Biophys J. 2025 Feb 18:S0006-3495(25)00107-9. doi: 10.1016/j.bpj.2025.02.016. Online ahead of print.

ABSTRACT

Humans have three known ATP-binding cassette (ABC) transporters in the inner mitochondrial membrane (ABCB7, ABCB8, and ABCB10). ABCB10, the most studied of them thus far, is essential for normal red blood cell development and protection against oxidative stress, and it was recently found to export biliverdin, a heme degradation product with antioxidant properties. The molecular mechanism underlying the function of ABC-transporters remains controversial. Their nucleotide binding domains (NBDs) must dimerize to hydrolyze ATP, but capturing the transporters in such conformation for structural studies has been experimentally difficult, especially for ABCB10 and related eukaryotic transporters. Purified transporters are commonly studied in detergent micelles, or after their reconstitution in nanodiscs, usually at non-physiological temperature and using non-hydrolysable ATP analogs or mutations that prevent ATP hydrolysis. Here, we have used Luminescence Resonance Energy Transfer (LRET) to evaluate the effect of experimental conditions on the NBDs dimerization of ABCB10. Our results indicate that all conditions used for determination of currently available ABCB10 structures have failed to induce NBDs dimerization. ABCB10 in detergent responded only to MgATP at 37oC, whereas reconstituted protein shifted towards dimeric NBDs more easily, including in response to MgAMP-PNP and even present NBDs dimerization with MgATP at room temperature. The nanodisc's size affects the nucleotide-free conformational equilibrium of ABCB10 and the response to ATP in absence of magnesium, but for all analyzed sizes (scaffold proteins MSP1D1, MSP1E3D1, and MSP2N2), a conformation with dimeric NBDs is clearly preferred during active ATP hydrolysis (MgATP, 37oC). These results highlight the sensitivity of this human ABC transporter to experimental conditions and the need for a more cautious interpretation of structural models obtained under far from physiological conditions. A dimeric NBDs conformation that has been elusive in prior studies, seems to be dominant during MgATP hydrolysis at physiological temperature.

PMID:39973007 | DOI:10.1016/j.bpj.2025.02.016

Phys Rev E. 2025 Jan;111(1-1):014420. doi: 10.1103/PhysRevE.111.014420.

ABSTRACT

The folding of cellular monolayers pervades embryonic development and disease, and is often caused by cell shape changes such as cell wedging. However, the function and mechanical role of different active cellular changes in different regions of folding tissues remain unclear in many cases, at least partially because the quantification of out-of-plane mechanical stresses in complex three-dimensional tissues has proved challenging. The gastrulationlike inversion process of the green alga Volvox provides a unique opportunity to overcome this difficulty: Combining laser ablation experiments and a mechanical model, we infer the mechanical properties of the curved tissue from its unfurling on ablation. We go on to reproduce the tissue shapes at different developmental timepoints quantitatively using our mechanical model. Strikingly, this reveals out-of-plane stresses associated with additional cell shape changes away from those regions where cell wedging bends the tissue. Moreover, the fits indicate an adaptive response of the tissue to these stresses. In this way, our paper provides not only the experimental and theoretical framework to quantify out-of-plane stresses in tissue folding, but it also shows how additional cell shape changes can provide another source of out-of-plane stresses in development complementing cell wedging.

PMID:39972828 | DOI:10.1103/PhysRevE.111.014420

Equine Vet J. 2025 Feb 19. doi: 10.1111/evj.14490. Online ahead of print.

ABSTRACT

BACKGROUND: Equine osteoarthritis (OA) is predominantly diagnosed through clinical examination and radiography, leading to detection only after significant joint pathology. The pathogenesis of OA remains unclear and while many medications modify the disease's inflammatory components, no curative or reversal treatments exist. Identifying differentially abundant metabolites and proteins correlated with osteoarthritis severity could improve early diagnosis, track disease progression, and evaluate responses to interventions.

OBJECTIVES: To identify molecular markers of osteoarthritis severity based on histological and macroscopic grading.

STUDY DESIGN: Cross-sectional study.

METHODS: Post-mortem synovial fluid was collected from 58 Thoroughbred racehorse joints and 83 joints from mixed breeds. Joints were histologically and macroscopically scored and categorised by OA and synovitis grade. Synovial fluid nuclear magnetic resonance metabolomic and mass spectrometry proteomic analyses were performed, individually and combined.

RESULTS: In Thoroughbreds, synovial fluid concentrations of metabolites 2-aminobutyrate, alanine and creatine were elevated for higher OA grades, while glutamate was reduced for both Thoroughbreds and mixed breeds. In mixed breeds, concentrations of three uncharacterised proteins, lipopolysaccharide binding protein and immunoglobulin kappa constant were lower for higher OA grades; concentrations of an uncharacterised protein were higher for OA grade 1 only, and apolipoprotein A1 concentrations were higher for OA grades 1 and 2 compared with lower grades. For Thoroughbreds, gelsolin concentrations were lower for higher OA grades, and afamin was lower at a higher synovitis grade. Correlation analyses of combined metabolomics and proteomics datasets revealed 58 and 32 significant variables for Thoroughbreds and mixed breeds, respectively, with correlations from -0.48 to 0.42 and -0.44 to 0.49.

MAIN LIMITATIONS: The study's reliance on post-mortem assessments limits correlation with clinical osteoarthritis severity.

CONCLUSIONS: Following stratification of equine OA severity through histological and macroscopic grading, synovial fluid metabolomic and proteomic profiling identified markers that may support earlier diagnosis and progression tracking. Further research is needed to correlate these markers with clinical osteoarthritis severity.

PMID:39972657 | DOI:10.1111/evj.14490

ACS Nano. 2025 Feb 19. doi: 10.1021/acsnano.4c16949. Online ahead of print.

ABSTRACT

Fast and cost-effective microbial classification is crucial for clinical diagnosis, environmental monitoring, and food safety. However, traditional methods encounter challenges including intricate procedures, skilled personnel needs, and sophisticated instrumentations. Here, we propose a cost-effective microbe classification system, also termed freeze-thaw-induced floating pattern of AuNPs (FTFPA), coupled with artificial intelligence, which is capable of identifying microbes at a cost of $0.0023 per sample. Specifically, FTFPA utilizes AuNPs for coincubation with microbes, resulting in distinct patterns upon freeze-thawing due to their weak interaction. These patterns are digitized to train models that distinguish nine microbes in various tasks. The positive sample detection model achieved an F1 score of 0.976 (n = 194), while the multispecies classification task reached a macro F1 score of 0.859 (n = 1728). To address scalability and lightweight requirements across diverse classification scenarios, we categorized microbes based on species classification levels. The macro F1 score of the hierarchical model (n = 5184), order level model (n = 5184), Enterobacteriales level model (n = 2550), and Bacillales level model (n = 1974) was 0.854, 0.907, 0.958, and 0.843. In summary, our method is user-friendly, requiring only simple equipment, is easy to operate, and convenient, providing a platform for microbial identification.

PMID:39972564 | DOI:10.1021/acsnano.4c16949

bioRxiv [Preprint]. 2025 Jan 27:2025.01.24.634771. doi: 10.1101/2025.01.24.634771.

ABSTRACT

Synthetic prostaglandin analogues, such as latanoprost, are first-line treatments to reduce intraocular pressure (IOP) in the management of glaucoma, treating millions of patients daily. Glaucoma is a leading cause of blindness, characterized by progressive optic neuropathy, with elevated IOP being the sole modifiable risk factor. Despite this importance, the underlying latanoprost mechanism is still not well defined, being associated with both acute and long term activities, and ocular side effects. Prostaglandins are eicosanoid lipid mediators. Yet, there has not been a comprehensive assessment of small lipid mediators in glaucomatous eyes. Here we performed a lipidomic screen of aqueous humour sampled from glaucoma patients or healthy control eyes. The resulting signature was surprisingly focused on significantly elevated levels of arachidonic acid (AA) and the potent proresolving mediator, lipoxin A 4 (LXA 4 ) in glaucoma eyes. Subsequent experiments revealed that this response is due to latanoprost actions, rather than a consequence of elevated IOP. We demonstrated that increased LXA 4 inhibits pro-inflammatory cues and promotes TGF-β 3 mediated tissue remodeling in the anterior chamber. In concert, an autocrine prostaglandin circuit mediates rapid IOP-lowering. This work reveals parallel mechanisms underlying acute and long-term latanoprost activities during the treatment of glaucoma.

PMID:39975338 | PMC:PMC11838192 | DOI:10.1101/2025.01.24.634771

Front Neurol. 2025 Feb 5;16:1530438. doi: 10.3389/fneur.2025.1530438. eCollection 2025.

ABSTRACT

BACKGROUND: Widespread use of ocrelizumab, an anti-CD20 monoclonal antibody, for treating patients with multiple sclerosis (MS), has led to an increase in reported adverse events following real-world observation. Among these, drug-induced colitis is a rare, but severe side effect, prompting a recent FDA statement regarding this safety concern. Objectives: We analyzed a cohort of ocrelizumab treated patients in our MS center to evaluate the incidence of drug-induced colitis.

METHODS: We present a critical review of the available literature on diagnosis and management of anti-CD20 induced colitis and display a case series of 3 suspected patients in our cohort.

RESULTS: Two patients met the full criteria for ocrelizumab-induced colitis, while a third partially met the criteria. Following symptomatic treatment and discontinuation of ocrelizumab, the patients showed favorable outcomes.

CONCLUSION: Ocrelizumab-induced colitis is a rare, but severe adverse event. Its incidence may be higher than expected, reaching 1,95% in our cohort of MS patients. Further reporting of such cases is essential to broaden our understanding of this side effect.

PMID:39974366 | PMC:PMC11835689 | DOI:10.3389/fneur.2025.1530438

Int J Risk Saf Med. 2024 Nov;35(4):317-333. doi: 10.1177/09246479241292008. Epub 2024 Nov 6.

ABSTRACT

Background: Of many pharmaceutical products launched for the benefit of humanity, a significant number have had to be recalled from the marketplace due to adverse events. A systematic review found market recalls for 462 pharmaceutical products between 1953 and 2013. In our current and remarkable period of medical history, excess mortality figures are high in many countries. Yet these statistics receive limited attention, often ignored or dismissed by mainstream news outlets. This excess mortality may include adverse effects caused by novel pharmaceutical agents that use gene-code technology.Objective: To examine key pharmaceutical product withdrawals and derive lessons that inform the current use of gene-based COVID-19 vaccines.Methods: Selective narrative review of historical pharmaceutical recalls and comparative issues with recent COVID-19 vaccines.Results: Parallels with past drug withdrawals and gene-based vaccines include distortion of clinical trial data, with critical adverse event data absent from high-impact journal publications. Delayed regulatory action on pharmacovigilance data to trigger market withdrawal occurred with Vioxx (rofecoxib) and is apparent with the gene-based COVID-19 vaccines.Conclusion: Public health requires access to raw clinical trial data, improved transparency from corporations and heightened, active pharmacovigilance worldwide.

PMID:39973420 | DOI:10.1177/09246479241292008

J Acquir Immune Defic Syndr. 2025 Feb 20. doi: 10.1097/QAI.0000000000003648. Online ahead of print.

ABSTRACT

BACKGROUND: Leronlimab is a humanized κ-IgG4 monoclonal antibody that blocks C-C chemokine receptor type 5 (CCR5). We investigated leronlimab as a treatment option for people living with multidrug-resistant HIV-1.

SETTING: and Methods: In a phase 2b/3, multicenter, randomized, double-blind, placebo-controlled study conducted in 21 hospital centers in the United States, treatment-experienced people living with HIV (PLWH) with documented drug resistance were randomly assigned once weekly leronlimab (350 mg subcutaneously) or matching placebo for one week overlapping existing failing antiretroviral treatment (ART), followed by a 24-week single-arm extension with weekly leronlimab combined with a new optimized background treatment (OBT). The primary endpoint was achieving ≥0.5 log10 reduction in plasma HIV-1 RNA from baseline at the end of the one-week double-blinded treatment period.

RESULTS: 52 participants were enrolled (25 leronlimab and 27 placebo). After the one-week randomized phase, by the intent-to-treat analysis 64.0% (16/25) receiving leronlimab achieved ≥0.5 log10 reduction in plasma HIV-1 RNA versus 23.1% (6/26) receiving placebo (p=0.0032), while by per protocol analysis 72.7% (16/22) receiving leronlimab achieved ≥0.5 log10 reduction in plasma HIV-1 RNA versus 24.0% (6/25) receiving placebo (p=0.0008). Leronlimab was generally well tolerated with no drug-related SAEs reported. Overall, 175 adverse events were reported by 34/52 participants, with 120 (68.6%) adverse events categorized as mild.

CONCLUSIONS: Leronlimab resulted in significantly reduced plasma HIV-1 within one week after addition to failing ART. After 24 weeks combined with an OBT, most participants had plasma HIV-1 RNA levels <50 copies per mL plasma, suggesting utility of leronlimab as a component of salvage therapy.

PMID:39972543 | DOI:10.1097/QAI.0000000000003648

ArXiv [Preprint]. 2025 Jan 20:arXiv:2411.15418v2.

ABSTRACT

Virtual screening of small molecules against protein targets can accelerate drug discovery and development by predicting drug-target interactions (DTIs). However, structure-based methods like molecular docking are too slow to allow for broad proteome-scale screens, limiting their application in screening for off-target effects or new molecular mechanisms. Recently, vector-based methods using protein language models (PLMs) have emerged as a complementary approach that bypasses explicit 3D structure modeling. Here, we develop SPRINT, a vector-based approach for screening entire chemical libraries against whole proteomes for DTIs and novel mechanisms of action. SPRINT improves on prior work by using a self-attention based architecture and structure-aware PLMs to learn drug-target co-embeddings for binder prediction, search, and retrieval. SPRINT achieves SOTA enrichment factors in virtual screening on LIT-PCBA, DTI classification benchmarks, and binding affinity prediction benchmarks, while providing interpretability in the form of residue-level attention maps. In addition to being both accurate and interpretable, SPRINT is ultra-fast: querying the whole human proteome against the ENAMINE Real Database (6.7B drugs) for the 100 most likely binders per protein takes 16 minutes. SPRINT promises to enable virtual screening at an unprecedented scale, opening up new opportunities for in silico drug repurposing and development. SPRINT is available on the web as ColabScreen: https://bit.ly/colab-screen.

PMID:39975427 | PMC:PMC11838698

ACS Pharmacol Transl Sci. 2025 Jan 23;8(2):308-338. doi: 10.1021/acsptsci.4c00679. eCollection 2025 Feb 14.

ABSTRACT

Skin cancer represents a major health concern due to its rising incidence and limited treatment options. Current treatments (surgery, chemotherapy, radiotherapy, immunotherapy, and targeted therapy) often entail high costs, patient inconvenience, significant adverse effects, and limited therapeutic efficacy. The search for novel treatment options is also marked by the high capital investment and extensive development involved in the drug discovery process. In response to these challenges, repurposing existing drugs for topical application and optimizing their delivery through nanotechnology could be the answer. This innovative strategy aims to combine the advantages of the known pharmacological background of commonly used drugs to expedite therapeutic development, with nanosystem-based formulations, which among other advantages allow for improved skin permeation and retention and overall higher therapeutic efficacy and safety. The present review provides a critical analysis of repurposed drugs such as doxycycline, itraconazole, niclosamide, simvastatin, leflunomide, metformin, and celecoxib, formulated into different nanosystems, namely, nanoemulsions and nanoemulgels, nanodispersions, solid lipid nanoparticles, nanostructured lipid carriers, polymeric nanoparticles, hybrid lipid-polymer nanoparticles, hybrid electrospun nanofibrous scaffolds, liposomes and liposomal gels, ethosomes and ethosomal gels, and aspasomes, for improved outcomes in the battle against skin cancer. Enhanced antitumor effects on melanoma and nonmelanoma research models are highlighted, with some nanoparticles even showing intrinsic anticancer properties, leading to synergistic effects. The explored research findings highly evidence the potential of these approaches to complement the currently available therapeutic strategies in the hope that these treatments might one day reach the pharmaceutical market.

PMID:39974652 | PMC:PMC11833728 | DOI:10.1021/acsptsci.4c00679

In Silico Pharmacol. 2025 Feb 17;13(1):30. doi: 10.1007/s40203-025-00316-6. eCollection 2025.

ABSTRACT

In COVID-19 patients, respiratory failure was reported due to damage to the respiratory centers of the brainstem. Molecular mimicry of three brainstem pre-Botzinger complex proteins (DAB1, AIFM and SURF1) was regarded as the underlying reason for respiratory failure and the autoimmune neurological sequelae. Of the three brainstem proteins mimicked by SARS CoV-2, corresponding sequences to two of the mimicry peptides were located in the N-protein of SARS CoV-2. N-protein is important for viral RNA synthesis and genome packaging. Here, we have used molecular modeling, docking and MD simulations to discern potential drugs which can inhibit molecular mimicry of DAB1 by SARS CoV-2 and also eliminate it by interfering in genome packaging. The binding site (drug target) for molecular docking was defined as the amino acid sequence extending from position 168-185 of the N-protein which was a SLiM region and also included the mimicry hexapeptide. Molecular docking after MD simulations was used to discern probable inhibitors of the drug-target from FDA-approved neurological drugs in the Broad Institute's Drug Repurposing Hub. Our results revealed that an anti-anxiety drug afobazole qualified the ADMET parameters, formed a stable complex with the drug-target and exhibited the highest binding energy (-88.21 kJ/mol). This suggests that afobazole can be repurposed against SARS CoV-2 for disrupting molecular mimicry of human DAB1 protein and also eliminate the etiopathological agent by interfering in viral genome packaging.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40203-025-00316-6.

PMID:39974371 | PMC:PMC11832858 | DOI:10.1007/s40203-025-00316-6

Arterioscler Thromb Vasc Biol. 2025 Feb 20. doi: 10.1161/ATVBAHA.124.321579. Online ahead of print.

ABSTRACT

BACKGROUND: Thiol isomerases play essential and nonredundant roles in platelet activation, aggregation, and thrombus formation. Thiol isomerase inhibitors have the potential to overcome the 2 major drawbacks of current antithrombotic therapies, as they target both arterial and venous thrombosis without enhancing bleeding risks. Recently, a Food and Drug Administration-approved drug, zafirlukast (ZAF), was shown to be a promising pan-thiol isomerase inhibitor. The objective of this study is to develop analogues of ZAF with optimized thiol isomerase inhibition and antithrombotic activity.

METHODS: Thirty-five ZAF analogues were tested in an insulin turbidometric assay for thiol isomerase inhibition. Analogues were tested for platelet activation, aggregation, P-selectin expression, and laser-induced thrombosis in mice and compared with the parent compound.

RESULTS: Of the 35 analogues, 12 retained activity, with 1, compound 21, that demonstrated a greater potency than that of ZAF, 5 had a similar potency to that of ZAF, and 6 had a weaker potency. Analogues demonstrated inhibition of platelet aggregation and P-selectin expression as compared with ZAF, consistent with their potencies. ZAF and compound 21 were shown to be reversible inhibitors of thiol isomerases, and not cytotoxic to cultured, lung, liver, and kidney cells. Finally, in an in vivo assessment of thrombus formation, compound 21 was able to significantly inhibit thrombus formation without affecting bleeding times.

CONCLUSIONS: A ZAF analogue, compound 21, with properties superior to those of ZAF was synthesized, demonstrating improved inhibition of platelet activation, aggregation, and thrombus formation as compared with the parent ZAF. This approach could yield a promising clinical candidate for treatment and prophylaxis of arterial and venous thrombosis.

PMID:39973747 | DOI:10.1161/ATVBAHA.124.321579

Infect Dis (Lond). 2025 Feb 20:1-8. doi: 10.1080/23744235.2025.2468510. Online ahead of print.

ABSTRACT

BACKGROUND: Tick-borne encephalitis (TBE) is a neurological disease caused by the tick-borne encephalitis virus (TBEV). Despite available vaccines, breakthrough infections occur, some fatal.

OBJECTIVES: As no antiviral therapy for TBE is currently approved, this study evaluated the in vitro activity of already licenced remdesivir (RDV) and sofosbuvir (SOF) for possible drug repurposing against TBEV.

METHODS: TBEV was cultured in A549 cells, and the inhibitory effects of RDV (GS-5734), its parent nucleotide GS-441524, and SOF (GS-7977) were assessed.

RESULTS: After 78 h, RDV demonstrated significantly lower EC50 values than SOF (0.14 vs. 11 µM) based on TBEV RNA levels measured by RT-qPCR. RDV also had a lower mean EC50 (0.55 µM) compared to GS-441524 and SOF (>8.9 and 13.1 µM, respectively) using crystal violet staining after 5 days. After 11 passages of TBEV in the presence of RDV, emergence of virus with a higher EC50 (1.32 vs. 0.55 µM) was detected with two mutations (L3122F and Y3278F) in NS5, the viral RNA-dependent RNA polymerase (RdRp), and one substitution in envelope (E) protein (E402G). Similarly, SOF resistance appeared after 20 passages, increasing EC50 values (35.5 vs. 10 µM).

CONCLUSION: RDV exhibits potent in vitro antiviral activity against TBEV via specific targeting of the viral RdRp as confirmed by the emergence of resistance-associated double NS5 substitutions in vitro in the presence of RDV. While the potential in vivo implications of the observed RDV resistance remain to be determined, these in vitro data support further assessment of RDV for the treatment of TBEV infection.

PMID:39973341 | DOI:10.1080/23744235.2025.2468510

Brief Bioinform. 2024 Nov 22;26(1):bbaf054. doi: 10.1093/bib/bbaf054.

ABSTRACT

Combination therapy has become increasingly important for treating complex diseases which often involve multiple pathways and targets. However, experimental screening of drug combinations is costly and time-consuming. The availability of large-scale transcriptomic datasets (e.g. CMap and LINCS) from in vitro drug treatment experiments makes it possible to computationally predict drug combinations with synergistic effects. Towards this end, we developed a computational approach, termed Identification of Drug Combinations via Multi-Set Operations (iDOMO), to predict drug synergy based on multi-set operations of drug and disease gene signatures. iDOMO quantifies the synergistic effect of a pair of drugs by taking into account the combination's beneficial and detrimental effects on treating a disease. We evaluated iDOMO, in a DREAM Challenge dataset with the matched, pre- and post-treatment gene expression data and cell viability information. We further evaluated the performance of iDOMO by concordance index and Spearman correlation on predicting the Highest Single Agency (HSA) synergy scores for four most common cancer types in two large-scale drug combination databases, showing that iDOMO significantly outperformed two existing popular drug combination approaches including the Therapeutic Score and the SynergySeq Orthogonality Score. Application of iDOMO to triple-negative breast cancer (TNBC) identified drug pairs with potential synergistic effects, with the combination of trifluridine and monobenzone being the most synergistic. Our in vitro experiments confirmed that the top predicted drug combination exerted a significant synergistic effect in inhibiting TNBC cell growth. In summary, iDOMO is an effective method for the in silico screening of synergistic drug combinations and will be a valuable tool for the development of novel therapeutics for complex diseases.

PMID:39973082 | DOI:10.1093/bib/bbaf054

Orphanet J Rare Dis. 2025 Feb 20;20(1):72. doi: 10.1186/s13023-025-03571-0.

ABSTRACT

Young patients can be uniquely vulnerable to the impacts of a rare disease, diagnosed in their critical years of identity formation, social development, and planning for the future. Drawing from my journey as both a rare disease patient and a medical student, this essay explores how the rare disease achalasia has shaped my life, alongside the experiences of another young patient, Isobel. Most importantly, this essay highlights the critical role that individual healthcare professionals play in shaping young patients' experiences of their condition. Although diagnosing and managing rare diseases can be challenging due to limited research and awareness, my own experiences demonstrate that individual, intentional changes can have profound impacts. By engaging with and believing young patients, individual healthcare providers can reduce misdiagnoses, alleviate isolation and uncertainty, and ultimately, improve healthcare outcomes for young people with rare diseases.

PMID:39972474 | DOI:10.1186/s13023-025-03571-0

Sci Rep. 2025 Feb 19;15(1):6011. doi: 10.1038/s41598-025-89651-4.

ABSTRACT

Traffic congestion, particularly in rapidly expanding urban centers, significantly impacts the timely delivery of emergency medical services (EMS), where every minute can mean the difference between life and death. Traditional traffic signal control systems often lack real-time adaptability to prioritize emergency vehicles, resulting in delays caused by congestion around ambulances. To address this critical issue, this paper presents an AI-driven real-time traffic management system designed to reduce EMS response times. The proposed solution incorporates three core components: Raspberry Pi-based traffic signal prioritization, deep learning-enabled audio-visual ambulance detection, and an advanced intelligent traffic management framework. For audio detection, raw data is transformed into spectrograms using Mel Frequency Cepstral Coefficients (MFCCs) and classified using a Long Short-Term Memory (LSTM) network. Visual data is processed through a ResNet18 convolutional neural network, pre-trained on ImageNet using inductive transfer learning. The outputs from the auditory and visual streams are integrated using empirical risk minimization, enabling accurate ambulance detection through multimodal data fusion. Performance evaluation demonstrates the effectiveness of the proposed system, achieving 98.3% accuracy in audio classification, 98.1% accuracy in visual classification, and 99% accuracy with the fused model. Additional metrics, including precision, recall, F1-score, and a confusion matrix, confirm the model's reliability. This innovative system has the potential to transform urban traffic networks into intelligent, adaptive systems, reducing delays caused by traffic congestion, enhancing emergency medical care response times, and ultimately saving lives. The framework offers a scalable blueprint for future smart city traffic management solutions, meticulously designed to support urban growth and expansion.

PMID:39971977 | DOI:10.1038/s41598-025-89651-4

Sci Rep. 2025 Feb 19;15(1):6035. doi: 10.1038/s41598-025-89348-8.

ABSTRACT

Diet, stress, genetics, and a sedentary lifestyle may all contribute to heart disease rates. Although recent studies propose comprehensive automated diagnostic systems, these systems tend to focus on one aspect, such as feature selection, prioritization, or predictive accuracy. A more complete approach that considers all of these factors can improve the efficiency of a cardiac prediction system. This study uses an appropriate strategy to overcome potential network design problems, design challenges, overfitting, and lack of robustness that can interfere with system performance. The research introduces an ideally designed deep trust network called ID-DTN to improve system performance. The Ruzzo-Tompa method is used to eliminate noncontributory features. The Seagull Optimization Algorithm (SOA) is introduced to optimize the trust depth network to achieve optimal network design. The study scrutinizes the deep trust network (ID-DTN) and the restricted Boltzmann machine (RBM) and sheds light on the system's operation. This proposal can optimize both network architecture and feature selection, which is the main novelty. The proposed method is analyzed using the below-mentioned metrics: Matthew's correlation coefficient, F1 score, accuracy, sensitivity, specificity, and accuracy. ID-DTN performs well compared to other state-of-the-art methods. The validation results confirm that the proposed method improves the prediction accuracy to 97.11% and provides reliable recommendations for patients with cardiovascular disease.

PMID:39971944 | DOI:10.1038/s41598-025-89348-8

J Imaging Inform Med. 2025 Feb 19. doi: 10.1007/s10278-025-01440-7. Online ahead of print.

ABSTRACT

Segmentation and classification of breast tumors are two critical tasks since they provide significant information for computer-aided breast cancer diagnosis. Combining these tasks leverages their intrinsic relevance to enhance performance, but the variability and complexity of tumor characteristics remain challenging. We propose a novel multi-task deep learning network (NMTNet) for the joint segmentation and classification of breast tumors, which is based on a convolutional neural network (CNN) and U-shaped architecture. It mainly comprises a shared encoder, a multi-scale fusion channel refinement (MFCR) module, a segmentation branch, and a classification branch. First, ResNet18 is used as the backbone network in the encoding part to enhance the feature representation capability. Then, the MFCR module is introduced to enrich the feature depth and diversity. Besides, the segmentation branch combines a lesion region enhancement (LRE) module between the encoder and decoder parts, aiming to capture more detailed texture and edge information of irregular tumors to improve segmentation accuracy. The classification branch incorporates a fine-grained classifier that reuses valuable segmentation information to discriminate between benign and malignant tumors. The proposed NMTNet is evaluated on both ultrasound and magnetic resonance imaging datasets. It achieves segmentation dice scores of 90.30% and 91.50%, and Jaccard indices of 84.70% and 88.10% for each dataset, respectively. And the classification accuracy scores are 87.50% and 99.64% for the corresponding datasets, respectively. Experimental results demonstrate the superiority of NMTNet over state-of-the-art methods on breast tumor segmentation and classification tasks.

PMID:39971818 | DOI:10.1007/s10278-025-01440-7

J Imaging Inform Med. 2025 Feb 19. doi: 10.1007/s10278-025-01446-1. Online ahead of print.

ABSTRACT

The rapid advancements of deep learning technology have revolutionized medical imaging diagnosis. However, training these models is often challenged by label imbalance and the scarcity of certain diseases. Most models fail to recognize multiple coexisting diseases, which are common in real-world clinical scenarios. Moreover, most radiological models rely solely on image data, which contrasts with radiologists' comprehensive approach, incorporating both images and other clinical information such as clinical history and laboratory results. In this study, we introduce a Multimodal Chest X-ray Network (MCX-Net) that integrates chest X-ray images and clinical history texts for multi-label disease diagnosis. This integration is achieved by combining a pretrained text encoder, a pretrained image encoder, and a pretrained image-text cross-modal encoder, fine-tuned on the public MIMIC-CXR-JPG dataset, to diagnose 13 diverse lung diseases on chest X-rays. As a result, MCX-Net achieved the highest macro AUROC of 0.816 on the test set, significantly outperforming unimodal baselines such as ViT-base and ResNet152, which scored 0.747 and 0.749, respectively (p < 0.001). This multimodal approach represents a substantial advancement over existing image-based deep-learning diagnostic systems for chest X-rays.

PMID:39971817 | DOI:10.1007/s10278-025-01446-1

BJOG. 2025 Feb 19. doi: 10.1111/1471-0528.18097. Online ahead of print.

ABSTRACT

Cardiotocography (CTG), introduced in the 1960s, was initially expected to prevent hypoxia-related deaths and neurological injuries. However, more than five decades later, evidence supporting the evidence of intrapartum CTG in preventing neonatal and long-term childhood morbidity and mortality remains inconclusive. At the same time, shortcomings in CTG interpretation have been recognised as important contributory factors to rising caesarean section rates and missed opportunities for timely interventions. An important limitation is its high false-positive rate and poor specificity, which undermines reliably identifying foetuses at risk of hypoxia-related injuries. These shortcomings are compounded by the technology's significant intra- and interobserver variability, as well as the subjective and complex nature of fetal heart rate interpretation. However, human factors and other environmental factors are equally significant. Advancements in fetal heart rate monitoring are crucial to support clinicians in improving health outcomes for newborns and their mothers, while at the same time avoiding unnecessary operative deliveries. These limitations highlight the clinical need to enhance neonatal outcomes while minimising unnecessary interventions, such as instrumental deliveries or caesarean sections. We believe that achieving this requires a paradigm shift from subjective interpretation of complex and nonspecific fetal heart rate patterns to evidence-based, quantifiable solutions that integrate hardware, engineering and clinical perspectives. Such transformation necessitates an international, multidisciplinary effort encompassing the entire continuum of pregnancy care and the broader healthcare ecosystem, with emphasis on well-defined, actionable health outcomes. Achieving this will depend on collaborations between researchers, clinicians, medical device manufacturers and other relevant stakeholders. This expert review paper outlines the most relevant and promising directions for research and strategic initiatives to address current challenges in fetal heart rate monitoring. Key themes include advancements in computerised fetal heart rate monitoring, the application of big data and artificial intelligence, innovations in home and remote monitoring and consideration of human factors.

PMID:39971749 | DOI:10.1111/1471-0528.18097