J Appl Clin Med Phys. 2025 Mar 14:e70012. doi: 10.1002/acm2.70012. Online ahead of print.

ABSTRACT

PURPOSE: Training deep learning dose prediction models for the latest cutting-edge radiotherapy techniques, such as AI-based nodal radiotherapy (AINRT) and Daily Adaptive AI-based nodal radiotherapy (DA-AINRT), is challenging due to limited data. This study aims to investigate the impact of transfer learning on the predictive performance of an existing clinical dose prediction model and its potential to enhance emerging radiotherapy approaches for head and neck cancer patients.

METHOD: We evaluated the impact and benefits of transfer learning by fine-tuning a Hierarchically Densely Connected U-net on both AINRT and DA-AINRT patient datasets, creating ModelAINRT (Study 1) and ModelDA-AINRT (Study 2). These models were compared against pretrained and baseline models trained from scratch. In Study 3, both fine-tuned models were tested using DA-AINRT patients' final adaptive sessions to assess ModelAINRT 's effectiveness on DA-AINRT patients, given that the primary difference is planning target volume (PTV) sizes between AINRT and DA-AINRT.

RESULT: Studies 1 and 2 revealed that the transfer learning model accurately predicted the mean dose within 0.71% and 0.86% of the prescription dose on the test data. This outperformed the pretrained and baseline models, which showed PTV mean dose prediction errors of 2.29% and 1.1% in Study 1, and 2.38% and 2.86% in Study 2 (P < 0.05). Additionally, Study 3 demonstrated significant improvements in PTV dose prediction error with ModelDA-AINRT, with a mean dose difference of 0.86% ± 0.73% versus 2.26% ± 1.65% (P < 0.05). This emphasizes the importance of training models for specific patient cohorts to achieve optimal outcomes.

CONCLUSION: Applying transfer learning to dose prediction models significantly improves prediction accuracy for PTV while maintaining similar dose performance in predicting organ-at-risk (OAR) dose compared to pretrained and baseline models. This approach enhances dose prediction models for novel radiotherapy methods with limited training data.

PMID:40087841 | DOI:10.1002/acm2.70012

Magn Reson Med. 2025 Mar 14. doi: 10.1002/mrm.30488. Online ahead of print.

ABSTRACT

PURPOSE: To develop a quantitative multislice chemical exchange saturation transfer (CEST) schedule optimization and pulse sequence that reduces the loss of sensitivity inherent to multislice sequences.

METHODS: A deep learning framework was developed for simultaneous optimization of scan parameters and slice order. The optimized sequence was tested in numerical simulations against a random schedule and an optimized single-slice schedule. The scan efficiency of each schedule was quantified. Three healthy subjects were scanned with the proposed sequence. Regions of interest in white matter (WM) and gray matter (GM) were defined. The sequence was compared with the single-slice sequence in vivo and differences quantified using Bland-Altman plots. Test-retest reproducibility was assessed, and the Lin's concordance correlation coefficient (CCC) was calculated for WM and GM. Intersubject variability was also measured with the CCC. Feasibility of whole-brain clinical imaging was tested using a multislab acquisition in 1 subject.

RESULTS: The optimized multislice sequence yielded a lower mean error than the random schedule for all tissue parameters and a lower error than the optimized single-slice schedule for four of six parameters. The optimized multislice sequence provided the highest scan efficiency. In vivo tissue-parameter values obtained with the proposed sequence agreed well with those of the optimized single-slice sequence and prior studies. The average WM/GM CCC was 0.8151/0.7779 for the test-retest scans and 0.7792/0.7191 for the intersubject variability experiment.

CONCLUSION: A multislice schedule optimization framework and pulse sequence were demonstrated for quantitative CEST. The proposed approach enables accurate and reproducible whole-brain quantitative CEST imaging in clinically relevant scan times.

PMID:40087839 | DOI:10.1002/mrm.30488

Biomark Res. 2025 Mar 14;13(1):45. doi: 10.1186/s40364-025-00758-2.

ABSTRACT

Artificial intelligence (AI) can transform drug discovery and early drug development by addressing inefficiencies in traditional methods, which often face high costs, long timelines, and low success rates. In this review we provide an overview of how to integrate AI to the current drug discovery and development process, as it can enhance activities like target identification, drug discovery, and early clinical development. Through multiomics data analysis and network-based approaches, AI can help to identify novel oncogenic vulnerabilities and key therapeutic targets. AI models, such as AlphaFold, predict protein structures with high accuracy, aiding druggability assessments and structure-based drug design. AI also facilitates virtual screening and de novo drug design, creating optimized molecular structures for specific biological properties. In early clinical development, AI supports patient recruitment by analyzing electronic health records and improves trial design through predictive modeling, protocol optimization, and adaptive strategies. Innovations like synthetic control arms and digital twins can reduce logistical and ethical challenges by simulating outcomes using real-world or virtual patient data. Despite these advancements, limitations remain. AI models may be biased if trained on unrepresentative datasets, and reliance on historical or synthetic data can lead to overfitting or lack generalizability. Ethical and regulatory issues, such as data privacy, also challenge the implementation of AI. In conclusion, in this review we provide a comprehensive overview about how to integrate AI into current processes. These efforts, although they will demand collaboration between professionals, and robust data quality, have a transformative potential to accelerate drug development.

PMID:40087789 | DOI:10.1186/s40364-025-00758-2

Biomed Eng Online. 2025 Mar 14;24(1):34. doi: 10.1186/s12938-025-01336-1.

ABSTRACT

BACKGROUND: In recent years, artificial intelligence and machine learning algorithms have been used more extensively to diagnose diabetic retinopathy and other diseases. Still, the effectiveness of these methods has not been thoroughly investigated. This study aimed to evaluate the performance and limitations of machine learning and deep learning algorithms in detecting diabetic retinopathy.

METHODS: This study was conducted based on the PRISMA checklist. We searched online databases, including PubMed, Scopus, and Google Scholar, for relevant articles up to September 30, 2023. After the title, abstract, and full-text screening, data extraction and quality assessment were done for the included studies. Finally, a meta-analysis was performed.

RESULTS: We included 76 studies with a total of 1,371,517 retinal images, of which 51 were used for meta-analysis. Our meta-analysis showed a significant sensitivity and specificity with a percentage of 90.54 (95%CI [90.42, 90.66], P < 0.001) and 78.33% (95%CI [78.21, 78.45], P < 0.001). However, the AUC (area under curvature) did not statistically differ across studies, but had a significant figure of 0.94 (95% CI [- 46.71, 48.60], P = 1).

CONCLUSIONS: Although machine learning and deep learning algorithms can properly diagnose diabetic retinopathy, their discriminating capacity is limited. However, they could simplify the diagnosing process. Further studies are required to improve algorithms.

PMID:40087776 | DOI:10.1186/s12938-025-01336-1

BMC Pulm Med. 2025 Mar 14;25(1):118. doi: 10.1186/s12890-025-03572-6.

ABSTRACT

OBJECTIVE: Respiratory muscle dysfunction in patients with idiopathic pulmonary fibrosis (IPF) is a big challenge for treatment and rehabilitation. To quantitatively assess diaphragm and chest wall dysfunction using dynamic Magnetic Resonance Imaging (Dyn-MRI) in patients with IPF.

METHODS: Ninety-six patients with IPF and 50 gender- and age-matched controls were prospectively included and underwent D-MRI with a dynamic fast spoiled gradient-recalled echo sequence. Respiratory muscles function were assessed with thoracic anterior-posterior (AP), left-right (LR), cranial-caudal (CC) metrics. Moreover, lung area ratios, height (DH), and area (DA) of diaphragm curvature between end-inspiration and end-expiration during both quiet and deep breathing.

RESULTS: During quiet breathing, the functional metrics of the diaphragm and chest wall were comparable between IPF patients and controls. However, during deep breathing, IPF patients exhibited significantly reduced ratios of AP, CC, and lung area compared to controls. Moreover, the median ratios of DH and DA were higher in IPF patients than in controls (DH: 0.96 vs. 0.81, p < 0.001; DA: 1.00 vs. 0.90, p < 0.001). Furthermore, the ratios of AP, CC, and lung area during deep breathing were found to correlate with pulmonary function, total lung volume, and 6-minute walk distance.

CONCLUSION: D-MRI demonstrated dysfunction in the diaphragm and chest wall among IPF patients, with respiratory muscle dysfunction showing a correlation with the severity of disease.

TRIAL REGISTRATION: This article presents a prospective observational study that does not include the outcomes of any healthcare interventions on human participants. The study was registered on September 11, 2018, under the registration number NCT03666234.

PMID:40087606 | DOI:10.1186/s12890-025-03572-6

Genetics. 2025 Mar 15:iyaf040. doi: 10.1093/genetics/iyaf040. Online ahead of print.

ABSTRACT

Autism spectrum disorder (ASD) is a neurodevelopmental disorder that affects nearly 3% of children and has a strong genetic component. While hundreds of ASD risk genes have been identified through sequencing studies, the genetic heterogeneity of ASD makes identifying additional risk genes using these methods challenging. To predict candidate ASD risk genes, we developed a simple machine learning model, ASiDentify (ASiD), using human genomic, RNA- and protein-based features. ASiD identified over 1,300 candidate ASD risk genes, over 300 of which have not been previously predicted. ASiD made accurate predictions of ASD risk genes using six features predictive of ASD risk gene status, including mutational constraint, synapse localization and gene expression in neurons, astrocytes and non-brain tissues. Particular functional groups of proteins found to be strongly implicated in ASD include RNA-binding proteins and chromatin regulators. We constructed additional logistic regression models to make predictions and assess informative features specific to RNA-binding proteins, including mutational constraint, or chromatin regulators, for which both expression level in excitatory neurons and mutational constraint were informative. The fact that RNA-binding proteins and chromatin regulators had informative features distinct from all protein-coding genes, suggests that specific biological pathways connect risk genes with different molecular functions to ASD.

PMID:40088463 | DOI:10.1093/genetics/iyaf040

STAR Protoc. 2025 Mar 14;6(2):103699. doi: 10.1016/j.xpro.2025.103699. Online ahead of print.

ABSTRACT

As the study of central control of multiple organ function becomes more prominent, there is an increasing need for the collection of fixed brain and unfixed organs and tissues from the same experimental animal. Here, we present a protocol for performing carotid artery cannulation, organ and tissue collection, in situ brain perfusion and fixation, and brain dissection in mice. We describe steps for cannulating the carotid artery, harvesting the heart and other organs, and perfusing, fixing, and dissecting the brain.

PMID:40088450 | DOI:10.1016/j.xpro.2025.103699

Health Econ Rev. 2025 Mar 15;15(1):20. doi: 10.1186/s13561-025-00609-8.

ABSTRACT

BACKGROUND: Fertility care represents a financial burden on patients and healthcare services alike and can represent a barrier to entry for many couples. Controlled ovarian stimulation (COH) is routinely used as part of in vitro fertilization and intracytoplasmic sperm injection (ICSI) procedures, as such the use of gonadotropins is a major contributing factor to the cost of the procedure. Recent studies have shown that myo-Inositol (myo-ins) may reduce the amount of gonadotrophins required in assisted reproductive technology (ART) procedures. This retrospective study measured the effect of myo-ins on the number of recombinant follicular stimulating hormone (rFSH) units used in IVF and ICSI and the relative cost to verify if this may be a cost saving strategy. We also investigated the oocyte and embryo quality, implantation rate, abortion rate, clinical pregnancy, and ovarian hyperstimulation syndrome.

METHODS: A total of 300 women undergoing either IVF or ICSI were distributed between two distinct and equal patient groups of 150 women. In control group (group A), folic acid (FA) alone was prescribed, meanwhile the treated group (group B) were prescribed FA, myo-Inositol (myo-ins) and alpha-lactalbumin (α-LA), both groups started this oral treatment in the middle of the luteal phase.

RESULTS: Myo-Ins supplementation in the treatment group significantly reduced the number of units of rFSH used in COH vs. the control group (2526 vs. 1647, p < 0.05); however, no changes were seen in other measured outcomes, likely due to the short treatment period.

CONCLUSIONS: The use of myo-Ins presents a safe method for reducing the amount and subsequent costs of rFSH usage in ART protocols.

TRIAL REGISTRATION: The trial was retrospectively registered with the Institutional Review Board of ALMA RES IVF Center, trial number n°2/2024.

PMID:40088331 | DOI:10.1186/s13561-025-00609-8

New Phytol. 2025 Mar 15. doi: 10.1111/nph.70068. Online ahead of print.

ABSTRACT

The green seaweed Ulva compressa is a promising model for functional biology. In addition to historical research on growth and development, -omics data and molecular tools for stable transformation are available. However, more efficient tools are needed to study gene function. Here, we expand the molecular toolkit for Ulva. We screened the survival of Ulva and its mutualistic bacteria on 14 selective agents and established that Blasticidin deaminases (BSD or bsr) can be used as selectable markers to generate stable transgenic lines. We show that Cas9 and Cas12a RNPs are suitable for targeted mutagenesis and can generate genomic deletions of up to 20 kb using the marker gene ADENINE PHOSPHORIBOSYLTRANSFERASE (APT). We demonstrate that the targeted insertion of a selectable marker via homology-directed repair or co-editing with APT is possible for nonmarker genes. We evaluated 31 vector configurations and found that the bicistronic fusion of Cas9 to a resistance marker or the incorporation of introns in Cas9 led to the most mutants. We used this to generate mutants in three nonmarker genes using a co-editing strategy. This expanded molecular toolkit now enables us to reliably make gain- and loss-of-function mutants; additional optimizations will be necessary to allow for vector-based multiplex genome editing in Ulva.

PMID:40088038 | DOI:10.1111/nph.70068

Mol Microbiol. 2025 Mar 14. doi: 10.1111/mmi.15357. Online ahead of print.

ABSTRACT

The opportunistic pathogen Pseudomonas aeruginosa relies on a large collection of two-component regulatory systems (TCSs) to sense and adapt to changing environments. Among them, the Roc (regulation of cup) system is a one-of-a-kind network of branched TCSs, composed of two histidine kinases (HKs-RocS1 and RocS2) interacting with three response regulators (RRs-RocA1, RocR, and RocA2), which regulate virulence, antibiotic resistance, and biofilm formation. Based on extensive work on the Roc system, previous data suggested the existence of other key regulators yet to be discovered. In this work, we identified PA4080, renamed RocA3, as a fourth RR that is activated by RocS1 and RocS2 and that positively controls the expression of the cupB operon. Comparative genomic analysis of the locus identified a gene-rocR3-adjacent to rocA3 in a subpopulation of strains that encodes a protein with structural and functional similarity to the c-di-GMP phosphodiesterase RocR. Furthermore, we identified a fourth branch of the Roc system consisting of the PA2583 HK, renamed RocS4, and the Hpt protein HptA. Using a bacterial two-hybrid system, we showed that RocS4 interacts with HptA, which in turn interacts with RocA1, RocA2, and RocR3. Finally, we mapped the pangenomic RRs repertoire, establishing a comprehensive view of the plasticity of such regulators among clades of the species. Overall, our work provides a comprehensive inter-species definition of the Roc system, nearly doubling the number of proteins known to be involved in this interconnected network of TCSs controlling pathogenicity in Pseudomonas species.

PMID:40087830 | DOI:10.1111/mmi.15357

Genome Biol. 2025 Mar 14;26(1):56. doi: 10.1186/s13059-025-03499-5.

ABSTRACT

BACKGROUND: The molecular underpinnings of organ dysfunction in severe COVID-19 and its potential long-term sequelae are under intense investigation. To shed light on these in the context of liver function, we perform single-nucleus RNA-seq and spatial transcriptomic profiling of livers from 17 COVID-19 decedents.

RESULTS: We identify hepatocytes positive for SARS-CoV-2 RNA with an expression phenotype resembling infected lung epithelial cells, and a central role in a pro-fibrotic TGFβ signaling cell-cell communications network. Integrated analysis and comparisons with healthy controls reveal extensive changes in the cellular composition and expression states in COVID-19 liver, providing the underpinning of hepatocellular injury, ductular reaction, pathologic vascular expansion, and fibrogenesis characteristic of COVID-19 cholangiopathy. We also observe Kupffer cell proliferation and erythrocyte progenitors for the first time in a human liver single-cell atlas. Despite the absence of a clinical acute liver injury phenotype, endothelial cell composition is dramatically impacted in COVID-19, concomitantly with extensive alterations and profibrogenic activation of reactive cholangiocytes and mesenchymal cells.

CONCLUSIONS: Our atlas provides novel insights into liver physiology and pathology in COVID-19 and forms a foundational resource for its investigation and understanding.

PMID:40087773 | DOI:10.1186/s13059-025-03499-5

Commun Biol. 2025 Mar 15;8(1):439. doi: 10.1038/s42003-025-07875-6.

ABSTRACT

Large-scale GWAS studies have uncovered hundreds of genomic loci linked to facial and brain shape variation, but only tens associated with cranial vault shape, a largely overlooked aspect of the craniofacial complex. Surrounding the neocortex, the cranial vault plays a central role during craniofacial development and understanding its genetics are pivotal for understanding craniofacial conditions. Experimental biology and prior genetic studies have generated a wealth of knowledge that presents opportunities to aid further genetic discovery efforts. Here, we use the conditional FDR method to leverage GWAS data of facial shape, brain shape, and bone mineral density to enhance SNP discovery for cranial vault shape. This approach identified 120 independent genomic loci at 1% FDR, nearly tripling the number discovered through unconditioned analysis and implicating crucial craniofacial transcription factors and signaling pathways. These results significantly advance our genetic understanding of cranial vault shape and craniofacial development more broadly.

PMID:40087503 | DOI:10.1038/s42003-025-07875-6

Nat Commun. 2025 Mar 15;16(1):2545. doi: 10.1038/s41467-025-57809-3.

ABSTRACT

In photosynthetic organisms, light acts as an environmental signal to control their development and physiology, as well as energy source to drive the conversion of CO2 into carbohydrates used for growth or storage. The main storage carbohydrate in green algae is starch, which accumulates during the day and is broken down at night to meet cellular energy demands. The signaling role of light quality in the regulation of starch accumulation remains unexplored. Here, we identify PHOTOTROPIN-MEDIATED SIGNALING KINASE 1 (PMSK1) as a key regulator of starch metabolism in Chlamydomonas reinhardtii. In its phosphorylated form (PMSK1-P), it activates GLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE (GAP1), promoting starch biosynthesis. We show that blue light, perceived by PHOTOTROPIN, induces PMSK1 dephosphorylation that in turn represses GAP1 mRNA levels and reduces starch accumulation. These findings reveal a previously uncharacterized blue light-mediated signaling pathway that advances our understanding of photoreceptor-controlled carbon metabolism in microalgae.

PMID:40087266 | DOI:10.1038/s41467-025-57809-3

Mol Biotechnol. 2025 Mar 14. doi: 10.1007/s12033-025-01397-6. Online ahead of print.

ABSTRACT

Drug-induced gingival enlargement (DIGE) is an abnormal overgrowth that may occur as a side effect in some patients when calcium channel blockers, immunosuppressants, or anticonvulsants are taken. The prevalence of DIGE was shown to be 70% for phenytoin (30% for other anticonvulsant medicines) and 50-80% for cyclosporine. The usage of these medications is increasing as new indications emerge. These drugs act through a common mechanism of action at the cellular level by inhibiting intracellular calcium influx. DIGE is characterized by the presence of varied quantities of inflammatory infiltrates, primarily plasma cells, and an excessive build-up of extracellular matrix like-collagen. Fibroblasts, the cells responsible for collagen synthesis, may become hyperactive, leading to the excessive production of collagen fibers. This increased collagen content can result in the enlargement of gingival tissues. As collagen deposits increase, it hinders normal oral care routines, masticatory processes, and esthetics. In this study, we compared the cytotoxicity of phenytoin, gabapentin, and cyclosporine on gingival fibroblast cells using the methyl thiazolyl-tetrazolium assay to understand their effect on gingival fibroblast cells. Phenytoin had the greatest half-maximal inhibitory concentration (IC50) with a value of 305.78 µg/ml, followed by gabapentin with a value of 260.44 µg/ml and cyclosporin with a value of 243.79 µg/ml. Understanding the cytotoxic thresholds of these medications is essential for improving patient outcomes and minimizing the incidence of gingival enlargement in those requiring long-term therapy. According to the study, cytotoxicity increases along with medication concentration. These findings will assist medical professionals in selecting the drug that poses the least risk of adverse effects on gingival health, ultimately guiding more informed prescribing practices.

PMID:40087263 | DOI:10.1007/s12033-025-01397-6

Do you want to learn more about how applications will be reviewed under the simplified review framework for most research project grants? Whether you’re a peer reviewer awaiting training or a researcher preparing an application, we invite you to review our reviewer guidance page for the simplified review framework. There, you can learn more about how reviewers will be instructed to evaluate applications under the new framework.  

Just remember, the simplified review framework is a new way of reviewing the same research strategy and is not expected to change how applicants prepare applications. See our applicant guide for more information on navigating the transition to the simplified review framework. 

Other Information 

• Simplifying Review of Research Project Grant Applications 

• Simplified Review Framework FAQs 

• NIH All About Grants Podcast – Simplified Review Framework 

Mol Genet Metab. 2025 Apr;144(4):109073. doi: 10.1016/j.ymgme.2025.109073. Epub 2025 Mar 1.

ABSTRACT

Rare diseases affect over 400 million people worldwide, with approved treatment available for less than 6 % of these diseases. Drug repurposing is a key strategy in the development of therapies for rare disease patients with large unmet medical needs. The process of repurposing drugs compared to novel drug development is a time-saving and cost-efficient method potentially resulting in higher success rates. To accelerate and ensure sustainability in therapy development for rare neurometabolic, neurological, and neuromuscular diseases, an international consortium SIMilarities in clinical and molecular PATHology (SIMPATHIC) has been established where we move away from the one drug one disease concept and move towards one drug targeting a pathomechanism shared between diseases, by applying parallel preclinical and clinical drug development. Here the consortium describes accelerators of drug repurposing pursued by the consortium, including 1) co-creation, 2) patient empowerment, 3) use of standardized induced pluripotent stem cell (iPSC)-derived disease models and cellular and molecular profiling, 4) high-throughput drug screening in neurons, 5) innovative clinical trial design, and 6) selection of appropriate exploitation and patient access models. In this way, a fast and effective drug repurposing pathway for several rare diseases will be established to reduce time from discovery to patient access.

PMID:40086177 | DOI:10.1016/j.ymgme.2025.109073

Comput Biol Med. 2025 Mar 13;189:109894. doi: 10.1016/j.compbiomed.2025.109894. Online ahead of print.

ABSTRACT

Advanced imaging techniques, like magnetic resonance imaging (MRI), have revolutionised cardiovascular disease diagnosis and monitoring in humans and animal models. Real-time (RT) MRI, which can capture a single slice during each consecutive heartbeat while the animal or patient breathes continuously, generates large data sets that necessitate automatic myocardium segmentation to fully exploit these technological advancements. While automatic segmentation is common in human adults, it remains underdeveloped in preclinical animal models. In this study, we developed and trained a fully automated 2D convolutional neural network (CNN) for segmenting the left and right ventricles and the myocardium in non-human primates (NHPs) using RT cardiac MR images of rhesus macaques, in the following referred to as PrimUNet. Based on the U-Net framework, PrimUNet achieved optimal performance with a learning rate of 0.0001, an initial kernel size of 64, a final kernel size of 512, and a batch size of 32. It attained an average Dice score of 0.9, comparable to human studies. Testing PrimUNet on additional RT MRI data from rhesus macaques demonstrated strong agreement with manual segmentation for left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), and left ventricular myocardial volume (LVMV). It also performs well on cine MRI data of rhesus macaques and acceptably on those of baboons. PrimUNet is well-suited for automatically segmenting extensive RT MRI data, facilitating strain analyses of individual heartbeats. By eliminating human observer variability, PrimUNet enhances the reliability and reproducibility of data analysis in animal research, thereby advancing translational cardiovascular studies.

PMID:40086292 | DOI:10.1016/j.compbiomed.2025.109894

Comput Biol Med. 2025 Mar 13;189:109924. doi: 10.1016/j.compbiomed.2025.109924. Online ahead of print.

ABSTRACT

- In recent times, the electrocardiogram (ECG) has been considered as a significant and effective screening mode in clinical practice to assess cardiac arrhythmias. Precise feature extraction and classification are considered as essential concerns in the automated prediction of heart disease. A deep bi-directional capsule network (Bi-CapsNet) uses a new method based on an intelligent deep learning (DL) classifier model to make the classification process very accurate. Initially, the input ECG signal data are acquired and the preprocessing steps such as DC drift, normalization, LPF filtering, spectrogram analysis, and artifact removal are applied. After preprocessing the data, the Deep Ensemble CNN-RNN approach is employed for feature extraction. Finally, the Deep Bi-CapsNet model is used to predict and classify the cardiac arrhythmia. For performance validation, the dataset is referred to the MIT-BIH arrhythmia database, which selects five different types of arrhythmias from the ECG waveform to estimate the proposed model. Various ECG arrhythmia categories, including Normal (NOR), Right Bundle Branch Block (RBBB), Premature Ventricular Contraction (PVC), Atrial Premature Beat (APB), and Left Bundle Branch Block (LBBB) have been identified. For performance analysis, the metrics such as precision, accuracy, F1-score, error rate, sensitivity, false positive rate, specificity, Mathew coefficient, Kappa coefficient, and outcomes are included and compared with the traditional methods to validate the effectiveness of the implemented scheme. The proposed scheme has achieved an overall accuracy rate of approximately 97.19 % compared to the traditional deep learning models like CNN (89.87 %), FTBO (85 %), and Capsule Network (97.0 %). The comparison results indicate that the proposed hybrid model outperforms these traditional models.

PMID:40086290 | DOI:10.1016/j.compbiomed.2025.109924

Mol Genet Metab. 2025 Mar 1;144(4):109073. doi: 10.1016/j.ymgme.2025.109073. Online ahead of print.

ABSTRACT

Rare diseases affect over 400 million people worldwide, with approved treatment available for less than 6 % of these diseases. Drug repurposing is a key strategy in the development of therapies for rare disease patients with large unmet medical needs. The process of repurposing drugs compared to novel drug development is a time-saving and cost-efficient method potentially resulting in higher success rates. To accelerate and ensure sustainability in therapy development for rare neurometabolic, neurological, and neuromuscular diseases, an international consortium SIMilarities in clinical and molecular PATHology (SIMPATHIC) has been established where we move away from the one drug one disease concept and move towards one drug targeting a pathomechanism shared between diseases, by applying parallel preclinical and clinical drug development. Here the consortium describes accelerators of drug repurposing pursued by the consortium, including 1) co-creation, 2) patient empowerment, 3) use of standardized induced pluripotent stem cell (iPSC)-derived disease models and cellular and molecular profiling, 4) high-throughput drug screening in neurons, 5) innovative clinical trial design, and 6) selection of appropriate exploitation and patient access models. In this way, a fast and effective drug repurposing pathway for several rare diseases will be established to reduce time from discovery to patient access.

PMID:40086177 | DOI:10.1016/j.ymgme.2025.109073

Respir Med. 2025 Mar 13:108016. doi: 10.1016/j.rmed.2025.108016. Online ahead of print.

ABSTRACT

Cystic Fibrosis (CF) is a life-limiting, inherited condition in which a novel class of oral medicine, CFTR modulators, has revolutionised symptoms and health indicators, providing an opportunity to evaluate traditional treatment regimens with the hope of reducing burden. Additionally, there is cautious optimism that life expectancy for people with CF born today could ultimately compare to that of the general population. Given this potential, there is a need and requirement to optimise treatment to balance burden with the best clinical outcomes for each person with CF in an individualised manner. Personalised data-Linkage to Understand Treatment Optimisation (PLUTO) is a clinical framework, developed within the 14-Centre UK CFHealthHub Learning Health System collaborative, designed for use at an individual level for people with CF taking CFTR modulators. The PLUTO framework encourages use of two routinely collected clinical outcome measure (FEV1 and BMI) to determine health status. Where FEV1 or BMI trends suggest that optimal health outcomes are not being achieved for a person with CF, PLUTO supports consideration of adherence to both CFTR modulators and inhaled therapy to help guide the next steps. PLUTO is designed to support people with CF and their clinical teams to individualise care and optimise outcomes for those taking CFTR modulators, using data available in routine clinical encounters.

PMID:40087032 | DOI:10.1016/j.rmed.2025.108016