Sci Rep. 2025 Jan 26;15(1):3296. doi: 10.1038/s41598-025-87993-7.

ABSTRACT

In response to the pressing need for the detection of Monkeypox caused by the Monkeypox virus (MPXV), this study introduces the Enhanced Spatial-Awareness Capsule Network (ESACN), a Capsule Network architecture designed for the precise multi-class classification of dermatological images. Addressing the shortcomings of traditional Machine Learning and Deep Learning models, our ESACN model utilizes the dynamic routing and spatial hierarchy capabilities of CapsNets to differentiate complex patterns such as those seen in monkeypox, chickenpox, measles, and normal skin presentations. CapsNets' inherent ability to recognize and process crucial spatial relationships within images outperforms conventional CNNs, particularly in tasks that require the distinction of visually similar classes. Our model's superior performance, demonstrated through rigorous evaluation, exhibits significant improvements in accuracy, precision, recall, and F1 score, even with limited data. The results highlight the potential of ESACN as a reliable tool for enhancing diagnostic accuracy in medical settings. In our case study, the ESACN model was applied to a dataset comprising 659 images across four classes: 178 images of Monkeypox, 171 of Chickenpox, 80 of Measles, and 230 of Normal skin conditions. This case study underscores the model's effectiveness in real-world applications, providing robust and accurate classification that could greatly aid in early diagnosis and treatment planning in clinical environments.

PMID:39865160 | DOI:10.1038/s41598-025-87993-7

Nat Commun. 2025 Jan 26;16(1):1057. doi: 10.1038/s41467-025-56322-x.

ABSTRACT

Enhancing the ion conduction in solid electrolytes is critically important for the development of high-performance all-solid-state lithium-ion batteries (LIBs). Lithium thiophosphates are among the most promising solid electrolytes, as they exhibit superionic conductivity at room temperature. However, the lack of comprehensive understanding of their ion conduction mechanism, especially the effect of structural disorder on ionic conductivity, is a long-standing problem that limits further innovations in all-solid-state LIBs. Here, we address this challenge by establishing and employing a deep learning potential to simulate Li3PS4 electrolyte systems with varying levels of disorder. The results show that disorder-driven diffusion dynamics significantly enhances the room-temperature conductivity. We further establish bridges between dynamical characteristics, local structural features, and atomic rearrangements by applying a machine learning-based structure fingerprint termed "softness". This metric allows the classification of the disorder-induced "soft" hopping lithium ions. Our findings offer insights into ion conduction mechanisms in complex disordered structures, thereby contributing to the development of superior solid-state electrolytes for LIBs.

PMID:39865086 | DOI:10.1038/s41467-025-56322-x

Acta Med Indones. 2024 Oct;56(4):563-570.

ABSTRACT

The burden of undiagnosed diabetes mellitus (DM) is substantial, with approximately 240 million individuals globally unaware of their condition, disproportionately affecting low- and middle-income countries (LMICs), including Indonesia. Without screening, DM and its complications will impose significant pressure on healthcare systems. Current clinical practices for screening and diagnosing DM primarily involve blood or laboratory-based testing which possess limitations on access and cost. To address these challenges, researchers have developed risk-scoring tools to identify high-risk populations. However, considering generalizability, artificial intelligence (AI) technologies offer a promising approach, leveraging diverse data sources for improved accuracy. AI models (i.e., machine learning and deep learning) have yielded prediction performances of up to 98% in various diseases. This article underscores the potential of AI-driven approaches in reducing the burden of DM through accurate prediction of undiagnosed diabetes while highlighting the need for continued innovation and collaboration in healthcare delivery.

PMID:39865054

Ultrasound Med Biol. 2025 Jan 25:S0301-5629(24)00469-1. doi: 10.1016/j.ultrasmedbio.2024.12.008. Online ahead of print.

ABSTRACT

OBJECTIVE: To develop and compare methods to automatically estimate regional ultrasound image quality for echocardiography separate from view correctness.

METHODS: Three methods for estimating image quality were developed: (i) classic pixel-based metric: the generalized contrast-to-noise ratio (gCNR), computed on myocardial segments (region of interest) and left ventricle lumen (background), extracted by a U-Net segmentation model; (ii) local image coherence: the average local coherence as predicted by a U-Net model that predicts image coherence from B-mode ultrasound images at the pixel level; (iii) deep convolutional network: an end-to-end deep-learning model that predicts the quality of each region in the image directly. These methods were evaluated against manual regional quality annotations provided by three experienced cardiologists.

RESULTS: The results indicated poor performance of the gCNR metric, with Spearman correlation to annotations of ρ = 0.24. The end-to-end learning model obtained the best result, ρ = 0.69, comparable to the inter-observer correlation, ρ = 0.63. Finally, the coherence-based method, with ρ = 0.58, out-performed the classical metrics and was more generic than the end-to-end approach.

CONCLUSION: The deep convolutional network provided the most accurate regional quality prediction, while the coherence-based method offered a more generalizable solution. gCNR showed limited effectiveness in this study. The image quality prediction tool is available as an open-source Python library at https://github.com/GillesVanDeVyver/arqee.

PMID:39864961 | DOI:10.1016/j.ultrasmedbio.2024.12.008

Neurol Med Chir (Tokyo). 2025 Jan 24. doi: 10.2176/jns-nmc.2024-0163. Online ahead of print.

ABSTRACT

Non-enhanced head computed tomography is widely used for patients presenting with head trauma or stroke, given acute intracranial hemorrhage significantly influences clinical decision-making. This study aimed to develop a deep learning algorithm, referred to as DeepCT, to detect acute intracranial hemorrhage on non-enhanced head computed tomography images and evaluate its clinical applicability. We retrospectively collected 1,815 computed tomography image sets from a single center for model training. Additional computed tomography sets from 3 centers were used to construct an independent validation dataset (VAL) and 2 test datasets (GPS-C and DICH). A third test dataset (US-TW) comprised 150 cases, each from 1 hospital in Taiwan and 1 hospital in the United States of America. Our deep learning model, based on U-Net and ResNet architectures, was implemented using PyTorch. The deep learning algorithm exhibited high accuracy across the validation and test datasets, with overall accuracy ranging from 0.9343 to 0.9820. Our findings show that the deep learning algorithm effectively identifies acute intracranial hemorrhage in non-enhanced head computed tomography studies. Clinically, this algorithm can be used for hyperacute triage, reducing reporting times, and enhancing the accuracy of radiologist interpretations. The evaluation of the algorithm on both United States and Taiwan datasets further supports its universal reliability for detecting acute intracranial hemorrhage.

PMID:39864839 | DOI:10.2176/jns-nmc.2024-0163

Biosci Trends. 2025 Jan 25. doi: 10.5582/bst.2024.01370. Online ahead of print.

ABSTRACT

In human-computer interaction, gesture recognition based on physiological signals offers advantages such as a more natural and fast interaction mode and less constrained by the environment than visual-based. Surface electromyography-based gesture recognition has significantly progressed. However, since individuals have physical differences, researchers must collect data multiple times from each user to train the deep learning model. This data acquisition process can be particularly burdensome for non-healthy users. Researchers are currently exploring transfer learning and data augmentation techniques to enhance the accuracy of small-sample gesture recognition models. However, challenges persist, such as negative transfer and limited diversity in training samples, leading to suboptimal recognition performance. Therefore, We introduce motion information into sEMG-based recognition and propose a multimodal optimal matching and augmentation method for small sample gesture recognition, achieving efficient gesture recognition with only one acquisition per gesture. Firstly, this method utilizes the optimal matching signal selection module to select the most similar signals from the existing data to the new user as the training set, reducing inter-domain differences. Secondly, the similarity calculation augmentation module enhances the diversity of the training set. Finally, the Modal-type embedding enhances the information interaction between each mode signal. We evaluated the effectiveness on Self-collected Stroke Patient, the Ninapro DB1 dataset and the Ninapro DB5 dataset and achieved accuracies of 93.69%, 91.65% and 98.56%, respectively. These results demonstrate that the method achieved performance comparable to traditional recognition models while significantly reducing the collected data.

PMID:39864830 | DOI:10.5582/bst.2024.01370

J Cardiovasc Magn Reson. 2025 Jan 24:101844. doi: 10.1016/j.jocmr.2025.101844. Online ahead of print.

ABSTRACT

PURPOSE: To investigate image quality and agreement of derived cardiac function parameters in a novel joint image reconstruction and segmentation approach based on disentangled representation learning, enabling real-time cardiac cine imaging during free-breathing.

METHODS: A multi-tasking neural network architecture, incorporating disentangled representation learning, was trained using simulated examinations based on data from a public repository along with MR scans specifically acquired for model development. An exploratory feasibility study evaluated the method on undersampled real-time acquisitions using an in-house developed spiral bSSFP pulse sequence in eight healthy participants and five patients with intermittent atrial fibrillation. Images and predicted LV segmentations were compared to the reference standard of ECG-gated segmented Cartesian cine with repeated breath-holds and corresponding manual segmentation.

RESULTS: On a 5-point Likert scale, image quality of the real-time breath-hold approach and Cartesian cine was comparable in healthy participants (RT-BH: 1.99 ±.98, Cartesian: 1.94 ±.86, p=.052), but slightly inferior in free-breathing (RT-FB: 2.40 ±.98, p<.001). In patients with arrhythmia, both real-time approaches demonstrated favourable image quality (RT-BH: 2.10 ± 1.28, p<.001, RT-FB: 2.40 ± 1.13, p<.01, Cartesian: 2.68 ± 1.13). Intra-observer reliability was good (ICC=.77,95%-confidence interval [.75,.79], p<.001). In functional analysis, a positive bias was observed for ejection fractions derived from the proposed model compared to the clinical reference standard (RT-BH mean: 58.5 ± 5.6%, bias: +3.47%, 95%-confidence interval [-.86, 7.79%], RT-FB mean: 57.9 ± 10.6%, bias: +1.45%, [-3.02, 5.91%], Cartesian mean: 54.9 ± 6.7%).

CONCLUSION: The introduced real-time MR imaging technique enables high-quality cardiac cine data acquisitions in 1-2minutes, eliminating the need for ECG gating and breath-holds. This approach offers a promising alternative to the current clinical practice of segmented acquisition, with shorter scan times, improved patient comfort, and increased robustness to arrhythmia and patient non-compliance.

PMID:39864743 | DOI:10.1016/j.jocmr.2025.101844

Osteoarthritis Cartilage. 2025 Jan 24:S1063-4584(25)00018-4. doi: 10.1016/j.joca.2025.01.001. Online ahead of print.

ABSTRACT

OBJECTIVE: Knee-adjacent subcutaneous fat (kaSCF) has emerged as a potential biomarker and risk factor for OA progression. This study aims to develop an AI-based tool for the automatic segmentation of kaSCF thickness and evaluate the cross-sectional associations between kaSCF, cartilage thickness, MRI-based cartilage T2 relaxation time, knee pain, and muscle strength independent of BMI.

DESIGN: Baseline 3.0T MR images of the right knee from the entire Osteoarthritis Initiative (OAI) cohort (n=4796) were used to quantify average values of kaSCF, cartilage thickness, and T2 using deep learning algorithms. Regression models (adjusted for age, gender, BMI, and race) were used to evaluate the associations between standardized kaSCF and outcomes of cartilage thickness, T2, pain, and knee extension strength.

RESULTS: Model prediction CVs for kaSCF thickness ranged from 3.57% to 9.87% across femoral and tibial regions. Greater average kaSCF was associated with thinner cartilage in men (std. β= -0.029, 95% CI: -0.050 to -0.007, p=0.010) and higher T2 in women (std. β=0.169, 95% CI: 0.072 to 0.265, p=0.001). Greater kaSCF was also associated with lower knee extension force (std. β= -15.36, 95% CI: -20.39 to -10.33, p<0.001) and higher odds of frequent knee pain (std. OR=1.156, 95% CI: 1.046 to 1.278, p=0.005) across all participants.

CONCLUSIONS: Greater kaSCF was associated with thinner cartilage in men, higher T2 in women, reduced knee strength, and greater knee pain, independent of BMI. These findings suggest a potential role of kaSCF as a predictor for KOA-related structural, functional, and clinical outcomes independent of the effects of BMI.

PMID:39864732 | DOI:10.1016/j.joca.2025.01.001

Protein Sci. 2025 Feb;34(2):e70023. doi: 10.1002/pro.70023.

ABSTRACT

Proteins' flexibility is a feature in communicating changes in cell signaling instigated by binding with secondary messengers, such as calcium ions, associated with the coordination of muscle contraction, neurotransmitter release, and gene expression. When binding with the disordered parts of a protein, calcium ions must balance their charge states with the shape of calcium-binding proteins and their versatile pool of partners depending on the circumstances they transmit. Accurately determining the ionic charges of those ions is essential for understanding their role in such processes. However, it is unclear whether the limited experimental data available can be effectively used to train models to accurately predict the charges of calcium-binding protein variants. Here, we developed a chemistry-informed, machine-learning algorithm that implements a game theoretic approach to explain the output of a machine-learning model without the prerequisite of an excessively large database for high-performance prediction of atomic charges. We used the ab initio electronic structure data representing calcium ions and the structures of the disordered segments of calcium-binding peptides with surrounding water molecules to train several explainable models. Network theory was used to extract the topological features of atomic interactions in the structurally complex data dictated by the coordination chemistry of a calcium ion, a potent indicator of its charge state in protein. Our design created a computational tool of CaXML, which provided a framework of explainable machine learning model to annotate ionic charges of calcium ions in calcium-binding proteins in response to the chemical changes in an environment. Our framework will provide new insights into protein design for engineering functionality based on the limited size of scientific data in a genome space.

PMID:39865355 | DOI:10.1002/pro.70023

Life Sci Space Res (Amst). 2025 Feb;44:10-22. doi: 10.1016/j.lssr.2024.11.004. Epub 2024 Nov 22.

ABSTRACT

The space environment presents unique stressors, such as microgravity and space radiation, which can induce molecular and physiological changes in living organisms. To identify key reproducible transcriptomic features and explore potential biological roles in space-flown C. elegans, we integrated transcriptomic data from C. elegans subjected to four spaceflights aboard the International Space Station (ISS) and identified 32 reproducibly differentially expressed genes (DEGs). These DEGs were enriched in pathways related to the structural constituent of cuticle, defense response, unfolded protein response, longevity regulation, extracellular structural organization, and signal receptor regulation. Among these 32 DEGs, 13 genes were consistently downregulated across four spaceflight conditions, primarily associated with the structural constituent of the cuticle. The remaining genes, involved in defense response, unfolded protein response, and longevity regulation pathway, exhibited distinct patterns depending on spaceflight duration: they were downregulated during short-term spaceflights but upregulated during long-term spaceflights. To explore the potential space stressors responsible for these transcriptomic changes, we performed qRT-PCR experiments on C. elegans exposed to simulated microgravity and low-dose radiation. Our results demonstrated that cuticle-related gene expression was significantly downregulated under both simulated microgravity and low-dose radiation conditions. In contrast, almost all genes involved in defense response, unfolded protein response, and longevity regulation pathway were downregulated under simulated microgravity but upregulated under low-dose radiation exposure. These findings suggest that both microgravity and space radiation inhibit cuticle formation; microgravity as the primary stressor inhibit defense response, unfolded protein response, and longevity regulation pathway during short-term spaceflights, while space radiation may promote these processes during long-term spaceflights. In summary, through integrated spaceflight transcriptomic analyses and simulated space experiments, we identified key transcriptomic features and potential biological functions in space-flown C. elegans, shedding light on the space stressors responsible for these changes. This study provides new insights into the molecular and physiological adaptations of C. elegans to spaceflight, highlighting the distinct impacts of microgravity and space radiation.

PMID:39864902 | DOI:10.1016/j.lssr.2024.11.004

Environ Toxicol Chem. 2025 Jan 6:vgae063. doi: 10.1093/etojnl/vgae063. Online ahead of print.

ABSTRACT

An adverse outcome pathway (AOP) framework maps the sequence of events leading to adverse outcomes from chemical exposures, providing a mechanistic understanding often absent in traditional methods. The quantitative AOP (qAOP) advances AOP by integrating quantitative data and mathematical modeling, thereby providing a more precise comprehension of relationships between molecular initiating events, key events, and adverse outcomes. This review critically examines three primary methodologies: systems toxicology, regression modeling, and Bayesian network modeling, highlighting their strengths, limitations, and specific data requirements within toxicology. Through an analysis of current methodologies and challenges, this review emphasizes the integration of experimental and computational approaches to elucidate key event relationships and proposes strategies for overcoming limitations through standardized protocols and advanced computational tools. By outlining future research directions and the potential of qAOPs to transform chemical risk assessment, this review aims to contribute to the advancement of regulatory science and the protection of public health and the environment.

PMID:39864436 | DOI:10.1093/etojnl/vgae063

Int J Med Inform. 2025 Jan 17;196:105797. doi: 10.1016/j.ijmedinf.2025.105797. Online ahead of print.

ABSTRACT

BACKGROUND: ANCA-associated vasculitis (AAV) is a rare but serious disease. Traditional case-identification methods using claims data can be time-intensive and may miss important subgroups. We hypothesized that a deep learning model analyzing electronic health records (EHR) can more accurately identify AAV cases.

METHODS: We examined the Mass General Brigham (MGB) repository of clinical documentation from 12/1/1979 to 5/11/2021, using expert-curated keywords and ICD codes to identify a large cohort of potential AAV cases. Three labeled datasets (I, II, III) were created, each containing note sections. We trained and evaluated a range of machine learning and deep learning algorithms for note-level classification, using metrics like positive predictive value (PPV), sensitivity, F-score, area under the receiver operating characteristic curve (AUROC), and area under the precision and recall curve (AUPRC). The hierarchical attention network (HAN) was further evaluated for its ability to classify AAV cases at the patient-level, compared with rule-based algorithms in 2000 randomly chosen samples.

RESULTS: Datasets I, II, and III comprised 6000, 3008, and 7500 note sections, respectively. HAN achieved the highest AUROC in all three datasets, with scores of 0.983, 0.991, and 0.991. The deep learning approach also had among the highest PPVs across the three datasets (0.941, 0.954, and 0.800, respectively). In a test cohort of 2000 cases, the HAN model achieved a PPV of 0.262 and an estimated sensitivity of 0.975. Compared to the best rule-based algorithm, HAN identified six additional AAV cases, representing 13% of the total.

CONCLUSION: The deep learning model effectively classifies clinical note sections for AAV diagnosis. Its application to EHR notes can potentially uncover additional cases missed by traditional rule-based methods.

PMID:39864108 | DOI:10.1016/j.ijmedinf.2025.105797

Cell Rep. 2025 Jan 25;44(2):115241. doi: 10.1016/j.celrep.2025.115241. Online ahead of print.

ABSTRACT

Hematopoietic stem cells (HSCs) possess the capacity to regenerate the entire hematopoietic system. However, the precise HSC dynamics in the early post-transplantation phase remain an enigma. Clinically, the initial hematopoiesis in the post-transplantation period is critical, necessitating strategies to accelerate hematopoietic recovery. Here, we uncovered the spatiotemporal dynamics of early active hematopoiesis, "hematopoietic cell inflation," using a highly sensitive in vivo imaging system. Hematopoietic cell inflation occurs in three peaks in the spleen after transplantation, with common myeloid progenitors (CMPs), notably characterized by HSC-like signatures, playing a central role. Leveraging these findings, we developed expanded CMPs (exCMPs), which exhibit a gene expression pattern that selectively proliferates in the spleen and promotes hematopoietic expansion. Moreover, universal exCMPs supported early hematopoiesis in allogeneic transplantation. Human universal exCMPs have the potential to be a viable therapeutic enhancement for all HSC transplant patients.

PMID:39864058 | DOI:10.1016/j.celrep.2025.115241

Radiol Med. 2025 Jan 26. doi: 10.1007/s11547-025-01957-5. Online ahead of print.

ABSTRACT

PURPOSE: To develop an artificial intelligence (AI) algorithm for automated measurements of spinopelvic parameters on lateral radiographs and compare its performance to multiple experienced radiologists and surgeons.

METHODS: On lateral full-spine radiographs of 295 consecutive patients, a two-staged region-based convolutional neural network (R-CNN) was trained to detect anatomical landmarks and calculate thoracic kyphosis (TK), lumbar lordosis (LL), sacral slope (SS), and sagittal vertical axis (SVA). Performance was evaluated on 65 radiographs not used for training, which were measured independently by 6 readers (3 radiologists, 3 surgeons), and the median per measurement was set as the reference standard. Intraclass correlation coefficient (ICC), mean absolute error (MAE), and standard deviation (SD) were used for statistical analysis; while, ANOVA was used to search for significant differences between the AI and human readers.

RESULTS: Automatic measurements (AI) showed excellent correlation with the reference standard, with all ICCs within the range of the readers (TK: 0.92 [AI] vs. 0.85-0.96 [readers]; LL: 0.95 vs. 0.87-0.98; SS: 0.93 vs. 0.89-0.98; SVA: 1.00 vs. 0.99-1.00; all p < 0.001). Analysis of the MAE (± SD) revealed comparable results to the six readers (TK: 3.71° (± 4.24) [AI] v.s 1.86-5.88° (± 3.48-6.17) [readers]; LL: 4.53° ± 4.68 vs. 2.21-5.34° (± 2.60-7.38); SS: 4.56° (± 6.10) vs. 2.20-4.76° (± 3.15-7.37); SVA: 2.44 mm (± 3.93) vs. 1.22-2.79 mm (± 2.42-7.11)); while, ANOVA confirmed no significant difference between the errors of the AI and any human reader (all p > 0.05). Human reading time was on average 139 s per case (range: 86-231 s).

CONCLUSION: Our AI algorithm provides spinopelvic measurements accurate within the variability of experienced readers, but with the potential to save time and increase reproducibility.

PMID:39864034 | DOI:10.1007/s11547-025-01957-5

Am J Hum Genet. 2025 Feb 6;112(2):428-449. doi: 10.1016/j.ajhg.2025.01.002. Epub 2025 Jan 24.

ABSTRACT

More than 50% of families with suspected rare monogenic diseases remain unsolved after whole-genome analysis by short-read sequencing (SRS). Long-read sequencing (LRS) could help bridge this diagnostic gap by capturing variants inaccessible to SRS, facilitating long-range mapping and phasing and providing haplotype-resolved methylation profiling. To evaluate LRS's additional diagnostic yield, we sequenced a rare-disease cohort of 98 samples from 41 families, using nanopore sequencing, achieving per sample ∼36× average coverage and 32-kb read N50 from a single flow cell. Our Napu pipeline generated assemblies, phased variants, and methylation calls. LRS covered, on average, coding exons in ∼280 genes and ∼5 known Mendelian disease-associated genes that were not covered by SRS. In comparison to SRS, LRS detected additional rare, functionally annotated variants, including structural variants (SVs) and tandem repeats, and completely phased 87% of protein-coding genes. LRS detected additional de novo variants and could be used to distinguish postzygotic mosaic variants from prezygotic de novos. Diagnostic variants were established by LRS in 11 probands, with diverse underlying genetic causes including de novo and compound heterozygous variants, large-scale SVs, and epigenetic modifications. Our study demonstrates LRS's potential to enhance diagnostic yield for rare monogenic diseases, implying utility in future clinical genomics workflows.

PMID:39862869 | DOI:10.1016/j.ajhg.2025.01.002

Magn Reson Imaging. 2025 Jan 23:110334. doi: 10.1016/j.mri.2025.110334. Online ahead of print.

ABSTRACT

Magnetic Resonance Imaging is a cornerstone of medical diagnostics, providing high-quality soft tissue contrast through non-invasive methods. However, MRI technology faces critical limitations in imaging speed and resolution. Prolonged scan times not only increase patient discomfort but also contribute to motion artifacts, further compromising image quality. Compressed Sensing (CS) theory has enabled the acquisition of partial k-space data, which can then be effectively reconstructed to recover the original image using advanced reconstruction algorithms. Recently, deep learning has been widely applied to MRI reconstruction, aiming to reduce the artifacts in the image domain caused by undersampling in k-space and enhance image quality. As deep learning continues to evolve, the undersampling factors in k-space have gradually increased in recent years. However, these layers are limited in compensating for reconstruction errors in the unsampled areas, impeding further performance improvements. To address this, we propose a learnable spatial-frequency difference-aware module that complements the learnable data consistency layer, mapping k-space domain differences to the spatial image domain for perceptual compensation. Additionally, inspired by wavelet decomposition, we introduce explicit priors by decomposing images into mean and residual components, enforcing a refined zero-mean constraint on the residuals while maintaining computational efficiency. Comparative experiments on the FastMRI and Calgary-Campinas datasets demonstrate that our method achieves superior reconstruction performance against seven state-of-the-art techniques. Ablation studies further confirm the efficacy of our model's architecture, establishing a new pathway for enhanced MRI reconstruction.

PMID:39863026 | DOI:10.1016/j.mri.2025.110334

Neuroimage. 2025 Jan 23:121054. doi: 10.1016/j.neuroimage.2025.121054. Online ahead of print.

ABSTRACT

Magnetic resonance electrical properties tomography can extract the electrical properties of in-vivo tissue. To estimate tissue electrical properties, various reconstruction algorithms have been proposed. However, physics-based reconstructions are prone to various artifacts such as noise amplification and boundary artifact. Deep learning-based approaches are robust to these artifacts but need extensive training datasets and suffer from generalization to unseen data. To address these issues, we introduce a joint three-plane physics-constrained deep learning framework for polynomial fitting MR-EPT by merging physics-based weighted polynomial fitting with deep learning. Within this framework, deep learning is used to discern the optimal polynomial fitting weights for a physics based polynomial fitting reconstruction on the complex B1+ data. For the prediction of optimal fitting coefficients, three neural networks were separately trained on simulated heterogeneous brain models to predict optimal polynomial weighting parameters in three orthogonal planes. Then, the network weights were jointly optimized to estimate the polynomial weights in each plane for a combined conductivity reconstruction. Based on this physics-constrained deep learning approach, we achieved an improvement of conductivity estimation accuracy in comparison to a single plane estimation and a reduction of computational load. The results demonstrate that the proposed method based on 3D data exhibits superior performance in comparison to conventional polynomial fitting methods in terms of capturing anatomical detail and homogeneity. Crucially, in-vivo application of the proposed method showed that the method generalizes well to in-vivo data, without introducing significant errors or artifacts. This generalization makes the presented method a promising candidate for use in clinical applications.

PMID:39863005 | DOI:10.1016/j.neuroimage.2025.121054

Mar Pollut Bull. 2025 Jan 24;212:117583. doi: 10.1016/j.marpolbul.2025.117583. Online ahead of print.

ABSTRACT

Marine pollution due to oil spills presents major risks to coastal areas and aquatic life, leading to serious environmental health concerns. Oil Spill detection using SAR data has transitioned from traditional segmentation to a variety of machine learning & deep learning models like UNET proving its efficiency for the task. This research paper proposes a GSCAT-UNET model for efficient oil spill detection and discrimination from lookalikes. The GSCAT-UNET is an advanced UNET architecture comprising of Spatial-Channel Attention Gates(SCAG), Three Level Attention Module(TLM) and Global Feature Module(GFM) for global level oil spill feature enhancement leading to effective oil spill detection and discrimination from lookalikes. Sentinel-1 Dual-Pol SAR dataset of 1112 images and respective labeled images (5 classes) including confirmed oil spills and lookalikes is used to demonstrate the efficacy of the GSCAT-UNET model. The GSCAT-UNET model significantly enhances segmentation accuracy and robustness for oil spill detection with 5% higher accuracy and 29% higher IoU i.e. 93.7% compared to the UNET segmentation model, addressing the challenges of SAR data complexities and imbalanced datasets. The strong performance of the GSCAT-UNET model demonstrates its potential as a critical tool for disaster response and environmental monitoring.

PMID:39862681 | DOI:10.1016/j.marpolbul.2025.117583

J Electrocardiol. 2025 Jan 22;89:153882. doi: 10.1016/j.jelectrocard.2025.153882. Online ahead of print.

ABSTRACT

PURPOSE OF REVIEW: WHO defines SCD as sudden unexpected death either within 1 h of symptom onset (witnessed) or within 24 h of having been observed alive and symptom-free (unwitnessed). Sudden cardiac arrest is a major cause of mortality worldwide, with survival to hospital discharge for hospital cardiac arrest and in-hospital cardiac arrest being only 9.3 % and 21.2 %, respectively, despite treatment highlighting the importance of effectively predicting and preventing cardiac arrest. This literature review aims to explore the role and application of AI (Artificial Intelligence) in predicting and preventing sudden cardiac arrest.

MATERIAL AND METHODS: Eligible studies were searched from PubMed and Web of Science. The inclusion criteria were fulfilled if sudden cardiac death prediction and prevention, artificial intelligence, machine learning, and deep learning were included.

CONCLUSIONS: Artificial intelligence, machine learning, and deep learning have shown remarkable prospects in SCA risk stratification, which can improve the survival rate from SCA. Nonetheless, they have not been adequately trained and tested, necessitating further studies with explainable techniques, larger sample sizes, external validation, more diverse patient samples, multimodal tools, ethics, and bias mitigation to unlock their full potential.

PMID:39862597 | DOI:10.1016/j.jelectrocard.2025.153882

Int J Med Inform. 2025 Jan 21;196:105793. doi: 10.1016/j.ijmedinf.2025.105793. Online ahead of print.

ABSTRACT

BACKGROUND: In the context of routine breast cancer diagnosis, the precise discrimination between benign and malignant breast masses holds utmost significance. Notably, few prior investigations have concurrently explored the integration of imaging histology features, deep learning characteristics, and clinical parameters. The primary objective of this retrospective study was to pioneer a multimodal feature fusion model tailored for the prediction of breast tumor malignancy, harnessing the potential of ultrasound images.

METHOD: We compiled a dataset that included clinical features from 1065 patients and 3315 image datasets. Specifically, we selected data from 603 patients for training our multimodal model. The comprehensive experimental workflow involves identifying the optimal unimodal model, extracting unimodal features, fusing multimodal features, gaining insights from these fused features, and ultimately generating prediction results using a classifier.

RESULTS: Our multimodal feature fusion model demonstrates outstanding performance, achieving an AUC of 0.994 (95 % CI: 0.988-0.999) and an F1 score of 0.971 on the primary multicenter dataset. In the evaluation on two independent testing cohorts (TCs), it maintains strong performance, with AUCs of 0.942 (95 % CI: 0.854-0.994) for TC1 and 0.945 (95 % CI: 0.857-1.000) for TC2, accompanied by corresponding F1 scores of 0.872 and 0.857, respectively. Notably, the decision curve analysis reveals that our model achieves higher accuracy within the threshold probability range of approximately [0.210, 0.890] (TC1) and [0.000, 0.850] (TC2) compared to alternative methods. This capability enhances its utility in clinical decision-making, providing substantial benefits.

CONCLUSION: The multimodal model proposed in this paper can comprehensively evaluate patients' multifaceted clinical information, achieve the prediction of benign and malignant breast ultrasound tumors, and obtain high performance indexes.

PMID:39862564 | DOI:10.1016/j.ijmedinf.2025.105793