ACS Meas Sci Au. 2024 Dec 27;5(1):109-119. doi: 10.1021/acsmeasuresciau.4c00077. eCollection 2025 Feb 19.

ABSTRACT

Direct-infusion mass spectrometry (DI-MS) and mass spectrometry imaging (MSI) are powerful techniques for lipidomics research. However, annotating isomeric and isobaric lipids with these methods is challenging due to the absence of chromatographic separation. Recently, cyclic ion mobility mass spectrometry (cIM-MS) has been proposed to overcome this limitation. However, fluctuations in room conditions can affect ion mobility multipass arrival times, potentially reducing annotation confidence. In this study, we developed a multipass arrival time correction method that proved effective across various dates, room temperatures, ion mobility settings, and laboratories using mixtures of reference standards. We observed slight variations in the linear correction lines between lipid and nonlipid molecules, underscoring the importance of choosing appropriate reference molecules. Based on these results, we demonstrated that an accurate multipass arrival time database can be constructed from corrected t 0 and t p for interlaboratory use and can effectively identify isomeric lipids in MSI using only a single measurement. This approach significantly simplifies the identification process compared to determining multipass collision cross-section, which requires multiple measurements that are both sample- and time-intensive for MSI. Additionally, we validated our multipass drift time correction method in shotgun lipidomics analyses of human and mouse serum samples and observed no matrix effect for the analysis. Despite variations in dates, room temperatures, instruments, and ion mobility settings, our approach reduced the mean drift time differences from over 2% to below 0.2%.

PMID:39991034 | PMC:PMC11843504 | DOI:10.1021/acsmeasuresciau.4c00077

bioRxiv [Preprint]. 2025 Feb 15:2025.02.11.637731. doi: 10.1101/2025.02.11.637731.

ABSTRACT

MicroRNA (miRNA) induced silencing complex (miRISC) is the targeting apparatus and arguably rate-limiting step of the miRNA-mediated regulatory subsystem - the major noise reducing though metabolically wasteful mechanism. Recently, we reported that miRISC channels miRNA-mediated regulatory activity back onto their own mRNAs to form negative self-feedback loops, a noise-reduction technique in engineering and synthetic/systems biology. Here, we describe mathematical modeling that predicts mRNA expression noise to correlate negatively with degradation rate (K deg ) and noise reduction by self-feedback control of K deg . We also calculated K deg and expression noise of mRNAs detected in a cutting-edge total-RNA single-cell RNA-seq (scRNA-seq) dataset. As predicted, miRNA-targeted mRNAs exhibited higher K deg values in conjunction with lower inter-cell expression noise. Moreover, as predicted by our self-feedback loop model, miRISC mRNAs (AGO1/2/3 and TNRC6A/B/C) exhibited further reduced expression noise. In short, mathematical-modeling and total-RNA scRNA-seq data-analysis shed insight into operational trade-off between noise reduction and metabolic/energetic expenditure in producing miRNA-targeted mRNAs destined for enhanced degradation and translational inhibition, as well as negative self-feedback loop reinforcement of miRISC - the core of miRNA-mediated noise-reduction subsystem. To our knowledge, this is the first report of concurrent mRNA degradation and expression noise analyses and of noise reduction by self-feedback control of mRNA degradation.

PMID:39990448 | PMC:PMC11844488 | DOI:10.1101/2025.02.11.637731

bioRxiv [Preprint]. 2025 Feb 16:2025.02.13.637616. doi: 10.1101/2025.02.13.637616.

ABSTRACT

The human microbiome plays a crucial role in metabolism and thereby influences health and disease. Constraint-based reconstruction and analysis (COBRA) has proven an attractive framework to generate mechanism-derived hypotheses along the nutrition-host-microbiome-disease axis within the computational systems biology community. Unlike for human, no large-scale visualisation resource for microbiome metabolism has been available to date. To address this gap, we created the MicroMap, a manually curated microbiome metabolic network visualisation, which captures the metabolic content of over a quarter million microbial genome-scale metabolic reconstructions. The MicroMap contains 5,064 unique reactions and 3,499 unique metabolites, including for 98 drugs. The MicroMap allows users to intuitively explore microbiome metabolism, inspect microbial metabolic capabilities, and visualise computational modelling results. Further, the MicroMap shall serve as an educational tool to make microbiome metabolism accessible to broader audiences beyond computational modellers. For example, we utilised the MicroMap to generate a comprehensive collection of 257,429 visualisations, corresponding to the entire scope of our current microbiome reconstruction resources, to enable users to visually compare and contrast the metabolic capabilities for different microbes. The MicroMap seamlessly integrates with the Virtual Metabolic Human (VMH, www.vmh.life) and the COBRA Toolbox (opencobra.github.io), and is freely accessible at the MicroMap dataverse (https://dataverse.harvard.edu/dataverse/micromap), in addition to all the generated reconstruction visualisations.

PMID:39990405 | PMC:PMC11844556 | DOI:10.1101/2025.02.13.637616

Front Cell Dev Biol. 2025 Feb 7;13:1528281. doi: 10.3389/fcell.2025.1528281. eCollection 2025.

NO ABSTRACT

PMID:39989984 | PMC:PMC11842372 | DOI:10.3389/fcell.2025.1528281

Exp Ther Med. 2025 Feb 5;29(4):66. doi: 10.3892/etm.2025.12815. eCollection 2025 Apr.

ABSTRACT

Intracranial tuberculoma represents one of the most severe complications of central nervous system tuberculosis (TB), with an incidence that is relatively low. In cases of intracranial tuberculoma, patients may develop drug toxicity and/or immune reconstitution inflammatory syndrome (IRIS) while receiving anti-TB treatment. The current study presented the case of a seven-year-old female patient with intracranial tuberculoma who developed drug-induced hepatotoxicity and IRIS during the course of treatment. During the follow-up of the patient, anti-TB drug-induced hepatitis developed, which led to the discontinuation of the drug twice. In the seventh month of treatment, cranial MRI showed the progression of tuberculoma lesions. The possibility of IRIS or treatment failure was considered and the treatment was restarted with steroids and non-hepatotoxic anti-TB drugs. With steroid and anti-TB treatment, the lesions regressed almost completely and the neurological deficit regressed. Patients receiving treatment should be followed up closely due to the possible side effects of anti-TB drugs, especially IRIS, which develops as an immune restructuring response during the recovery of the immune system.

PMID:39991721 | PMC:PMC11843197 | DOI:10.3892/etm.2025.12815

Clin Epidemiol. 2025 Feb 17;17:87-104. doi: 10.2147/CLEP.S494215. eCollection 2025.

ABSTRACT

PURPOSE: This study aims to investigate the potential association between non-selective RET kinase inhibitors and thyroid dysfunction (TD) by conducting a pharmacovigilance analysis using data from the US FDA Adverse Event Reporting System (FAERS).

METHODS: Data for non-selective RET MKIs were obtained from the FAERS database, spanning the first quarter of 2015 to the fourth quarter of 2023. Disproportionality analysis was used to quantify the AE signals associated with non-selective RET MKIs and to identify TD AEs. Subgroup analyses and multivariate logistic regressions were used to assess the factors influencing the occurrence of TD AEs. Time-to-onset (TTO) analysis and the Weibull Shape Parameter (WSP) test were also performed.

RESULTS: Descriptive analysis revealed an increasing trend in TD adverse events linked to non-selective RET MKIs, with a notable proportion of serious reactions reported. Disproportionality analysis using ROR, PRR, BCPNN, and EBGM algorithms consistently demonstrated a positive association between Sunitinib, Cabozantinib, and Lenvatinib with TD adverse events. Subgroup analyses highlighted differential susceptibility to TD based on age, gender, and weight, with varying patterns observed for each inhibitor. Logistic regression analyses identified factors independently influencing the occurrence of TD adverse events, emphasizing the importance of age, gender, and weight in patient stratification. Time-to-onset analysis indicated early manifestation of TD adverse events following treatment with non-selective RET MKIs, with a decreasing risk over time.

CONCLUSION: The results of our study indicate a correlation between the use of non-selective RET MKIs and the occurrence of TD AEs. This may provide support for the clinical monitoring and risk identification of non-selective RET MKIs. Nevertheless, further clinical studies are required to substantiate the findings of this study.

PMID:39989882 | PMC:PMC11844211 | DOI:10.2147/CLEP.S494215

Circulation. 2025 Feb 24. doi: 10.1161/CIRCULATIONAHA.124.071954. Online ahead of print.

ABSTRACT

BACKGROUND: To date, the only effective treatment of severe aortic stenosis is valve replacement. With the introduction of transcatheter aortic valve replacement and extending indications to younger patients, the use of bioprosthetic valves (BPVs) has considerably increased. The main inconvenience of BPVs is their limited durability because of mechanisms similar as the fibro-calcifying processes observed in native aortic stenosis. One of the major gaps of the field is to identify therapeutic targets to prevent or slow the fibro-calcifying process leading to severe and symptomatic aortic stenosis. The aim was to identify new targets for anticalcification drugs to prevent aortic and BPV calcification using an unbiased translational approach.

METHODS: Explanted valves were collected from patients and organ donor hearts. A comparative transcriptomic analysis was performed on valvular interstitial cells (VIC) obtained from calcified (bicuspid and tricuspid) versus control valves. The mechanisms and consequences of aldehyde dehydrogenase 1 family member A1 (ALDH1A1) downregulation were analyzed in VIC cultures from control human aortic valves. ALDH1A1 was inhibited or silenced and its impact on osteogenic marker expression and calcification processes assessed in VIC. The effect of all-trans retinoic acid on calcification was tested on human VIC cultures and on 2 animal models: the model of subcutaneous implantation of bovine pericardium in rats and the model of xenograft aortic valve replacement in juvenile sheep.

RESULTS: Transcriptome analysis of human VIC identified ALDHA1 as the most downregulated gene in VIC from calcified versus control valves. In human VIC, ALDH1A1 expression is downregulated by TGF-β1 in a SMAD2/3-dependent manner. ALDH1A1 inhibition promotes an osteoblast-like VIC phenotype and increases calcium deposition through inhibition of retinoic acid receptor alpha signaling. Conversely, VIC treatment with retinoids decreases calcium deposition and attenuates VIC osteoblast activity. Last, all-trans retinoic acid inhibits calcification development of aortic BPV in both in vivo models and improves aortic valve echocardiographic parameters in the xenograft sheep model.

CONCLUSIONS: These results show that ALDH1A1 is downregulated in calcified valves, hence promoting VIC transition into an osteoblastic phenotype. Retinoic acid receptor alpha agonists, including all-trans retinoic acid through a drug repositioning strategy, represent a promising and innovative pharmacological approach to prevent calcification of native aortic valves and BPV.

PMID:39989358 | DOI:10.1161/CIRCULATIONAHA.124.071954

Exp Dermatol. 2025 Feb;34(2):e70060. doi: 10.1111/exd.70060.

ABSTRACT

Globally, chronic wounds impact the health of millions of people, negatively affecting quality of life and healthcare budgets. Some of the crucial steps and pathways in healing mechanisms are the hypoxic response and the expression of host defence peptides, which are decreased in diseases related to chronic wounds such as diabetes mellitus and cardiovascular diseases. It has been shown that histone deacetylase inhibitors can induce the expression of Host Defence Peptides (HDP) by inducing the stabilisation and activation of hypoxia-inducible factor 1-α (HIF-1α), promoting wound healing pathways, although their high cost and side effects limit clinical research. With the help of bioinformatics tools, we found potential histone deacetylase inhibitor candidates in an FDA-approved drugs database. The candidates, 1,3-Diphenylurea (DiPU), 2'-Aminoacetanilide (Ace), and Tert-butyl (2-aminophenyl) carbamate (N-boc), show wound healing effects in HaCaT cells, increasing cell migration possibly via HIF-1α, inducing the expression of LL-37 and vascular endothelial growth factor (VEGF), while in a mouse ring angiogenesis model, Ace and N-boc have angiogenic effects. In a model of basal primary keratinocytes from donors with diabetes mellitus (DM), without DM, and from Diabetic Foot Ulcers (DFU), it was observed that only DiPU is capable of inducing LL-37 in all scenarios. There is limited information about histone deacetylase inhibitors and wound healing but in this paper, we observe promising results and a proposed mechanism that involved specifically Histone Deacetylase 1 inhibition (HDAC1).

PMID:39989310 | DOI:10.1111/exd.70060

Br J Haematol. 2025 Feb 23. doi: 10.1111/bjh.20026. Online ahead of print.

ABSTRACT

We sought to identify new candidate drugs for repurposing to myelodysplastic syndromes (MDS). Connectivity map analysis was performed on gene expression signatures generated from bone marrow CD34+ cells of splicing factor mutant MDS patients. Celastrol and Withaferin A (WA), two top-ranking compounds identified, markedly inhibited proliferation, arrested the cell cycle and induced apoptosis in leukaemia cells. These compounds also inhibited the viability of primary bone marrow MDS cells. We showed that Celastrol and WA inhibit interleukin-1 receptor-associated kinase 4-mediated nuclear factor kappa-light-chain-enhancer of activated B cells signalling activation in splicing factor mutant MDS and leukaemia cells. Celastrol and WA may represent novel candidate drugs for the treatment of MDS.

PMID:39988885 | DOI:10.1111/bjh.20026

Intern Med J. 2025 Feb 24. doi: 10.1111/imj.16658. Online ahead of print.

ABSTRACT

The role of non-invasive ventilation (NIV) in patients with cystic fibrosis (pwCF) includes use in both the management of hypercapnic respiratory failure and as an adjunct to airway clearance techniques. We performed a retrospective review of the Australian Cystic Fibrosis Data Registry to analyse the characteristics of pwCF requiring NIV. We demonstrated that despite improvements in overall health in pwCF there is still a significant role of NIV in this population.

PMID:39989367 | DOI:10.1111/imj.16658

Adv Mater. 2025 Feb 23:e2417478. doi: 10.1002/adma.202417478. Online ahead of print.

ABSTRACT

The emergence of deep learning (DL) has provided great opportunities for the high-throughput analysis of atomic-resolution micrographs. However, the DL models trained by image patches in fixed size generally lack efficiency and flexibility when processing micrographs containing diversified atomic configurations. Herein, inspired by the similarity between the atomic structures and graphs, a few-shot learning framework based on an equivariant graph neural network (EGNN) to analyze a library of atomic structures (e.g., vacancies, phases, grain boundaries, doping, etc.) is described, showing significantly promoted robustness and three orders of magnitude reduced computing parameters compared to the image-driven DL models, which is especially evident for those aggregated vacancy lines with flexible lattice distortion. Besides, the intuitiveness of graphs enables quantitative and straightforward extraction of the atomic-scale structural features in batches, thus statistically unveiling the self-assembly dynamics of vacancy lines under electron beam irradiation. A versatile model toolkit is established by integrating EGNN sub-models for single structure recognition to process images involving varied configurations in the form of a task chain, leading to the discovery of novel doping configurations with superior electrocatalytic properties for hydrogen evolution reactions. This work provides a powerful tool to explore structure diversity in a fast, accurate, and intelligent manner.

PMID:39988855 | DOI:10.1002/adma.202417478

J Chem Inf Model. 2025 Feb 23. doi: 10.1021/acs.jcim.4c01752. Online ahead of print.

ABSTRACT

Proteins play a fundamental role in biology, and their thermostability is essential for their proper functionality. The precise measurement of thermostability is crucial, traditionally relying on resource-intensive experiments. Recent advances in deep learning, particularly in protein language models (PLMs), have significantly accelerated the progress in protein thermostability prediction. These models utilize various biological characteristics or deep representations generated by PLMs to represent the protein sequences. However, effectively incorporating structural information, based on the PLM embeddings, while not considering atomic protein structures, remains an open and formidable challenge. Here, we propose a novel Protein Contrast-enhanced Structure-Aware (ProCeSa) model that seamlessly integrates both sequence and structural information extracted from PLMs to enhance thermostability prediction. Our model employs a contrastive learning scheme guided by the categories of amino acid residues, allowing it to discern intricate patterns within protein sequences. Rigorous experiments conducted on publicly available data sets establish the superiority of our method over state-of-the-art approaches, excelling in both classification and regression tasks. Our results demonstrate that ProCeSa addresses the complex challenge of predicting protein thermostability by utilizing PLM-derived sequence embeddings, without requiring access to atomic structural data.

PMID:39988825 | DOI:10.1021/acs.jcim.4c01752

Genetics. 2025 Feb 24:iyaf026. doi: 10.1093/genetics/iyaf026. Online ahead of print.

ABSTRACT

The contributions of genetic interactions to natural trait variation are challenging to estimate experimentally, as current approaches for detecting epistasis are often underpowered. Powerful mapping approaches such as bulk segregant analysis, wherein individuals with extreme phenotypes are pooled for genotyping, obscure epistasis by averaging over genotype combinations. To accurately characterize and quantify epistasis underlying natural trait variation, we have engineered strains of the budding yeast Saccharomyces cerevisiae to enable crosses where one parent's chromosome is fixed while the rest of the chromosomes segregate. These crosses allow us to use bulk segregant analysis to identify quantitative trait loci (QTL) whose effects depend on alleles on the fixed parental chromosome, indicating a genetic interaction with that chromosome. Our method, which we term epic-QTL (for epistatic-with-chromosome QTL) analysis, can thus identify interaction loci with high statistical power. Here we perform epic-QTL analysis of copper resistance with chromosome I or VIII fixed in a cross between divergent naturally derived strains. We find seven loci that interact significantly with chromosome VIII and none that interact with chromosome I, the smallest of the 16 budding yeast chromosomes. Each of the seven interactions alters the magnitude, rather than the direction, of an additive QTL effect. We also show that fixation of one source of variation-in this case chromosome VIII, which contains the large-effect QTL mapping to CUP1-increases power to detect the contributions of other loci to trait differences.

PMID:39989051 | DOI:10.1093/genetics/iyaf026

Comput Biol Med. 2025 Apr;188:109848. doi: 10.1016/j.compbiomed.2025.109848. Epub 2025 Feb 22.

ABSTRACT

Phase I clinical trials are the first-in-human studies that primarily focus on the safety profile of drugs. Traditionally, the primary aim of a phase I clinical trial is to establish the maximum tolerated dose and characterize the toxicity profile of the tested agents. As a secondary aim, some phase I studies also include studies to obtain preliminary efficacy information about the experimental agents. In our research, we consider the optimal design of experiments in extended phase I clinical trials where both efficacy and toxicity are measured and the maximum tolerated dose has been established. We represent the response of both outcomes using a bivariate probit model for correlated responses and propose systematic numerical approaches based on Semidefinite Programming to address the problem. We construct locally optimal experimental designs for the following situations: (i) responses with efficacy and toxicity strongly correlated versus non-correlated, by varying the correlation parameter; (ii) a priori known correlation versus unknown correlation; (iii) unconstrained versus constrained designs, where the constraints represent safety limits, budget constraints and probability bounds; (iv) single versus combined drugs. Additionally, we consider four distinct optimality criteria: D-, A-, E-, and K-optimality. Our methodologies are extensively tested, and we demonstrate the optimality of the designs using equivalence theorems. To enrich our analysis, an equivalence theorem for the K-optimality criterion is derived.

PMID:39987700 | DOI:10.1016/j.compbiomed.2025.109848

J Colloid Interface Sci. 2025 Feb 20;688:79-92. doi: 10.1016/j.jcis.2025.02.133. Online ahead of print.

ABSTRACT

A non-experimental approach to developing high-performance EMI shielding materials is urgently needed to reduce costs and manpower. In this investigation, a multimodal data fusion neural network model is proposed to predict the EMI shielding performances of silver-modified four-pronged zinc oxide/waterborne polyurethane/barium ferrite (Ag@F-ZnO/WPU/BF) aerogels. First, 16 Ag@F-ZnO/WPU/BF samples with varying Ag@F-ZnO and BF contents were successfully prepared using the pre-casting and directional freezing techniques. The experimental results demonstrate that these aerogels perform well in terms of averaged EMI shielding effectiveness (SET) up to 78.6 dB and absorption coefficient as high as 0.96. On the basis of composite ingredients and microstructural images, the established multimodal neural network model can effectively predict the EMI shielding performances of Ag@F-ZnO/WPU/BF aerogels. Notably, the multimodal model of fully connected neural network (FCNN) and residual neural network (ResNet) utilizing GatedFusion method yields the best root mean squared error (RMSE) and mean absolute error (MAE) values of 0.7626 and 0.4918, respectively, and correlation coefficient (R) of 0.9885. In addition, this multimodal model successfully predicts the EMI performances of four new aerogels with an average error of less than 5 %, demonstrating its strong generalization capability. The accuracy and efficiency of material property prediction based on multimodal neural network model are largely improved by integrating multiple data sources, offering new possibility for reducing experimental burdens, accelerating the development of new materials, and gaining a deeper understanding of material mechanisms.

PMID:39987843 | DOI:10.1016/j.jcis.2025.02.133

Biomed Eng Online. 2025 Feb 23;24(1):23. doi: 10.1186/s12938-025-01349-w.

ABSTRACT

This paper reviews the recent advancements in the application of deep learning combined with electrocardiography (ECG) within the domain of cardiovascular diseases, systematically examining 198 high-quality publications. Through meticulous categorization and hierarchical segmentation, it provides an exhaustive depiction of the current landscape across various cardiovascular ailments. Our study aspires to furnish interested readers with a comprehensive guide, thereby igniting enthusiasm for further, in-depth exploration and research in this realm.

PMID:39988715 | DOI:10.1186/s12938-025-01349-w

BMC Oral Health. 2025 Feb 23;25(1):293. doi: 10.1186/s12903-025-05652-9.

ABSTRACT

BACKGROUND: Identifying fractured endodontic instruments (FEIs) in periapical radiographs (PAs) is a critical yet challenging aspect of root canal treatment (RCT) due to anatomical complexities and overlapping structures. Deep learning (DL) models offer potential solutions, yet their comparative performance in this domain remains underexplored.

METHODS: A dataset of 700 annotated PAs, including 381 teeth with FEIs, was divided into training, validation, and test sets (60/20/20 split). Five DL models-DenseNet201, EfficientNet B0, ResNet-18, VGG-19, and MaxVit-T-were trained using transfer learning and data augmentation techniques. Performance was evaluated using accuracy, AUC and MCC. Statistical analysis included the Friedman test with post-hoc corrections.

RESULTS: DenseNet201 achieved the highest AUC (0.900) and MCC (0.810), outperforming other models in FEI detection. ResNet-18 demonstrated robust results, while EfficientNet B0 and VGG-19 provided moderate performance. MaxVit-T underperformed, with metrics near random guessing. Statistical analysis revealed significant differences among models (p < 0.05), but pairwise comparisons were not significant.

CONCLUSIONS: DenseNet201's superior performance highlights its clinical potential for FEI detection, while ResNet-18 offers a balance between accuracy and computational efficiency. The findings highlight the need for model-task alignment and optimization in medical imaging applications.

PMID:39988714 | DOI:10.1186/s12903-025-05652-9

Alzheimers Dement. 2025 Feb;21(2):e14593. doi: 10.1002/alz.14593.

ABSTRACT

INTRODUCTION: Foundational models suggest Alzheimer's disease (AD) can be diagnosed using retinal images, but the specific structural features remain poorly understood. This study investigates retinal vascular changes in individuals with cognitive impairment in three East Asian regions.

METHODS: A multicenter study was conducted in Shanghai, Hong Kong, and Ningxia, collecting retinal images from 176 patients with mild cognitive impairment (MCI) or AD and 264 controls. The VC-Net deep learning model segmented arterial/venous networks, extracting 36 vascular features.

RESULTS: Significant reductions in vessel length, segment number, and vascular density were observed in cognitively impaired patients, while venous structure and complexity were correlated with the level of cognitive function.

DISCUSSION: Retinal vascular changes may serve as indicators of cognitive impairment, requiring validation in larger cohorts and exploration of the underlying mechanisms.

HIGHLIGHTS: A deep learning segmentation model extracted diverse retinal vascular features. Significant alterations in the structure of retinal arterial/venous networks were identified. Partitioning vessel-rich retinal zones improved detection of vascular changes. Decreases in vessel length, segment number, and vascular density were found in CI individuals.

PMID:39988572 | DOI:10.1002/alz.14593

Health Informatics J. 2025 Jan-Mar;31(1):14604582241295912. doi: 10.1177/14604582241295912.

ABSTRACT

Objectives: Mechanical ventilator plays a vital role in saving millions of lives. Patients with COVID-19 symptoms need a ventilator to survive during the pandemic. Studies have reported that the mortality rates rise from 50% to 97% in those requiring mechanical ventilation during COVID-19. The pumping of air into the patient's lungs using a ventilator requires a particular air pressure. High or low ventilator pressure can result in a patient's life loss as high air pressure in the ventilator causes the patient lung damage while lower pressure provides insufficient oxygen. Consequently, precise prediction of ventilator pressure is a task of great significance in this regard. The primary aim of this study is to predict the airway pressure in the ventilator respiratory circuit during the breath. Methods: A novel hybrid ventilator pressure predictor (H-VPP) approach is proposed. The ventilator exploratory data analysis reveals that the high values of lung attributes R and C during initial time step values are the prominent causes of high ventilator pressure. Results: Experiments using the proposed approach indicate H-VPP achieves a 0.78 R2, mean absolute error of 0.028, and mean squared error of 0.003. These results are better than other machine learning and deep learning models employed in this study. Conclusion: Extensive experimentation indicates the superior performance of the proposed approach for ventilator pressure prediction with high accuracy. Furthermore, performance comparison with state-of-the-art studies corroborates the superior performance of the proposed approach.

PMID:39988551 | DOI:10.1177/14604582241295912

Artif Intell Med. 2025 Feb 19:103088. doi: 10.1016/j.artmed.2025.103088. Online ahead of print.

ABSTRACT

Neonatal jaundice is a common and potentially fatal health condition in neonates, especially in low and middle income countries, where it contributes considerably to neonatal morbidity and death. Traditional diagnostic approaches, such as Total Serum Bilirubin (TSB) testing, are invasive and could lead to discomfort, infection risk, and diagnostic delays. As a result, there is a rising interest in non-invasive approaches for detecting jaundice early and accurately. An in-depth analysis of non-invasive techniques for detecting neonatal jaundice is presented by this review, exploring several AI-driven techniques, such as Machine Learning (ML) and Deep Learning (DL), which have demonstrated the ability to enhance diagnostic accuracy by evaluating complex patterns in neonatal skin color and other relevant features. It is identified that AI models incorporating variants of neural networks achieve an accuracy rate of over 90% in detecting jaundice when compared to traditional methods. Furthermore, satisfactory outcomes in field settings have been demonstrated by mobile-based applications that use smartphone cameras to estimate bilirubin levels, providing a practical alternative for resource-constrained areas. The potential impact of AI-based solutions on reducing neonatal morbidity and mortality is evaluated by this review, with a focus on real-world clinical challenges, highlighting the effectiveness and practicality of AI-based strategies as an assistive tool in revolutionizing neonatal care through early jaundice diagnosis, while also addressing the ethical and practical implications of integrating these technologies in clinical practice. Future research areas, such as the development of new imaging technologies and the incorporation of wearable sensors for real-time bilirubin monitoring, are recommended by the paper.

PMID:39988547 | DOI:10.1016/j.artmed.2025.103088