Sci Rep. 2025 Mar 15;15(1):9015. doi: 10.1038/s41598-025-93011-7.

ABSTRACT

Physicians evaluate a patient's respiratory health during a physical examination by visual assessment of the work of breathing (WoB) to determine respiratory stability, and by detecting abnormal lung sounds via lung auscultation using a stethoscope to identify common pathological lung diseases, such as chronic obstructive pulmonary disease (COPD) and pneumonia. Since these assessment methods are subjective, a low-profile device used for an accurate and quantitative monitoring approach could provide valuable preemptive insights into respiratory health, proving to be clinically beneficial. To achieve this goal, we have developed a miniature patch consisting of a sensitive wideband multi-axis seismometer that can be placed on the anatomical areas of a patient's lungs to enable an effective quantification of a patient's WoB and lung sounds. When used on a patch, the seismometer captures chest wall vibrations due to respiratory muscle effort, known as high-frequency mechanomyogram (MMG), during tidal breathing as well as seismic pulmonary-induced vibrations (PIVs) during deep breathing due to normal and/or adventitious lung sounds like crackles, while simultaneously recording respiration rate and phase. A system comprised of multiple patches was evaluated on 124 patients in the hospital setting and shown to accurately assess and quantify a patent's physical signs of WoB by measuring the average respiratory effort extracted from high-frequency MMG signals, demonstrating statistical significance of this method in comparison to clinical bedside observation of WoB and respiration rate. A data fusion deep learning model was developed which combined the inputs of PIVs lung sounds and the corresponding respiration phase to detect crackle, wheeze and normal breath sound features. The model exhibited high accuracy, sensitivity, specificity, precision and F1 score of 93%, 93%, 97%, 93% and 93% respectively, with area under the curve (AUC) of precision recall (PR) of 0.97 on the test set. Additionally, the PIVs with corresponding respiration phase captured from each auscultation point generated an acoustic map of the patient's lung, which correlated with traditional lung radiographic findings.

PMID:40089574 | DOI:10.1038/s41598-025-93011-7

Sci Rep. 2025 Mar 15;15(1):9005. doi: 10.1038/s41598-025-88687-w.

ABSTRACT

Social media has become a powerful tool for public discourse, shaping opinions and the emotional landscape of communities. The extensive use of social media has led to a massive influx of online content. This content includes instances where negativity is amplified through hateful speech but also a significant number of posts that provide support and encouragement, commonly known as hope speech. In recent years, researchers have focused on the automatic detection of hope speech in languages such as Russian, English, Hindi, Spanish, and Bengali. However, to the best of our knowledge, detection of hope speech in Urdu and English, particularly using translation-based techniques, remains unexplored. To contribute to this area we have created a multilingual dataset in English and Urdu and applied a translation-based approach to handle multilingual challenges and utilized several state-of-the-art machine learning, deep learning, and transfer learning based methods to benchmark our dataset. Our observations indicate that a rigorous process for annotator selection, along with detailed annotation guidelines, significantly improved the quality of the dataset. Through extensive experimentation, our proposed methodology, based on the Bert transformer model, achieved benchmark performance, surpassing traditional machine learning models with accuracies of 87% for English and 79% for Urdu. These results show improvements of 8.75% in English and 1.87% in Urdu over baseline models (SVM 80% English and 78% in Urdu).

PMID:40089522 | DOI:10.1038/s41598-025-88687-w

NPJ Sci Food. 2025 Mar 15;9(1):31. doi: 10.1038/s41538-025-00393-z.

ABSTRACT

Veterinary drug residues in poultry and livestock products present persistent challenges to food safety, necessitating precise and efficient detection methods. Surface-enhanced Raman scattering (SERS) has been identified as a powerful tool for veterinary drug residue analysis due to its high sensitivity and specificity. However, the development of reliable SERS substrates and the interpretation of complex spectral data remain significant obstacles. This review summarizes the development process of SERS substrates, categorizing them into metal-based, rigid, and flexible substrates, and highlighting the emerging trend of multifunctional substrates. The diverse application scenarios and detection requirements for these substrates are also discussed, with a focus on their use in veterinary drug detection. Furthermore, the integration of deep learning techniques into SERS-based detection is explored, including substrate structure design optimization, optical property prediction, spectral preprocessing, and both qualitative and quantitative spectral analyses. Finally, key limitations are briefly outlined, such as challenges in selecting reporter molecules, data imbalance, and computational demands. Future trends and directions for improving SERS-based veterinary drug detection are proposed.

PMID:40089516 | DOI:10.1038/s41538-025-00393-z

Sci Rep. 2025 Mar 15;15(1):8938. doi: 10.1038/s41598-025-92994-7.

ABSTRACT

Cracks are common defects in physical structures, and if not detected and addressed in a timely manner, they can pose a severe threat to the overall safety of the structure. In recent years, with advancements in deep learning, particularly the widespread use of Convolutional Neural Networks (CNNs) and Transformers, significant breakthroughs have been made in the field of crack detection. However, CNNs still face limitations in capturing global information due to their local receptive fields when processing images. On the other hand, while Transformers are powerful in handling long-range dependencies, their high computational cost remains a significant challenge. To effectively address these issues, this paper proposes an innovative modification to the VM-UNet model. This modified model strategically integrates the strengths of the Mamba architecture and UNet to significantly improve the accuracy of crack segmentation. In this study, we optimized the original VM-UNet architecture to better meet the practical needs of crack segmentation tasks. Through comparative experiments on the Crack500 and Ozgenel public datasets, the results clearly demonstrate that the improved VM-UNet achieves significant advancements in segmentation accuracy. Compared to the original VM-UNet and other state-of-the-art models, VM-UNet++ shows a 3% improvement in mDS and a 4.6-6.2% increase in mIoU. These results fully validate the effectiveness of our improvement strategy. Additionally, VM-UNet++ demonstrates lower parameter count and floating-point operations, while maintaining a relatively satisfactory inference speed. These improvements make VM-UNet++ advantageous for practical applications.

PMID:40089495 | DOI:10.1038/s41598-025-92994-7

Expert Rev Pharmacoecon Outcomes Res. 2025 Mar 16. doi: 10.1080/14737167.2025.2480718. Online ahead of print.

ABSTRACT

INTRODUCTION: Objective: To systematically review studies on the cost-effectiveness of pirfenidone compared to nintedanib in patients with idiopathic pulmonary fibrosis (IPF).

METHODS: Data sources: PubMed, EMBASE, Scopus, and Web of Science. Inclusion criteria: Full economic evaluations comparing pirfenidone versus nintedanib in IPF patients. Assessment: Quality of Health Economic Studies (QHES) tool for study quality.

RESULTS: Nine studies met the inclusion criteria with QHES scores of 0.91 or higher. The incremental cost-effectiveness ratios (ICERs) ranged from $66,434 to $1,668,321 per quality-adjusted life year (QALY) in the United States. Nintedanib was found to be cost-effective in five studies.

CONCLUSIONS: Nine studies met the inclusion criteria with QHES scores of 0.91 or higher. The incremental cost-effectiveness ratios (ICERs) ranged from $66,434 to $1,668,321 per quality-adjusted life year (QALY) in the United States. Nintedanib was found to be cost-effective in five studies.Nintedanib appears to be a more cost-effective treatment for IPF compared to pirfenidone. Further research is needed, particularly in low- and middle-income countries, considering healthcare system perspectives and varied willingness-to-pay thresholds.

PMID:40089935 | DOI:10.1080/14737167.2025.2480718

BMC Health Serv Res. 2025 Mar 15;25(1):385. doi: 10.1186/s12913-025-12506-1.

ABSTRACT

OBJECTIVES: Novel non-invasive machine learning algorithms may improve accuracy and reduce the need for biopsy when diagnosing idiopathic pulmonary fibrosis (IPF). We conducted a cost-effectiveness analysis of diagnostic strategies for IPF.

METHODS: We developed a decision analytic model to evaluate diagnostic strategies for IPF in the United States. To assess the full spectrum of costs and benefits, we compared four interventions: a machine learning diagnostic algorithm, a genomic classifier, a biopsy-all strategy, and a treat-all strategy. The analysis was conducted from the health sector perspective with a lifetime horizon. The primary outcome measures were costs, Quality-Adjusted Life-Years (QALYs) gained, and Incremental Cost-Effectiveness Ratios (ICERs) based on the average of 10,000 probabilistic runs of the model.

RESULTS: Compared to a biopsy-all strategy the machine learning algorithm and genomic classifer reduced diagnostic-related costs by $14,876 and $3,884, respectively. Use of the machine learning algorithm consistently reduced diagnostic costs. When including downstream treatment costs and benefits of anti-fibrotic treatment, the machine learning algorithm had an ICER of $331,069 per QALY gained compared to the biopsy-all strategy. The genomic classifier had a higher ICER of $390,043 per QALY gained, while the treat-all strategy had the highest ICER of $3,245,403 per QALY gained. Results were sensitive to changes in various input parameters including IPF treatment costs, sensitivity and specificity of novel screening tools, and the rate of additional diagnostics following inconclusive results. High treatment costs were found to drive overall cost regardless of the diagnostic method. As treatment costs lowered, the supplemental diagnostic tools became increasingly cost-effective.

CONCLUSIONS: Novel tools for diagnosing IPF reduced diagnostic costs, while overall incremental cost-effectiveness ratios were high due to treatment costs. New IPF diagnosis approaches may become more favourable with lower-cost treatments for IPF.

PMID:40089758 | DOI:10.1186/s12913-025-12506-1

Plant J. 2025 Mar;121(6):e70095. doi: 10.1111/tpj.70095.

ABSTRACT

When pollen lands on a receptive stigma, it germinates and extends a tube inside the transmitting tissue of the pistil to deliver the sperm cells for double fertilization. The growth of the pollen tube triggers significant alterations in its gene expression. The extent to which these changes occur in the vegetative cell or extend to the sperm cells transported by the tube is unclear but important to determine since sperm cells are believed to acquire a competency for fertilization during pollen-pistil interactions. To address these questions, we compared the transcriptomes of Arabidopsis thaliana sperm cells and vegetative nuclei isolated from mature pollen grains with those isolated from in vitro-grown pollen tubes. Importantly, we also compared transcriptomes of sperm cells obtained from pollen tubes grown under semi-in vivo conditions where tubes passed through a pistil section. Our data show that extensive transcriptomic changes occur in sperm cells during pollen tube growth, some of which are elicited only as sperms are carried through the pistil. Their analysis reveals a host of previously unidentified transcripts that may facilitate sperm maturation and gamete fusion. The vegetative cell undergoes even more extensive transcriptomic reprogramming during pollen tube growth, mainly through the upregulation of genes associated with pollen tube growth and vesicle-mediated transport. Interestingly, ATAC-seq data show that the promoters of genes upregulated in sperm during pollen tube growth are already accessible in sperm chromatin of mature pollen grains, suggesting pre-configured promoter accessibility. This study's expression data can be further explored here: https://bar.utoronto.ca/eFP-Seq_Browser/.

PMID:40089905 | DOI:10.1111/tpj.70095

Plant J. 2025 Mar;121(6):e70061. doi: 10.1111/tpj.70061.

ABSTRACT

Diatoms are prominent microalgae found in all aquatic environments. Over the last 20 years, thanks to the availability of genomic and genetic resources, diatom species such as Phaeodactylum tricornutum and Thalassiosira pseudonana have emerged as valuable experimental model systems for exploring topics ranging from evolution to cell biology, (eco)physiology, and biotechnology. Since the first genome sequencing projects initiated more than 20 years ago, numerous genome-enabled datasets have been generated, based on RNA-Seq and proteomics experiments, epigenomes, and ecotype variant analysis. Unfortunately, these resources, generated by various laboratories, are often in disparate formats and challenging to access and analyze. Here we present DiatOmicBase, a genome portal gathering comprehensive omics resources from P. tricornutum and T. pseudonana to facilitate the exploration of dispersed public datasets and the design of new experiments based on the prior-art. DiatOmicBase provides gene annotations, transcriptomic profiles and a genome browser with ecotype variants, histone and methylation marks, transposable elements, non-coding RNAs, and read densities from RNA-Seq experiments. We developed a semi-automatically updated transcriptomic module to explore both publicly available RNA-Seq experiments and users' private datasets. Using gene-level expression data, users can perform exploratory data analysis, differential expression, pathway analysis, biclustering, and co-expression network analysis. Users can create heatmaps to visualize pre-computed comparisons for selected gene subsets. Automatic access to other bioinformatic resources and tools for diatom comparative and functional genomics is also provided. Focusing on the resources currently centralized for P. tricornutum, we showcase several examples of how DiatOmicBase strengthens molecular research on diatoms, making these organisms accessible to a broad research community.

PMID:40089834 | DOI:10.1111/tpj.70061

BMC Infect Dis. 2025 Mar 15;25(1):363. doi: 10.1186/s12879-025-10712-0.

ABSTRACT

BACKGROUND: Mycoplasma pneumoniae (MP) is a common bacterial respiratory infection that can cause pneumonia, particularly in children. Previously published data have highlighted the high incidence of viral co-infections and the problem of increasing macrolide resistance in MP worldwide.

AIMS: (1) to estimate the impact of viral infections circulating in a local population on the spectrum of co-infection in hospitalized children with Mycoplasma pneumoniae pneumonia (MPP), (2) to determine if there are differences in resistance mutation rate for samples from hospitals of Russia located in the European and Far East, (3) to describe genomic characteristics of MP from Russian patients during the MPP outbreaks in the fall-winter of 2023-2024.

METHODS: The carriage of viral pathogens was analyzed by real-time PCR in children with MPP from the European Part and Far East of Russian Federation and compared with the infections from two control groups. The V region of the 23S gene and the quinolone resistance-determining regions (QRDRs) of the parC and gyrA genes were sequenced to detect resistance-associated mutations in MP. Whole-genome sequencing method was used to determine the genetic relationship of a Russian MP isolate with known MP isolates.

RESULTS: The 62% of patients with MPP had a viral co-infection, with HPIV and SARS-CoV-2 predominating at 47% and 12.4%, respectively. The 15% of patients were infected with two or more viruses. In the control groups, 21% of healthy children and 43% of healthy adults were infected with Coronaviruses and Human Parainfluenza Viruses (HPIV-3 and -4), respectively. The 2063 A/G mutation of the 23S gene was found in 40.8% of patients from European Russia and in 35.7% of patients from the Far East. The result of core genes demonstrates that the sequence obtained from Russia clusters with sequences from clade 1.

CONCLUSIONS: Both HPIV and SARS-CoV-2 circulated in the population among healthy children and adults in December 2023 and they also were predominated in children with MPP. The rate of macrolide resistance was ⁓40%, which is higher than in European countries and significantly lower than in patients from Asian countries. Phylogenetic analysis showed the MP genome form Russia related to P1 type 1 (clade 1).

PMID:40089690 | DOI:10.1186/s12879-025-10712-0

Oncogene. 2025 Mar 16. doi: 10.1038/s41388-025-03337-9. Online ahead of print.

ABSTRACT

Chemoresistance poses a significant challenge in colorectal cancer (CRC) treatment. However, the mechanisms underlying chemoresistance remain unclear. CBX3 promoted proliferation and metastasis in CRC. However, the role and mechanism of CBX3 in chemoresistance remain unknown. Therefore, we aimed to investigate the effects and mechanisms of CBX3 on multidrug resistance in CRC. Our studies showed that higher levels of CBX3 expression were associated with poor survival, especially in groups with progression following chemotherapy. CBX3 overexpression increased Irinotecan and Oxaliplatin resistance, whereas CBX3 knockdown suppressed multidrug resistance in CRC cells. Additionally, CBX3 inhibited ferroptosis associated with multidrug resistance, and the ferroptosis activators prevented CBX3 overexpression-mediated cell survival. RNA sequencing revealed that the NRF2-signaling pathway was involved in this process. CBX3-upregulated NRF2 protein expression by directly binding to the promoter of Cullin3 (CUL3) to suppress CUL3 transcription and CUL3-mediated NRF2 degradation. Moreover, Glutathione Peroxidase 2 (GPX2) was downstream of the CBX3-NRF2 pathway in CRC chemoresistance. ML385, an NRF2 inhibitor, suppressed GPX2 expression, and increased ferroptosis in PDX models. Our study identified CBX3/NRF2/GPX2 axis may be a novel signaling pathway that mediates multidrug resistance in CRC. This study proposes developing novel strategies for cancer treatment to overcome drug resistance in the future.

PMID:40089640 | DOI:10.1038/s41388-025-03337-9

J Mol Biol. 2025 Mar 13:169089. doi: 10.1016/j.jmb.2025.169089. Online ahead of print.

ABSTRACT

Dr. Donita C. Brady is the Harrison McCrea Dickson, MD, and Clifford C. Baker, MD Presidential Associate Professor of Cancer Biology at the University of Pennsylvania Perelman School of Medicine. She earned her BS in Chemistry from Radford University and her PhD in Pharmacology from UNC-Chapel Hill before completing postdoctoral training at Duke University with Dr. Christopher Counter. At Penn, Dr. Brady leads a research program pioneeringmetalloallostery, where redox-active metals regulate kinase activity. Her lab investigates the intersection of kinase signaling and copper (Cu) homeostasis, identifying Cu-dependent kinases and developing targeted therapies through drug repurposing and novel drug design. Her work has advanced our understanding of metals in nutrient signaling, energy homeostasis, and cancer metabolism. Dr. Brady has received numerous honors, including being a Pew Biomedical Scholar, a V Foundation Scholar, and the recipient of the Perelman School of Medicine's Michael S. Brown New Investigator Research Award. A dedicated advocate for diversity, equity, inclusion, and accessibility (DEIA), she has spent the past decade addressing barriers to representation in STEM. In 2021, she was appointed the inaugural Assistant Dean for Inclusion, Diversity, and Equity (IDE) in Research Training at Penn, leading efforts to foster an inclusive research environment. For these contributions, she received the 2022 Vanderbilt Basic Science Juneteenth Icon Award and the Penn Biomedical Graduate Studies Cell and Molecular Biology Graduate Group Community Service Award.

PMID:40089146 | DOI:10.1016/j.jmb.2025.169089

Drug Discov Today. 2025 Mar 13:104331. doi: 10.1016/j.drudis.2025.104331. Online ahead of print.

ABSTRACT

Ovarian cancer (OC) has poor survival statistics and increasing prevalence. One of the new options for its therapy could be overcoming platinum resistance. In this review, we have considered the idea of repositioning angiotensin-converting enzyme inhibitors (ACE-Is) as chemosensitizers. These drugs have been shown to suppress angiogenesis and OC cell migration in preclinical studies. Moreover, clinical data have shown that using ACE-Is with standard chemotherapy prolongs patient survival. Based on this rationale, we discuss the available in vitro models of OC for future studies with ACE-Is and demonstrate an in silico approach that has enabled us to select the most promising molecules: perindopril, ramipril, trandolapril and their diketopiperazine derivatives.

PMID:40089017 | DOI:10.1016/j.drudis.2025.104331

Comput Biol Med. 2025 Mar 14;189:109984. doi: 10.1016/j.compbiomed.2025.109984. Online ahead of print.

ABSTRACT

The deployment of artificial intelligence (AI) is revolutionizing neuropharmacology and drug development, allowing the modulation of neurotransmitter systems at the personal level. This review focuses on the neuropharmacology and regulation of neurotransmitters using predictive modeling, closed-loop neuromodulation, and precision drug design. The fusion of AI with applications such as machine learning, deep-learning, and even computational modeling allows for the real-time tracking and enhancement of biological processes within the body. An exemplary application of AI is the use of DeepMind's AlphaFold to design new GABA reuptake inhibitors for epilepsy and anxiety. Likewise, Benevolent AI and IBM Watson have fast-tracked drug repositioning for neurodegenerative conditions. Furthermore, we identified new serotonin reuptake inhibitors for depression through AI screening. In addition, the application of Deep Brain Stimulation (DBS) settings using AI for patients with Parkinson's disease and for patients with major depressive disorder (MDD) using reinforcement learning-based transcranial magnetic stimulation (TMS) leads to better treatment. This review highlights other challenges including algorithm bias, ethical concerns, and limited clinical validation. Their proposal to incorporate AI with optogenetics, CRISPR, neuroprosthesis, and other advanced technologies fosters further exploration and refinement of precision neurotherapeutic approaches. By bridging computational neuroscience with clinical applications, AI has the potential to revolutionize neuropharmacology and improve patient-specific treatment strategies. We addressed critical challenges, such as algorithmic bias and ethical concerns, by proposing bias auditing, diverse datasets, explainable AI, and regulatory frameworks as practical solutions to ensure equitable and transparent AI applications in neurotransmitter modulation.

PMID:40088712 | DOI:10.1016/j.compbiomed.2025.109984

Eur J Pharmacol. 2025 Mar 13:177509. doi: 10.1016/j.ejphar.2025.177509. Online ahead of print.

ABSTRACT

The endoplasmic reticulum (ER) plays a fundamental role in maintaining cellular homeostasis by ensuring proper protein folding, lipid metabolism, and calcium regulation. However, disruptions to ER function, known as ER stress, activate the unfolded protein response (UPR) to restore balance. Chronic or unresolved ER stress contributes to metabolic dysfunctions, including insulin resistance, non-alcoholic fatty liver disease (NAFLD), and neurodegenerative disorders such as Parkinson's disease and Alzheimer's disease. Recent studies have also highlighted the importance of mitochondria-ER contact sites (MERCs) and ER-associated inflammation in disease progression. This review explores the current pharmacological landscape targeting ER stress, focusing on therapeutic strategies for rare metabolic and neurodegenerative diseases. It examines small molecules such as tauroursodeoxycholic acid (TUDCA) and 4-phenylbutyric acid (4-PBA), repurposed drugs like 17-AAG (17-N-allylamino-17demethoxygeldanamycin (tanespimycin)) and berberine, and phytochemicals such as resveratrol and hesperidin. Additionally, it discusses emerging therapeutic areas, including soluble epoxide hydrolase (sEH) inhibitors for metabolic disorders and MERCs modulation for neurological diseases. The review emphasizes challenges in translating these therapies to clinical applications, such as toxicity, off-target effects, limited bioavailability, and the lack of large-scale randomized controlled trials (RCTs). It also highlights the potential of personalized medicine approaches and pharmacogenomics in optimizing ER stress-targeting therapies.

PMID:40089262 | DOI:10.1016/j.ejphar.2025.177509

J Allergy Clin Immunol. 2025 Mar 13:S0091-6749(25)00274-X. doi: 10.1016/j.jaci.2025.02.037. Online ahead of print.

ABSTRACT

BACKGROUND: Inhaled corticosteroids (ICS) are frequently prescribed medications for asthma symptoms, but higher doses can increase risks of adrenal insufficiency through suppression of endogenous cortisol production. Understanding which patients may be at increased risk for developing adrenal suppression related to ICS use may help providers improve treatment regimens for asthma patients; however, the mechanisms underlying ICS-related adrenal insufficiency have not been clarified.

OBJECTIVE: This study identifies metabolite signatures and biochemical pathways associated with ICS-related adrenal insufficiency in asthma patients.

METHODS: Global metabolite profiling (metabolomics) was integrated with electronic medical records data including the development of adrenal suppression, in two independent asthma cohorts. The discovery cohort, Pharmacogenomics of Adrenal Suppression with Inhaled Corticosteroids (PhASIC), included 711 adult asthma patients on ICS. Untargeted metabolomic profiling identified 1,397 metabolites, of which 810 were selected for further analysis. Using plasma cortisol as a biomarker for adrenal status (outcome), linear regression models were implemented to identify associations between metabolites and plasma cortisol, adjusted for potential confounders. Metabolite associations were validated in an additional 575 patients on ICS. Pathway and network analyses were performed using bioinformatic approaches to identify altered metabolic pathways related to the outcome.

RESULTS: Of 810 endogenous metabolites, 12 demonstrated significant associations with adrenal insufficiency after correction for multiple comparisons. In the validation cohort, three of these 12 replicated, including two steroid metabolites (tetrahydrocortisol glucuronide and tetrahydrocortisol glucuronide (5)) and homocitrulline. Pathway and network analyses revealed alterations in biochemical pathways related to the metabolism of steroids, bile acids, urea cycle and long-chain polyunsaturated fatty acids.

CONCLUSIONS: We have identified specific metabolites within steroid and non-steroid metabolic pathways that are associated with adrenal insufficiency with ICS use.

PMID:40089116 | DOI:10.1016/j.jaci.2025.02.037

J Cyst Fibros. 2025 Mar 14:S1569-1993(25)00075-X. doi: 10.1016/j.jcf.2025.03.007. Online ahead of print.

ABSTRACT

BACKGROUND: Elexacaftor/tezacaftor/ivacaftor (ETI) has been highly effective for improving pulmonary disease and nutritional outcomes. However, the effect of this therapy on glycemic control in people with cystic fibrosis related diabetes (CFRD) is unclear. This study aimed to examine real-world effects of ETI on glycemia as captured by hemoglobin A1c (HbA1c) in people with pre-existing CFRD.

METHODS: Retrospective chart review was performed at 4 US CF centers. Individuals with CFRD included in the study started ETI before December 2020, and had an HbA1c within 1 year before and up to 2 years after ETI initiation. A sub-analysis comparing CGM data and insulin dosing within the year before and after ETI was performed. Summary statistics were calculated and within-subject results compared.

RESULTS: A total 175 individuals with CFRD had HbA1c data before and after ETI. Mean (±SD) age was 32.4 (±12.4) years, 49.1 % female. HbA1c were compared a median (IQR) of -40 (-93, 0) days before and 290 (107, 441) days after ETI initiation. Median (IQR) HbA1c decreased from 6.4 % (5.8, 7.2) to 6.0 % (5.5, 6.8), p<0.001. A subgroup of 13 individuals had CGM and basal insulin data for comparison. No changes were observed in CGM metrics, however, basal insulin dose in these patients decreased (p=0.03).

CONCLUSION: Findings suggest clinical improvements in glycemia following ETI initiation in people with CFRD. Further studies are required to better understand the mechanisms by which ETI may modulate insulin and glucose dynamics in individuals with existing CFRD.

PMID:40089409 | DOI:10.1016/j.jcf.2025.03.007

J Cyst Fibros. 2025 Mar 14:S1569-1993(25)00076-1. doi: 10.1016/j.jcf.2025.03.008. Online ahead of print.

ABSTRACT

BACKGROUND: The clinical efficacy of elexacaftor-tezacaftor-ivacaftor (ETI) in children with cystic fibrosis (cwCF) is variable; some respond, while others do not or have side effects. The pharmacokinetics (PK) of ETI are poorly described in published research, particularly when it comes to children. Knowledge of the PK in this population may provide more insight into the exposure-response relationship of the drugs and its corresponding inter-patient variability (IIV). The aim of this study was to evaluate the PK of ETI in cwCF in a real-world setting.

METHODS: A prospective, observational PK study was conducted in cwCF starting with ETI. PK samples were collected at home using dried blood spots (DBS), and during regular outpatient hospital visits. Clinical efficacy and safety data were gathered and evaluated. Population PK (popPK) models were developed using nonlinear mixed-effects modelling.

RESULTS: A total of 29 children were included in this study. Novel popPK models were developed for ETI and its main metabolites. There was significant variability in AUC of ETI within and between age groups, aligning with the references in the product information. All children had concentrations within or above the range needed for a clinical response. An exploratory exposure-response analysis found no direct linear relationship between AUC and sweat chloride, or ppFEV1.

CONCLUSIONS: This study is the first analysis of ETI popPK in cwCF. The developed popPK models may be used to further study the exposure-response relationship and its variability within cwCF, as a basis for more personalized dosing.

PMID:40089408 | DOI:10.1016/j.jcf.2025.03.008

Magn Reson Imaging. 2025 Mar 13:110370. doi: 10.1016/j.mri.2025.110370. Online ahead of print.

ABSTRACT

OBJECTIVE: To evaluate whether deep learning analysis (DL) of intratumor subregion based on dynamic contrast-enhanced MRI (DCE-MRI) can help predict Ki-67 expression level in breast cancer.

MATERIALS AND METHODS: A total of 290 breast cancer patients from two hospitals were retrospectively collected. A k-means clustering algorithm confirmed subregions of tumor. DL features of whole tumor and subregions were extracted from DCE-MRI images based on 3D ResNet18 pre-trained model. The logistic regression model was constructed after dimension reduction. Model performance was assessed using the area under the curve (AUC), and clinical value was demonstrated through decision curve analysis (DCA).

RESULTS: The k-means clustering method clustered the tumor into two subregions (habitat 1 and habitat 2) based on voxel values. Both the habitat 1 model (validation set: AUC = 0.771, 95 %CI: 0.642-0.900 and external test set: AUC = 0.794, 95 %CI: 0.696-0.891) and the habitat 2 model (AUC = 0.734, 95 %CI: 0.605-0.862 and AUC = 0.756, 95 %CI: 0.646-0.866) showed better predictive capabilities for Ki-67 expression level than the whole tumor model (AUC = 0.686, 95 %CI: 0.550-0.823 and AUC = 0.680, 95 %CI: 0.555-0.804). The combined model based on the two subregions further enhanced the predictive capability (AUC = 0.808, 95 %CI: 0.696-0.921 and AUC = 0.842, 95 %CI: 0.758-0.926), and it demonstrated higher clinical value than other models in DCA.

CONCLUSIONS: The deep learning model derived from subregion of tumor showed better performance for predicting Ki-67 expression level in breast cancer patients. Additionally, the model that integrated two subregions further enhanced the predictive performance.

PMID:40089082 | DOI:10.1016/j.mri.2025.110370

Int J Radiat Oncol Biol Phys. 2025 Mar 13:S0360-3016(25)00243-3. doi: 10.1016/j.ijrobp.2025.03.012. Online ahead of print.

ABSTRACT

BACKGROUND: Patients with identical stages or similar tumor volumes can vary significantly in their responses to radiotherapy (RT) due to individual characteristics, making personalized RT for non-small cell lung cancer (NSCLC) challenging. This study aimed to develop a deep learning (DL) model by integrating pre- and mid-treatment computed tomography (CT) to predict the treatment response in NSCLC patients.

METHODS AND MATERIAL: We retrospectively collected data from 168 NSCLC patients across three hospitals. Data from A (35 patients) and B (93 patients) were used for model training and internal validation, while data from C (40 patients) was used for external validation. DL, radiomics, and clinical features were extracted to establish a varying time-interval long short-term memory network (VTI-LSTM) for response prediction. Furthermore, we derived a model-deduced personalize dose escalation (DE) for patients predicted to have suboptimal gross tumor volume (GTV) regression. The area under the receiver operating characteristic curve (AUC) and predicted absolute error (PAE) were used to evaluate the predictive Response Evaluation Criteria in Solid Tumors (RECIST) classification and proportion of GTV residual. DE was calculated as biological equivalent dose (BED) using an α/β ratio of 10 Gy.

RESULTS: The model using only pre-treatment CT achieved the highest AUC of 0.762 and 0.687 in internal and external validation respectively, while the model integrating both pre- and mid-treatment CT achieved AUC of 0.869 and 0.798, with PAE of 0.137 and 0.185. We performed personalized DE for 29 patients. Their original BED was approximately 72 Gy, within the range of 71.6 Gy to 75 Gy. DE ranged from 77.7 to 120 Gy for 29 patients, with 17 patients exceeding 100 Gy and eight patients reaching the model's preset upper limit of 120 Gy.

CONCLUSIONS: Combining pre- and mid-treatment CT enhances prediction performance for RT response and offers a promising approach for personalized DE in NSCLC.

PMID:40089073 | DOI:10.1016/j.ijrobp.2025.03.012

Neural Netw. 2025 Mar 10;187:107337. doi: 10.1016/j.neunet.2025.107337. Online ahead of print.

ABSTRACT

Deep learning for Electroencephalography (EEG) has become dominant in the tasks of discrimination and evaluation of brain disorders. However, despite its significant successes, this approach has long been facing challenges due to the limited availability of labeled samples and the individuality of subjects, particularly in complex scenarios such as Autism Spectrum Disorders (ASD). To facilitate the efficient optimization of EEG discrimination models in the face of these limitations, this study has developed a framework called STEM (Self-Training EEG Model). STEM accomplishes this by self-training the model, which involves initializing it with limited labeled samples and optimizing it with self-constructed samples. (1) Model initialization with multi-task learning: A multi-task model (MAC) comprising an AutoEncoder and a classifier offers guidance for subsequent pseudo-labeling. This guidance includes task-related latent EEG representations and prediction probabilities of unlabeled samples. The AutoEncoder, which consists of depth-separable convolutions and BiGRUs, is responsible for learning comprehensive EEG representations through the EEG reconstruction task. Meanwhile, the classifier, trained using limited labeled samples through supervised learning, directs the model's attention towards capturing task-related features. (2) Model optimization aided by pseudo-labeled samples construction: Next, trustworthy pseudo-labels are assigned to the unlabeled samples, and this approach (PLASC) combines the sample's distance relationship in the feature space mapped by the encoder with the sample's predicted probability, using the initial MAC model as a reference. The constructed pseudo-labeled samples then support the self-training of MAC to learn individual information from new subjects, potentially enhancing the adaptation of the optimized model to samples from new subjects. The STEM framework has undergone an extensive evaluation, comparing it to state-of-the-art counterparts, using resting-state EEG data collected from 175 ASD-suspicious children spanning different age groups. The observed results indicate the following: (1) STEM achieves the best performance, with an accuracy of 88.33% and an F1-score of 87.24%, and (2) STEM's multi-task learning capability outperforms supervised methods when labeled data is limited. More importantly, the use of PLASC improves the model's performance in ASD discrimination across different age groups, resulting in an increase in accuracy (3%-8%) and F1-scores (4%-10%). These increments are approximately 6% higher than those achieved by the comparison methods.

PMID:40088831 | DOI:10.1016/j.neunet.2025.107337