Critical reviews in biomedical engineering最新文献

{"title":"Diffusion Tensor Imaging for Brain Injury Assessment: Methodological Foundations and Clinical Insights.","authors":"Nicholas Simard, Michael D Noseworthy","doi":"10.1615/CritRevBiomedEng.2025059839","DOIUrl":"10.1615/CritRevBiomedEng.2025059839","url":null,"abstract":"<p><p>Diffusion tensor imaging (DTI) has emerged as a powerful neuroimaging modality for investigating white matter microstructure and its alterations following brain injury. This review presents a comprehensive overview of DTI, encompassing its physical principles, mathematical modeling of diffusion tensors, and known limitations of the technique. We explore key methodological considerations, including acquisition protocols, preprocessing pipelines, vendor-related variability, atlas registration, and the role of diffusion phantoms in calibration. With the rise of big data in medical imaging, we highlight the influence of large-scale, multisite datasets and open-source neuroimaging repositories in advancing DTI research. A central focus is placed on the application of DTI in mild traumatic brain injury, a condition that often eludes detection in conventional imaging settings. We evaluate emerging computational strategies, including Z-score analysis, principal component analysis, random forests, and generative adversarial networks-that improve the sensitivity, specificity, and interpretability of DTI metrics in both clinical and research settings. By bridging methodological rigor with translational insight, this review underscores the evolving potential of DTI as a neuroimaging biomarker for brain injury assessment.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 1","pages":"49-66"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147346192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Improved YOLOv8-Based Object Detection Algorithm for Skin Diseases.","authors":"Xinglong Yu","doi":"10.1615/CritRevBiomedEng.2025059866","DOIUrl":"10.1615/CritRevBiomedEng.2025059866","url":null,"abstract":"<p><p>Accurate and effective diagnosis of skin diseases is crucial for clinical decision-making. However, there are still some challenges, including irregular lesion morphologies, class imbalance between rare and common types, and performance degradation in complex scenarios characterized by noise or occlusion. To address these issues, an improved skin disease detection algorithm is proposed based on YOLOv8n, which features three core innovations. First, it integrates deformable large kernel attention (D-LKA) into deep backbone layers to capture global contextual relationships of lesions; then it embeds DCNv3 deformable convolutions in mid-layers to adaptively sample irregular lesion boundaries; finally, it designs an EMA-Slide Loss function to dynamically weight hard-to-classify samples, thereby reducing bias toward common categories. After evaluating on the International Skin Imaging Collaboration (ISIC) dataset (with labels validated by board-certified dermatologists), the algorithm can achieve 96.58% mAP50 and 88.32% mAP50-95, 2.44% and 2.52% higher than the baseline YOLOv8n, respectively. It maintains a real-time inference speed of 31 ms per image, making it suitable for edge devices such as portable dermatoscopes. Supporting nine common types of skin diseases, with extensibility to accommodate rare types and multi-modal data fusion, this work provides a clinically actionable tool for automated skin lesion analysis, bridging the gap between algorithmic performance and real-world clinical demands.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 1","pages":"1-16"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147346204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Denoising of Medical Ultrasound Images Using Variational Autoencoders Guided by Attention Mechanisms.","authors":"Taher Slimi, Houda Akremi, Anouar Ben Khalifa","doi":"10.1615/CritRevBiomedEng.2026060806","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2026060806","url":null,"abstract":"<p><p>Ultrasound imaging, while widely used and clinically valuable, continues to face a major limitation, speckle noise. This type of noise interferes with the visibility of fine anatomical structures and can lower the clinician's diagnostic confidence. Traditional denoising methods often struggle to strike the right balance between effectively reducing noise and preserving essential image details. As a result, their usefulness in real-world, high-demand clinical settings remains limited. To address this ongoing challenge, we introduce a new denoising framework that combines large kernel attention (LKA) with a convolutional variational autoencoder (CVAE). This hybrid approach takes advantage of LKA's ability to capture broad contextual information, improving feature representation, while the generative power of the CVAE helps reconstruct clean, speckle-free ultrasound images with greater reliability. Extensive evaluation across six diverse ultrasound datasets demonstrates that LKA-CVAE consistently outperforms state-of-the-art methods. The model achieves an average peak signal-to-noise ratio (PSNR) of 53.64 dB and structural similarity index measure (SSIM) of 0.913, representing improvements of 10 dB and 0.041, respectively, over the best competing approach. These gains are statistically significant and accompanied by minimal variability, with standard deviations of 0.028 for PSNR and 0.03 for SSIM. These results establish LKA-CVAE as a robust, high-performance tool for ultrasound image enhancement, supporting more confident and reliable clinical interpretation.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 2","pages":"1-20"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust Detection of Major Depressive Disorder Using Empirical Mode Decomposition-Based EEG Signal Analysis.","authors":"Khadidja Gouizi","doi":"10.1615/CritRevBiomedEng.2026063181","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2026063181","url":null,"abstract":"<p><p>Depression is universally known as a mood disorder, which is a serious and common disease that causes moral weakness and suicidal mortality. Most psychophysiologists, sought to use physiological data to infer this psychiatric illness. This work focuses on developing a major depression disorder detection system using EEG signals. In this study, a publicly available dataset is used. 19 channels of EEG signals data are collected from 16 depressive subjects and 16 healthy subjects with closed eyes and subsequently subjected to preprocessing procedures. Using correlation coefficients and standard deviation parameters, just 14 channels are selected. The first minute recording of EEG channels is processed using empirical mode decomposition (EMD) to extract some features, which are then used for the detection of major depressive disorder (MDD). These features are extracted, first by the decomposition of EEG channels into IMFs; then by the extraction of instantaneous amplitude and instantaneous frequency of each IMF. These features show significant differences between MDD subjects and healthy subjects. Using the linear kernel of the support vector machine (SVM), the classification results achieve higher accuracy, after dimensionality reduction with the principal component analysis method. All performance measures including accuracy, precision, sensitivity, specificity, and negative predictive value reach 100%. Thus, the findings confirm the robustness of the proposed system in detecting MDD.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 2","pages":"21-40"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterization of Vibration and Sound Exposure during Neonatal Transport: A Scoping Review of Methods and Findings.","authors":"Keely Gibb, Patrick Kehoe, Eleanor Gerson, Kim J Greenwood, Jean Ngoie, Stephanie Redpath, Andrew J Law, Adrian D C Chan, James R Green, Robert Langlois","doi":"10.1615/CritRevBiomedEng.2026062445","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2026062445","url":null,"abstract":"<p><p>Neonatal transport plays a vital role in ensuring that newborns who require specialized medical care can safely access appropriate services. Standard, regionalized transport protocols and equipment are required to improve access, timeliness, and consistency to the level of care provided. A key concern in neonatal transport is the potential exposure of newborns to elevated sound and whole-body vibration from the variety of transport incubators, medical equipment, and the vehicle platforms used in practice. Although international research groups are actively pursuing strategies to mitigate harmful sound and whole-body vibration, characterizing the transport environment remains a critical prerequisite for identifying resonance-driven amplification and informing targeted mitigation. This scoping review explores the methods and findings of vibration and sound characterization studies in neonatal transport. A total of 33 articles were included in this review, 28 of which evaluated vibration and 16 measured sound. The exposure levels of the different modes of transport and positions within the vehicles were compared. With the exception of a single study, sound measurements exceeded recommended limits for transport incubators, and vibration levels commonly surpassed \"uncomfortable\" thresholds for adults. Current whole-body vibration standards apply only to healthy adults, which results in ambiguity in their application to newborns and emphasizes a critical gap in defining safe exposure limits for neonates. Defining the risks, magnitude, and acceptable limits of these environmental stressors is essential to ensuring safer transport for vulnerable newborns.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 4","pages":"1-35"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Local-Global-Graph Network-Based Biokey Generation with Electrocardiogram Signal and Lightweight Authentication in Cloud-Based Internet of Medical Things Networks.","authors":"Sanjeev Kumar Adhinki Nagarathinam, Ramesh Naik Bhukya","doi":"10.1615/CritRevBiomedEng.2025058925","DOIUrl":"10.1615/CritRevBiomedEng.2025058925","url":null,"abstract":"<p><p>The internet of medical things (IoMT) is regarded as a promising framework, which is used to expand and improve telemedicine services. Cloud-based IoMT refers to the integration of medical devices and sensors with cloud computing infrastructure, enabling real-time remote data collection, processing, storage, and analysis. This architecture supports the efficient management of patient health information and facilitates advanced telemedicine services by offering scalable, secure, and accessible healthcare solutions. Ensuring secure access and communication in such systems is critical, as vulnerabilities in the network can expose sensitive patient data to significant risks. Among various security measures, authentication using biomedical signals, particularly electrocardiogram (ECG) signals, is gaining attention due to their unique, individual-specific characteristics. Therefore, this paper develops a new approach called local-global-graph network-based biokey generation (LGGNet-BioKey) for authentication in Cloud-based IoMT. Initially, the Cloud-based IoMT network is simulated, and it includes three entities, like cloud server, gateway, and patient. First, the public key and security parameters are initialized, and then the entities are registered with the cloud server. Next, the key generation is done using LGGNet, and then the BioKey generation is performed using an ECG signal. Next, the lightweight authentication is done and lastly, attribute-based encryption and decryption are performed in the data preservation phase. Furthermore, the LGGNet-BioKey model measured an execution time, memory usage, and key generation time of 3.772 sec, 9.096 MB, and 3.771 sec.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 1","pages":"67-95"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147346219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Critical Review of Patient-Reported Inputs, Clinical Workflows, and Machine Learning Opportunities for State-of-the-Art Spinal Cord Stimulation.","authors":"Charlotte Wade Henry, Mary Fendley, Erich Wolf","doi":"10.1615/CritRevBiomedEng.2026061703","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2026061703","url":null,"abstract":"<p><p>Spinal cord stimulation (SCS) is an established treatment for chronic pain, but long-term outcomes remain inconsistent. Failures may originate from poor therapeutic targeting and dosing, both of which can stem from low-resolution patient-reported inputs and the subjective, trial-and-error programming in clinical workflows. These challenges are especially evident in patient-preferred, nominally paresthesia-free stimulation, which lacks somatotopic feedback, thereby complicating effective targeting and dosing. This review identifies deficiencies in current SCS clinical workflows: (1) lack of high-resolution, spatially contextualized communication tools for patient-reported pain and paresthesia, and (2) subjectivity and inefficiency of manual programming. Common patient-to-provider communication tools, such as the numeric rating scale, compress complex pain experiences into single values, which limit their clinical utility and suggest an opportunity for higher-resolution, contextualized communication tools. Recent research suggests that graphical user interfaces for pain mapping, particularly those incorporating advanced spatial annotation tools or 3D capabilities, may overcome these communication barriers while also generating structured, traceable, spatial datasets. When combined with stimulation metadata, these datasets may provide a foundation for data-driven lead placement and therapy personalization, which is particularly important for the effective application of nominally paresthesia-free stimulation paradigms. Integrating machine learning (ML) into the SCS clinical workflow, made possible by leveraging a large dataset of these spatial inputs, has the potential to create an algorithm that automates programming. This integration has the capacity to improve the effectiveness of pain relief and expand access to care through remote services. However, challenges still exist in validating the accuracy of patient-reported maps, accounting for psychosocial influences, and navigating regulatory pathways for ML-driven SCS.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 3","pages":"79-99"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Patient-Centric Blockchain-Enabled Framework for Secure Medical Data Sharing in Indian EHR Systems.","authors":"Namdev Manik Sawant, Joanne Gomes","doi":"10.1615/CritRevBiomedEng.2026059613","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2026059613","url":null,"abstract":"<p><p>Electronic health records (EHRs) are integral to modern hospital systems, especially in India's rapidly digitizing healthcare landscape. EHRs connect hospital records through shared networks, enabling quick and accurate data sharing. However, these systems are targeted by attackers as they store large volumes of sensitive patient information. Thus, this research proposed a secure EHR framework named the sequential belief neural network (SeqBNNet). The SeqBNNet integrates deep belief networks (DBN) with sequential convolutional neural networks (SCNN) to encrypt EHRs. The images are split into N-shares, stored in the InterPlanetary File System (IPFS), and indexed using blockchain hashes for secure access. Data retrieval involves blockchain verification and decryption. The proposed model demonstrated strong performance with a value of 52.134 dB peak signal-to-noise ratio, 0.066 MSE, 0.518 s encryption time, 0.436 s decryption time, 0.576 s latency, and 97.754 Mbps throughput. This architecture offers a secure and scalable solution for managing patient data in Indian hospitals.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 3","pages":"21-46"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanical Testing and Computational Modeling of Scapula Fracture Implants: A Review.","authors":"Radovan Zdero, George S Athwal, Emil H Schemitsch, Pawel Brzozowski","doi":"10.1615/CritRevBiomedEng.2025060183","DOIUrl":"https://doi.org/10.1615/CritRevBiomedEng.2025060183","url":null,"abstract":"<p><p>This is the first review of mechanical testing or computational modeling articles that analyzed the biomechanical influence of parametrically changing scapula fracture implant (SFI) variables to optimize or characterize performance. PubMed and Embase were explored for articles using inclusion and exclusion criteria: (i) SFI articles that permutated implant variables were included, but articles using only one SFI configuration were excluded; (ii) SFIs comprising of plates (with screws), non-plate isolated \"lag\" screws, or cerclage wires were included, but anchors/sutures, bone cements, bone grafts, or soft-tissue grafts were excluded; (iii) scapula fractures were included, but adjacent injuries of the clavicle, humerus, labrum/capsule, ligaments/tendons, or muscles were excluded; (iv) cadaveric, synthetic, or computer-generated scapulas were included, but other anatomic sites or biomaterials were excluded; (v) published on any date in any language. The 16 eligible articles considered different scapula regions (i.e., acromion/spine, body, neck), SFI variables (i.e., plate geometry, plate hole type, plate position, plate number, screw angle), and engineering outcome metrics: (i) interfragmentary displacement (i.e., chance of callus growth); (ii) stresses in bones, plates, or screws (i.e., risk of failure); (iii) stresses in the bone under the plate (i.e., risk of bone \"stress shielding\"); (iv) the number of loading cycles to failure (i.e., fatigue life); (v) overall stiffness (i.e., physical flexibility); or (vi) failure strength (i.e., peak load achieved). This review advises future work for implant variables, bone characteristics, and study protocols, while suggesting SFI configurations with maximum mechanical stability.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"54 3","pages":"47-64"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mathematical Probe of Nanoparticle Aggregation in Capillary-Tissue System Applying Fractal Model.","authors":"Bhawini Prasad","doi":"10.1615/CritRevBiomedEng.2024055532","DOIUrl":"10.1615/CritRevBiomedEng.2024055532","url":null,"abstract":"<p><p>Targeted drug delivery using nanoparticle-based technology represents an advance in tumor treatment aiming to improve drug retention in tumors and minimize side effects. This study explores nanoparticle aggregation as a mechanism of enhanced retention and controlled dispersion of therapeutic agents in tumor tissues. Unlike existing models that primarily focus on single-particle diffusion, this research investigates the aggregation dynamics of nanoparticles upon diffusion from capillaries into the surrounding tissue, using a fractal-based mathematical model. By incorporating fractal geometry, this model uniquely captures the complexity of nanoparticle interactions with heterogeneous tumor environments. The equations, solved using MATLAB, reveal that nanoparticles form aggregates of approximately 75 nm in the capillary, with an optimal fractal dimension of 2.8 promoting efficient aggregation and retention. These findings provide a new perspective on aggregation-controlled drug delivery systems, offering insights for enhancing nanoparticle bioavailability and therapeutic efficacy in tumors.</p>","PeriodicalId":94308,"journal":{"name":"Critical reviews in biomedical engineering","volume":"53 5","pages":"1-22"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144786341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}