Int Immunopharmacol. 2025 Jun 3;161:114975. doi: 10.1016/j.intimp.2025.114975. Online ahead of print.

ABSTRACT

Severe bacterial infections (e.g., pneumonia, sepsis) serve as key contributors to acute lung injury (ALI), underscoring the necessity of concurrent anti-infective therapy. The pathogenesis of ALI primarily mediates through two intertwined pathological processes: oxidative stress and cytokine storm. Minocycline, a semisynthetic tetracycline derivative with established clinical applications, exhibits promising potential as a therapeutic candidate for ALI due to its anti-inflammatory and antioxidant pharmacological actions. This investigation employed the lipopolysaccharide (LPS)-induced ALI mice model and RAW264.7 cells inflammation model to evaluate the pulmonary protective effects of minocycline. Our findings demonstrated that minocycline ameliorated symptoms of ALI in LPS-induced mice, including attenuating inflammatory cell infiltration, suppressing cytokine storm, mitigating oxidative stress damage, alleviating pulmonary edema and reducing microvascular permeability. Parallel in vitro experiments revealed that minocycline exhibited inhibitory effects on inflammatory response and oxidative stress in LPS-stimulated RAW264.7 cells. These results suggested that minocycline attenuated cytokine storm and oxidative stress, thereby protecting mice against lung injury. Therefore, minocycline may offer superior benefits compared to other antibiotics for ALI patients infected with susceptible bacteria. While preclinical investigations have unveiled emerging clinical application prospects, rigorous clinical trials remain imperative to substantiate minocycline's therapeutic efficacy in human populations.

PMID:40466614 | DOI:10.1016/j.intimp.2025.114975

Eur J Med Chem. 2025 May 31;295:117816. doi: 10.1016/j.ejmech.2025.117816. Online ahead of print.

ABSTRACT

The unique characteristics of cancer are crucial for comprehending the processes underlying cancer initiation, development, and maintenance. These hallmarks guide the development of novel therapeutic strategies aimed at fundamental traits of cancer, resulting in more targeted therapies with the possibility for sustained effectiveness and minimized adverse effects. Drug repurposing, a novel approach that leverages the known safety and pharmacological properties of existing drugs, has surfaced as a viable alternative to traditional drug development. This method expedites the timescale for introducing novel medicines into clinical practice, often demonstrating reduced failure rates in clinical trials. Recent data substantiates the therapeutic efficacy of many repurposed medications in the management of oral squamous cell carcinomas (OSCC), a highly aggressive and treatment-resistant malignancy. Prominent instances include metformin, phenformin, propranolol, acetylsalicylic acid, celecoxib, itraconazole, statins, dihydroartemisinin, and methotrexate. These pharmaceuticals demonstrated diverse anticancer actions, rendering them valuable tools in the therapy of OSCC. This review provides a comprehensive overview of molecular signaling in the reprocessing of drugs for OSCC.

PMID:40466285 | DOI:10.1016/j.ejmech.2025.117816

Biochem Pharmacol. 2025 Jun 2:117016. doi: 10.1016/j.bcp.2025.117016. Online ahead of print.

ABSTRACT

The pregnane X receptor (PXR) activator rifampicin (RIF) plays a critical role in drug-drug interactions (DDIs) by inducing cytochrome P450 (CYP) 3A4 expression. Increasing evidence indicates that n6-methyladenosine (m6A) modification is involved in the regulation of CYP basal expression. Here, we showed its effect on RIF-induced CYP3A4 expression. This study found that m6A levels and methyltransferase-like 3 (METTL3) expression were upregulated in HepG2 and LS174T cells treated with RIF, as well as in mice treated with pregnenolone-16α-carbonitrile (PCN, a typical PXR activator in mice), associated with the expression of CYP3A4 and Cyp3a11 (mouse homolog of human CYP3A4). Specifically, knockdown or overexpression of METTL3 downregulated and upregulated the basal and RIF-induced expression of CYP3A4, respectively. Similar results were obtained in a mouse model with liver-specific knockdown of Mettl3 (homolog of human METTL3) treated with PCN, indicating the involvement of m6A methylation in RIF-induced CYP3A4 expression. Mechanistically, METTL3 promotes PXR nuclear translocation and protein translation, while potentially affecting CYP3A4 mRNA stability through its binding to CYP3A4 mRNA (enhanced by RIF), thereby contributing to RIF-induced upregulation of CYP3A4 expression. Functionally, we observed that METTL3 knockdown or treatment with the METTL3 inhibitor STM2457 reduced the cytotoxicity induced by RIF combined with ritonavir. In conclusion, this study identified a novel mechanism of m6A modification in the regulation of RIF-induced CYP3A4 expression, providing valuable insights into CYP3A4-mediated DDIs.

PMID:40466953 | DOI:10.1016/j.bcp.2025.117016

EBioMedicine. 2025 Jun 3;117:105785. doi: 10.1016/j.ebiom.2025.105785. Online ahead of print.

ABSTRACT

With the rapid evolution of precision medicine, diagnostic, and therapeutic strategies for lupus nephritis (LN) are progressively shifting towards greater personalisation and accuracy. On the diagnostic front, the discovery and deployment of novel molecular biomarkers, together with the integration of artificial intelligence (AI) and machine learning (ML) technologies, have opened new avenues for delineating disease subtypes and forecasting treatment responses, thereby optimising clinical decision-support systems. In the therapeutic domain, innovations in targeted biologics and the precise application of diverse targeted modalities have shown promise in enhancing efficacy while reducing adverse effects. The triadic synergy of molecular biomarker breakthroughs, iterative advances in intelligent computational tools and pioneering targeted therapies is poised to usher LN in a new era of customised diagnosis and intervention.

PMID:40466435 | DOI:10.1016/j.ebiom.2025.105785

Hum Immunol. 2025 Jun 3;86(4):111333. doi: 10.1016/j.humimm.2025.111333. Online ahead of print.

ABSTRACT

BACKGROUND: High HIV/AIDS prevalence in China calls for personalized therapies like pharmacogenomics to improve antiretroviral treatment efficacy. This study explored associations between single nucleotide polymorphisms (SNPs) and outcomes of highly active antiretroviral therapy (HAART) in HIV-1 patients.

METHODS: We collected blood samples and clinical data from 503 HIV-1-infected patients undergoing HAART for 12 months. Using bioinformatics databases, we identified 81 SNPs in genes related to drug transport, immunity, and HIV infection. These SNPs were genotyped and correlated with highly active antiretroviral therapy efficacy and side effects using the Mass Array system and SPSS. FDR multiple correction was performed on all positive SNPs.

RESULTS: After 12 months of highly active antiretroviral therapy with Tenofovir disoproxil fumarate (TDF) and Lamivudine (3TC), the patients showed significant viral suppression and immune recovery.CD4 response was associated with GABPB1 rs12594956 (OR = 0.378, P = 0.002, FDR-P = 0.032). Viral load response was associated with CCR5 rs2734648 (OR = 22.812, P = 0.018, FDR-P = 0.045), IL2RA rs1323657 (OR = 18.312, P = 0.020, FDR-P = 0.045) and IL2 rs2069772(OR = 139.173, P = 0.002, FDR-P = 0.02). Rash risk was elevated with SLC22A2 rs316009 (OR = 16.077, P = 0.013, FDR-P = 0.048) and NRF2 rs1806649 (OR = 35.328, P = 0.002, FDR-P = 0.011). Liver toxicity was associated with SLC22A2 rs316009 (OR = 10.057, P = 0.005, FDR-P = 0.030). IFNL3 rs4803219 (OR = 2.461, P = 0.008, FDR-P = 0.024) and NRF1 rs11557288 (OR = 2.106, P = 0.035, FDR-P = 0.035) were linked to syphilis co-infection. NRF1 rs6949152 (OR = 0.329, P = 0.002, FDR-P = 0.030) were associated with HBsAg positivity in HBV co-infection.These results were verified in dominant or recessive models.

CONCLUSION: This study establishes a foundation for using SNPs as predictive biomarkers for highly active antiretroviral therapy outcomes in Chinese HIV/AIDS patients, offering insights into personalized treatment strategies.

PMID:40466234 | DOI:10.1016/j.humimm.2025.111333

Nat Commun. 2025 Jun 4;16(1):5180. doi: 10.1038/s41467-025-60441-w.

ABSTRACT

Human airways contain specialized rare epithelial cells including CFTR-rich ionocytes that regulate airway surface physiology and chemosensory tuft cells that produce asthma-associated inflammatory mediators. Here, using a lung cell atlas of 311,748 single cell RNA-Seq profiles, we identify 687 ionocytes (0.45%). In contrast to prior reports claiming a lack of ionocytes in the small airways, we demonstrate that ionocytes are present in small and large airways in similar proportions. Surprisingly, we find only 3 mature tuft cells (0.002%), and demonstrate that previously annotated tuft-like cells are instead highly replicative progenitor cells. These tuft-ionocyte progenitor (TIP) cells produce ionocytes as a default lineage. However, Type 2 and Type 17 cytokines divert TIP cell lineage in vitro, resulting in the production of mature tuft cells at the expense of ionocyte differentiation. Our dataset thus provides an updated understanding of airway rare cell composition, and further suggests that clinically relevant cytokines may skew the composition of disease-relevant rare cells.

PMID:40467553 | DOI:10.1038/s41467-025-60441-w

J Cyst Fibros. 2025 Jun 3:S1569-1993(25)01488-2. doi: 10.1016/j.jcf.2025.05.003. Online ahead of print.

ABSTRACT

BACKGROUND: There are great changes in cystic fibrosis (CF) disease following introduction of modulator treatments. We aimed to focus on the evolution of hepatobiliary involvement following lumacaftor-ivacaftor (LI) and elexacaftor-tezacaftor-ivacaftor (ETI) initiation.

METHODS: A retrospective monocentric observational study included 62 CF children treated with CFTR modulators. Data were collected at initiation and after one year of treatment. The primary objective was to describe the evolution of hepatobiliary involvement under CFTR modulator treatment.

RESULTS: We identified hepatobiliary involvement before treatment in 37 patients (59.7 %). Fifteen had persistently (during >6 months) elevated liver enzymes (mostly ALT); 17 had abnormal ultrasound including 3 with nodular liver and 3 with pathological elastography; 5 had isolated splenomegaly. Biliary involvement was found in 19 patients. The evolution of hepatic parameters in the overall population was not significant (p > 0.05). However, we observed a trend towards improvement in laboratory values under treatment. There was only one inaugural diagnosis of nodular liver under LI and none under ETI. All patients had preserved liver function (PT>50 %).

CONCLUSIONS: We did not find a significant improvement or worsening of hepatobiliary involvement under CFTR modulators. We hypothesize that it could be stabilized with these treatments, but this will need confirmation through further studies with longer follow-up and larger cohorts. The other hypothesis proposed is that biological monitoring may not be an accurate assessment of the hepatobiliary response to modulators. This study supports the safety of CFTR modulator use.

PMID:40467431 | DOI:10.1016/j.jcf.2025.05.003

J Cyst Fibros. 2025 Jun 3:S1569-1993(25)01489-4. doi: 10.1016/j.jcf.2025.05.004. Online ahead of print.

ABSTRACT

BACKGROUND: Air travel may cause significant hypoxemia in patients with cystic fibrosis (CF). A pre-flight algorithm has previously been validated for patients with chronic obstructive pulmonary disease (COPD). No such tools are available for CF patients. The aim of this study was to evaluate if the pre-flight algorithm for COPD patients can be used by CF patients.

METHODS: In this prospective cross-sectional study, oxygen saturation at sea level (SpO2 SL) and during exercise (SpO2 6MWT) were used to evaluate whether CF patients a) are fit for flight without further assessment, b) require in-flight supplemental oxygen, or c) need further evaluation with hypoxia-altitude simulation test (HAST). HAST was used as reference method, and SpO2 HAST ≤85 % was the criterion for recommending in-flight supplemental oxygen.

RESULTS: 79 CF patients (41 men), age 38.0 ± 13.4 years, with FEV1 of 71±23 % of predicted underwent HAST (SpO2 HAST 89.2 ± 4.0 %). Categories for SpO2 SL were >95 % (N = 53), 92-95 % (N = 25), and <92 %, (N = 1), and the cut-off value for SpO2 6MWT was <84 %. HAST showed that CF patients with SpO2 SL >95 % combined with SpO2 6MWT ≥84 % can travel by air without further assessment. Supplemental oxygen is recommended if SpO2 SL is 92-95 % combined with SpO2 6MWT <84 %, or if SpO2 SL<92 %. Otherwise, HAST should be performed. Only 21 patients (27 %) would have needed referral to HAST. The algorithm correctly identified those who needed and those did not need in-flight supplemental oxygen.

CONCLUSIONS: The algorithm for COPD patients may be used in the pre-flight evaluation of adult CF patients.

CLINICALTRIALS: gov (NCT03843723).

PMID:40467430 | DOI:10.1016/j.jcf.2025.05.004

Endocr Pract. 2025 Jun 2:S1530-891X(25)00896-1. doi: 10.1016/j.eprac.2025.05.744. Online ahead of print.

ABSTRACT

OBJECTIVE: Hypoglycemia in hospitalized patients is a persistent adverse event. Three quality improvement interventions were implemented with the aim of reducing hypoglycemia. Each intervention was targeted at one component of typical inpatient insulin management (basal, prandial, and correction) to attempt to achieve this singular quality improvement aim.

METHODS: Incidence of hypoglycemia in non-obstetrics patients ≥ 19 years of age at a tertiary hospital receiving scheduled insulin before and after the implementation of quality improvement initiatives was compared. Incidence was defined as the number of unique patients with a hypoglycemic event in each month, divided by all admissions for that month. The interventions included integrating weight-based insulin guidance into the electronic medical record (EMR), the addition of a carbohydrate-limited diet, and increasing the threshold for correction insulin administration from 150 mg/dL to 180 mg/dL.

RESULTS: After implementation of the interventions, there was a significantly lower incidence of hypoglycemia associated with prandial insulin (p = 0.02) and correction insulin (p < 0.001). There was not a significant decrease in hypoglycemia associated with basal insulin in the overall sample (p =0.25). There was a significant decrease in a subgroup analysis focused on hospital-associated hyperglycemia and type 2 diabetes (via exclusion of patients with type 1 diabetes or cystic fibrosis-related diabetes) (p = 0.005). Notably, following the interventions, there was a reduction in institutional blood glucose readings within goal range (71-179 mg/dL), which presumably translates to an increase in hyperglycemia given the known decrease in hypoglycemia (p value <0.0001).

CONCLUSION: Through a multipronged approach consisting of three unique QI interventions - each targeting one aspect of inpatient insulin management - our academic institution was able to significantly reduce the number of inpatient hypoglycemic events.

PMID:40467034 | DOI:10.1016/j.eprac.2025.05.744

Genome Biol. 2025 Jun 4;26(1):156. doi: 10.1186/s13059-025-03619-1.

ABSTRACT

BACKGROUND: Whole-genome sequencing (WGS) data has facilitated genome-wide identification of rare noncoding variants. However, elucidating these variants' associations with complex diseases remains challenging. A previous study utilized a deep-learning-based framework and reported a significant brain-related association signal of autism spectrum disorder (ASD) detected from de novo noncoding variants in the Simons Simplex Collection (SSC) WGS cohort.

RESULTS: We revisit the reported significant brain-related ASD association signal attributed to deep-learning and show that local GC content can capture similar association signals. We further show that the association signal appears driven by variants from male proband-female sibling pairs that are upstream of assigned genes. We then develop Expression Neighborhood Sequence Association Study (ENSAS), which utilizes gene expression correlations and sequence information, to more systematically identify phenotype-associated variant sets. Applying ENSAS to the same set of de novo variants, we identify gene expression-based neighborhoods showing significant ASD association signal, enriched for synapse-related gene ontology terms. For these top neighborhoods, we also identify chromatin state annotations of variants that are predictive of the proband-sibling local GC content differences.

CONCLUSIONS: Overall, our work simplifies a previously reported ASD signal and provides new insights into associations of noncoding de novo mutations in ASD. We also present a new analytical framework for understanding disease impact of de novo mutations, applicable to other phenotypes.

PMID:40468385 | DOI:10.1186/s13059-025-03619-1

Head Face Med. 2025 Jun 4;21(1):44. doi: 10.1186/s13005-025-00521-w.

ABSTRACT

BACKGROUND: We aimed to comprehensively analyze the application of machine learning (ML) in dentistry and oral surgery using bibliometric methods to identify research trends, hotspots, and future directions.

METHODS: Publications related to ML in dentistry and oral surgery published between 2010 and 2024 were retrieved from the Science Citation Index Expanded by the Web of Science Core Collection (WoSCC). A total of 2234 unique publications were identified after screening. Bibliometric analysis was performed using the VOSviewer and CiteSpace software, focusing on parameters such as the number of publications, countries, institutions, journals, co-cited references, and keyword bursts.

RESULTS: The number of publications increased significantly from 2018 to 2024. China and the United States were the leading countries in terms of number of publications and citation counts. Prominent institutions include Seoul National University, Sichuan University, and Charite Universitätsmedizin Berlin. Journals such as BMC Oral Health and the Journal of Dentistry have a large number of publications. Analysis of the co-cited references revealed clusters related to disease diagnosis and risk prediction, treatment planning, clinical decision support systems, and dental education. Keyword bursts indicate the evolution of research focus from traditional machine learning algorithms to deep learning algorithms and the emerging importance of multimodal data and foundation models.

CONCLUSION: ML has made remarkable progress in dentistry and oral surgery. Although clinicians can benefit from the application of ML models in their practice, they should conduct comprehensive clinical validations to ensure the accuracy and reliability of these models. Moreover, challenges, such as data availability and security, algorithmic biases, and "black-box models", must be addressed. Future research should focus on integrating multimodal data and leveraging foundation models to improve the accuracy of diagnosis, treatment planning, and educational tools in dentistry and oral surgery.

PMID:40468381 | DOI:10.1186/s13005-025-00521-w

BMC Med Inform Decis Mak. 2025 Jun 4;25(1):207. doi: 10.1186/s12911-025-03027-2.

ABSTRACT

The detection and classification of blood cells are important in diagnosing and monitoring a variety of blood-related illnesses, such as anemia, leukemia, and infection, all of which may cause significant mortality. Accurate blood cell identification has a high clinical relevance in these patients because this would help to prevent false-negative diagnosis and to treat them in a timely and effective manner, thus reducing their clinical impacts.Our research aims to automate the process and eliminate manual efforts in blood cell counting. While our primary focus is on detection and classification, the output generated by our approach can be useful for disease prediction. This follows a two-step approach, where YOLO-based detection is first performed to locate blood cells, followed by classification using a hybrid CNN model to ensure accurate identification. We conducted a thorough and extensive comparison with other state-of-the-art models, including MobileNetV2, ShuffleNetV2, and DarkNet, for blood cell detection and classification. In terms of real-time performance, YOLOv10 outperforms other object detection models with better detection rates and classification accuracy. But MobileNetV2 and ShuffleNetV2 are more computationally efficient, which becomes more appropriate for resource-constrained environments. In contrast, DarkNet outperformed in terms of feature extraction performance, and the fine blood cell type classification. Additionally, an annotated blood cell data set was generated for this study. A diverse set of blood cell images with fine-grained annotations is contained in this dataset to make it useful for deep learning models training and evaluation. Because the present dataset will be an important resource for researchers and developers working on automatic blood cell detection and classification systems, we will make it publicly available under the open-access nature in order to accelerate the collaboration and progress in this field.

PMID:40468312 | DOI:10.1186/s12911-025-03027-2

BMC Med Inform Decis Mak. 2025 Jun 4;25(1):206. doi: 10.1186/s12911-025-03047-y.

ABSTRACT

BACKGROUND: Deep learning models have shown considerable potential to improve diagnostic accuracy across medical fields. Although YOLACT has demonstrated real-time detection and segmentation in non-medical datasets, its application in medical settings remains underexplored. This study evaluated the performance of a YOLACT-derived Real-time Polyp Delineation Model (RTPoDeMo) for real-time use on prospectively recorded colonoscopy videos.

METHODS: Twelve combinations of architectures, including Mask-RCNN, YOLACT, and YOLACT++, paired with backbones such as ResNet50, ResNet101, and DarkNet53, were tested on 2,188 colonoscopy images with three image resolution sizes. Dataset preparation involved pre-processing and segmentation annotation, with optimized image augmentation.

RESULTS: RTPoDeMo, using YOLACT-ResNet50, achieved 72.3 mAP and 32.8 FPS for real-time instance segmentation based on COCO annotations. The model performed with a per-image accuracy of 99.59% (95% CI: [99.45 - 99.71%]), sensitivity of 90.63% (95% CI: [78.95 - 93.64%]), specificity of 99.95% (95% CI: [99.93 - 99.97%]) and a F1-score of 0.94 (95% CI: [0.87-0.98]). In validation, out of 36 polyps detected by experts, RTPoDeMo missed only one polyp, compared to six missed by senior endoscopists. The model demonstrated good agreement with experts, reflected by a Cohen's Kappa coefficient of 0.72 (95% CI: [0.54-1.00], p < 0.0001).

CONCLUSIONS: Our model provides new perspectives in the adaptation of YOLACT to the real-time delineation of colorectal polyps. In the future, it could improve the characterization of polyps to be resected during colonoscopy.

PMID:40468304 | DOI:10.1186/s12911-025-03047-y

Med Phys. 2025 Jun 4. doi: 10.1002/mp.17910. Online ahead of print.

ABSTRACT

BACKGROUND: The reconstruction of a computed tomography (CT) image can be compromised by artifacts, which, in many cases, reduce the diagnostic value of the image. These artifacts often result from missing or corrupt regions in the projection data, for example, by truncation, metal, or limited angle acquisitions.

PURPOSE: In this work, we introduce a novel deep learning-based framework, latent space reconstruction (LSR), which enables correction of various types of artifacts arising from missing or corrupted data.

METHODS: First, we train a generative neural network on uncorrupted CT images. After training, we iteratively search for the point in the latent space of this network that best matches the compromised projection data we measured. Once an optimal point is found, forward-projection of the generated CT image can be used to inpaint the corrupted or incomplete regions of the measured raw data.

RESULTS: We used LSR to correct for truncation and metal artifacts. For the truncation artifact correction, images corrected by LSR show effective artifact suppression within the field of measurement (FOM), alongside a substantial high-quality extension of the FOM compared to other methods. For the metal artifact correction, images corrected by LSR demonstrate effective artifact reduction, providing a clearer view of the surrounding tissues and anatomical details.

CONCLUSIONS: The results indicate that LSR is effective in correcting metal and truncation artifacts. Furthermore, the versatility of LSR allows its application to various other types of artifacts resulting from missing or corrupt data.

PMID:40468155 | DOI:10.1002/mp.17910

Cell Regen. 2025 Jun 5;14(1):21. doi: 10.1186/s13619-025-00239-9.

ABSTRACT

Mechanical properties of cells have been proposed as potential biophysical markers for cell phenotypes and functions since they are vital for maintaining biological activities. However, current approaches used to measure single-cell mechanics suffer from low throughput, high technical complexity, and stringent equipment requirements, which cannot satisfy the demand for large-scale cell sample testing. In this study, we proposed to evaluate cell stiffness at the single-cell level using deep learning. The image-based deep learning models could non-invasively predict the stiffness ranges of mesenchymal stem cells (MSCs) and macrophages in situ with high throughput and high sensitivity. We further applied the models to evaluate MSC functions including senescence, stemness, and immunomodulatory capacity as well as macrophage diversity in phenotypes and functions. Our image-based deep learning models provide potential techniques and perspectives for cell-based mechanobiology research and clinical translation.

PMID:40468050 | DOI:10.1186/s13619-025-00239-9

Nat Comput Sci. 2025 Jun 4. doi: 10.1038/s43588-025-00810-z. Online ahead of print.

ABSTRACT

Quantum many-body methods provide a systematic route to computing electronic properties of molecules and materials, but high computational costs restrict their use in large-scale applications. Owing to the complexity in many-electron wavefunctions, machine learning models capable of capturing fundamental many-body physics remain limited. Here we present a deep learning framework targeting the many-body Green's function, which unifies predictions of electronic properties in ground and excited states, while offering physical insights into many-electron correlation effects. By learning the many-body perturbation theory or coupled-cluster self-energy from mean-field features, our graph neural network achieves competitive performance in predicting one- and two-particle excitations and quantities derivable from a one-particle density matrix. We demonstrate its high data efficiency and good transferability across chemical species, system sizes, molecular conformations and correlation strengths in bond breaking, through multiple molecular and nanomaterial benchmarks. This work opens up opportunities for utilizing machine learning to solve many-electron problems.

PMID:40468046 | DOI:10.1038/s43588-025-00810-z

Eur Spine J. 2025 Jun 4. doi: 10.1007/s00586-025-08995-7. Online ahead of print.

NO ABSTRACT

PMID:40468028 | DOI:10.1007/s00586-025-08995-7

Med Phys. 2025 Jun 4. doi: 10.1002/mp.17911. Online ahead of print.

ABSTRACT

BACKGROUND: Cone-beam CT (CBCT) scans that are affected by motion often require motion compensation to reduce artifacts or to reconstruct 4D (3D+time) representations of the patient. To do so, most existing strategies rely on some sort of gating strategy that sorts the acquired projections into motion bins. Subsequently, these bins can be reconstructed individually before further post-processing may be applied to improve image quality. While this concept is useful for periodic motion patterns, it fails in case of non-periodic motion as observed, for example, in irregularly breathing patients.

PURPOSE: To address this issue and to increase temporal resolution, we propose the deep single angle-based motion compensation (SAMoCo).

METHODS: To avoid gating, and therefore its downsides, the deep SAMoCo trains a U-net-like network to predict displacement vector fields (DVFs) representing the motion that occurred between any two given time points of the scan. To do so, 4D clinical CT scans are used to simulate 4D CBCT scans as well as the corresponding ground truth DVFs that map between the different motion states of the scan. The network is then trained to predict these DVFs as a function of the respective projection views and an initial 3D reconstruction. Once the network is trained, an arbitrary motion state corresponding to a certain projection view of the scan can be recovered by estimating DVFs from any other state or view and by considering them during reconstruction.

RESULTS: Applied to 4D CBCT simulations of breathing patients, the deep SAMoCo provides high-quality reconstructions for periodic and non-periodic motion. Here, the deviations with respect to the ground truth are less than 27 HU on average, while respiratory motion, or the diaphragm position, can be resolved with an accuracy of about 0.75 mm. Similar results were obtained for real measurements where a high correlation with external motion monitoring signals could be observed, even in patients with highly irregular respiration.

CONCLUSIONS: The ability to estimate DVFs as a function of two arbitrary projection views and an initial 3D reconstruction makes deep SAMoCo applicable to arbitrary motion patterns with single-view temporal resolution. Therefore, the deep SAMoCo is particularly useful for cases with unsteady breathing, compensation of residual motion during a breath-hold scan, or scans with fast gantry rotation times in which the data acquisition only covers a very limited number of breathing cycles. Furthermore, not requiring gating signals may simplify the clinical workflow and reduces the time needed for patient preparation.

PMID:40467957 | DOI:10.1002/mp.17911

Sci Rep. 2025 Jun 4;15(1):19665. doi: 10.1038/s41598-025-04676-z.

ABSTRACT

Breast cancer remains one of the most prevalent and life-threatening diseases among women worldwide, necessitating early and accurate detection methods. Traditional diagnostic approaches often face limitations in sensitivity and specificity, highlighting the need for advanced computational techniques. This study proposes BCDGAN, a novel deep learning framework integrating a Generative Adversarial Network (GAN) and a Hybrid Deep Learning (HDL) model for breast cancer detection using thermogram imagery. The aim is to enhance diagnostic accuracy by synthesizing critical regions of interest (ROIs) and leveraging deep feature extraction for improved classification performance. The proposed GAN-HDL-BCD method first employs a hybrid deep learning model to extract features from thermogram images, followed by a GAN-based approach to generate synthetic ROIs, augmenting the dataset and improving model generalization. Experimental validation using the DMR-IR benchmark dataset demonstrates that the proposed framework achieves an accuracy of 98.56%, outperforming conventional deep-learning models. These findings suggest that BCDGAN can be effectively integrated into clinical decision-support applications for thermogram-based breast cancer screening, offering a promising tool for enhancing early detection and patient outcomes.

PMID:40467954 | DOI:10.1038/s41598-025-04676-z

Int Immunopharmacol. 2025 Jun 3;161:114962. doi: 10.1016/j.intimp.2025.114962. Online ahead of print.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a rare, chronic, and progressive interstitial lung disease characterized by an unclear etiology and pathogenesis. Current anti-fibrotic therapies frequently fall short in effectively halting disease progression. A critical aspect of IPF involves the role of macrophages, which exhibit distinct polarized phenotypes that significantly influence the initiation and progression of fibrosis within the lung immune microenvironment. Recent evidence highlights the importance of the JAK-STAT signaling pathway in regulating macrophage polarization, suggesting that its inhibition may offer a promising therapeutic strategy for IPF. In this study, Ruxolitinib, a JAK1/2 inhibitor that is approved for the treatment of myelofibrosis, was investigated for its effects on pulmonary fibrosis for the first time. The in vivo studies were conducted utilizing a bleomycin-induced pulmonary fibrosis model, and in vitro experiments were induced pro-inflammatory and pro-fibrotic macrophage polarization using LPS/IFN-γ and IL-4/13, respectively. Notably, our findings reveal that Ruxolitinib diminishes pro-inflammatory polarization, thereby promoting a more favorable pulmonary inflammatory microenvironment. Furthermore, Ruxolitinib inhibits fibrotic macrophage polarization, effectively curtailing myofibroblast activation and displaying clear anti-fibrotic effects. The underlying regulatory mechanism of Ruxolitinib is through inhibition of JAK1/2-mediated STAT signaling, which interrupts the pathways leading to the polarization of fibrotic macrophages and the activation of pro-inflammatory macrophages. Collectively, these results underline the potential of Ruxolitinib as a therapeutic option for IPF treatment, representing a pivotal advance in addressing a disease that has previously evaded effective pharmacological intervention.

PMID:40466612 | DOI:10.1016/j.intimp.2025.114962