631003","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":"Null subtraction imaging combined with modified delay multiply-and-sum beamforming for coherent plane-wave compounding.","authors":"Yijun Xu, Yaoting Yue, Hao Wang, Wenting Gu, Boyi Li, Yaqing Chen, Xin Liu","doi":"10.1007/s11517-025-03364-4","DOIUrl":"10.1007/s11517-025-03364-4","url":null,"abstract":"<p><p>Coherent plane-wave compounding, while efficient for ultrafast ultrasound imaging, yields lower image quality due to unfocused waves. Delay multiply-and-sum (DMAS) beamformer is one of the representative coherence-based methods which can improve images quality, but suffers from poor speckle quality brought by oversuppression. Current DMAS-based methods involve trade-offs between contrast, resolution, and speckle preservation. To overcome this limitation, a new beamformer method combining the null subtraction imaging (NSI) and DMAS is investigated. The proposed method explores the DMAS on different beamformers which employs NSI and delay and sum (DAS) at receive and do multiply-and-sum on different beamformers across transmitting dimension, thereby simultaneously possessing the speckle quality of DAS and the high resolution of NSI. The effectiveness of the proposed method is evaluated through simulation, phantom, and in vivo datasets. From the experimental study, in comparison with NSI, the proposed method has improved contrast ratio by 10.02%, speckle signal-to-noise ratio by 45.19%, and generalized contrast-to-noise ratio by 12.37%. The method has also improved the full width at half maximum by up to 0.24 mm. The results indicate that the proposed method achieves better resolution and contrast, while also alleviating the issue of excessive compression.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2781-2793"},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144058444","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":"Artificial intelligence in antibody design and development: harnessing the power of computational approaches.","authors":"Soudabeh Kavousipour, Mahdi Barazesh, Shiva Mohammadi","doi":"10.1007/s11517-025-03429-4","DOIUrl":"https://doi.org/10.1007/s11517-025-03429-4","url":null,"abstract":"<p><p>Antibodies are a key therapeutic class in pharma, enabling precise targeting of disease agents. Traditional methods for their design are slow, costly, and limited. Advances in high-throughput data and artificial intelligence (AI) including machine learning, deep learning, and reinforcement learning have revolutionized antibody sequence design, 3D structure prediction, and optimization of affinity and specificity. Computational approaches enable rapid library generation and efficient screening, reduce experimental sampling, and support rational design with improved immune response. Combining AI with experimental methods allows for de novo, multifunctional antibody development. AI also accelerates the discovery process, target identification, and candidate prioritization by analyzing large datasets, predicting interactions, and guiding modifications to enhance efficacy and safety. Despite challenges, ongoing research continues to expand the potential of AI and transform antibody development and the pharmaceutical industry. Antibodies are a key therapeutic class in pharma, enabling precise targeting of disease agents. Traditional methods for their design are slow, costly, and limited. Advances in high-throughput data and artificial intelligence (AI) including machine learning, deep learning, and reinforcement learning have revolutionized antibody sequence design, 3D structure prediction, and optimization of affinity and specificity. Computational approaches enable rapid library generation and efficient screening, reduce experimental sampling, and support rational design with improved immune response. Combining AI with experimental methods allows for de novo, multifunctional antibody development. AI also accelerates the discovery process, target identification, and candidate prioritization by analyzing large datasets, predicting interactions, and guiding modifications to enhance efficacy and safety. Despite challenges, ongoing research continues to expand the potential of AI and transform antibody development and the pharmaceutical industry.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":""},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144976398","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":"GatedSegDiff: a gated fusion diffusion model for skin lesion segmentation.","authors":"Rui Wang, Liucheng Yao, Jiawen Zeng, Xiaofei Chen, Haiquan Wang, Chunhua Qian, Xiangyang Wang","doi":"10.1007/s11517-025-03337-7","DOIUrl":"10.1007/s11517-025-03337-7","url":null,"abstract":"<p><p>Skin lesion segmentation is a vital process in skin disease diagnosis, crucial for maintaining diagnostic precision. Despite progress in existing image segmentation methods, challenges remain in handling the fuzzy boundaries of skin lesion areas. To address this, we developed GatedSegDiff-a dedicated end-to-end framework for melanoma skin lesion image segmentation. Innovatively integrating the semantic representation capabilities of denoising networks with a novel gated attention fusion module, this model effectively merges feature maps across various scales, enhancing segmentation precision. We evaluate our model on the ISIC 2017, ISIC 2018, and PH2 image datasets. For the IoU score, our model achieved an average increase of 4.3% across three datasets, while the HD95 score decreased by 1.5%. GatedSegDiff outperforms existing advanced methods across multiple performance metrics, showing significant progress in skin lesion segmentation tasks and validating its effectiveness within this specific domain. Impact statement-The GatedSegDiff model's innovative application in medical image segmentation, particularly in skin lesion segmentation, significantly enhances diagnostic precision and efficiency. By concentrating on information in lesion boundary areas, it substantially improves segmentation accuracy for lesions with fuzzy boundaries, which is crucial for the early diagnosis of serious skin diseases like melanoma. Additionally, it provides a solution to the shortcomings of general medical image segmentation methods in handling specific skin lesions, its applicability to other types of medical images requires further investigation. The model's outstanding performance on multiple skin lesion datasets highlights its potential for application in digital dermatological diagnosis, offering faster and more reliable services to patients, with significant implications for clinical use in the field of skin disease diagnosis. Melanin segmentation can be applied to medical integrated classification techniques to help experts select the most suitable treatment options for patients.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2637-2650"},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143651678","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":"Reimagining cancer tissue classification: a multi-scale framework based on multi-instance learning for whole slide image classification.","authors":"Zixuan Wu, Haiyong He, Xiushun Zhao, Zhenghui Lin, Yanyan Ye, Jing Guo, Wanming Hu, Xiaobing Jiang","doi":"10.1007/s11517-025-03341-x","DOIUrl":"10.1007/s11517-025-03341-x","url":null,"abstract":"<p><p>In cancer pathology diagnosis, analyzing Whole Slide Images (WSI) encounters challenges like invalid data, varying tissue features at different magnifications, and numerous hard samples. Multiple Instance Learning (MIL) is a powerful tool for addressing weakly supervised classification in WSI-based pathology diagnosis. However, existing MIL frameworks cannot simultaneously tackle these issues. To address these challenges, we propose an integrated recognition framework comprising three complementary components: a preprocessing selection method, an Efficient Feature Pyramid Network (EFPN) model for multi-instance learning, and a Similarity Focal Loss. The preprocessing selection method accurately identifies and selects representative image patches, effectively reducing invalid data interference and enhancing subsequent model training efficiency. The EFPN model, inspired by pathologists' diagnostic processes, captures different tissue features in WSI images by constructing a multi-scale feature pyramid, enhancing the model's ability to recognize tumor tissue features. Additionally, the Similarity Focal Loss further improves the model's discriminative power and generalization performance by focusing on hard samples and emphasizing classification boundary information. The test accuracy for binary tumor classification on the CAMELYON16 and two private datasets reached 93.58%, 84.74%, and 99.91%, respectively, all of which outperform existing techniques.</p>","PeriodicalId":49840,"journal":{"name":"Medical & Biological Engineering & Computing","volume":" ","pages":"2617-2635"},"PeriodicalIF":2.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143634908","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}