J Clin Densitom. 2025 Jan-Mar;28(1):101559. doi: 10.1016/j.jocd.2024.101559. Epub 2024 Dec 28.

ABSTRACT

The 24th Annual Santa Fe Bone Symposium (SFBS) was held in Santa Fe, New Mexico, USA, on August 2-3, 2024. This was a "hybrid" meeting, with in-person and real-time remote participants representing a broad range of geographical locations and medical disciplines. The focus was on new developments in the care of patients with osteoporosis, other metabolic bone diseases, and inherited skeletal disorders. The most current medical evidence was presented and discussed with consideration of implications for patient management. Topics included an update on clinical uses of osteoanabolic agents, management of patients discontinuing denosumab, bone health optimization for orthopedic surgery, estrogen and testosterone in the management of osteoporosis, osteoporosis treatment in the very old, overview of rare bone diseases, treat-to-target for osteoporosis, and a progress report on global activities of Bone Health ECHO. There were two highly interactive faculty panel discussions - one with case presentations by attendees and another with open microphone for all topics of interest. Endocrinology fellows, selected from attendees of the Santa Fe Fellows Workshop on Metabolic Bone Diseases, held the two days preceding the SFBS, participated with presentations of oral abstracts. Ancillary events addressed modern approaches to menopause and bone health, case studies of management of patients at very high fracture risk, and management of patients with rare bone diseases, such as hypophosphatasia, fibrodysplasia ossificans progressiva, X-linked hypophosphatemia, and hypoparathyroidism. These proceedings of the SFBS present the clinical highlights of the plenary sessions and the discussions that followed.

PMID:39826229 | DOI:10.1016/j.jocd.2024.101559

Commun Biol. 2025 Jan 18;8(1):85. doi: 10.1038/s42003-025-07519-9.

ABSTRACT

Phosphodiesterase-5 (PDE5) inhibitors have shown promise as anti-cancer agents in malignancies. However, their specific effects on gastric cancer (GC) and the underlying mechanisms remain elusive. Our aim was to investigate this by combining evidence from population-based studies with data obtained from in vivo and in vitro experiments. By combing a couple of nationwide Swedish registers, GC patients who received PDE5 inhibitors were compared to matched controls while adjusting for confounding factors. The anti-tumor effect and mechanism of the PDE5 inhibitor sildenafil were evaluated via using tumor cells, patient-derived tumor organoids and xenograft animal models in GC. A total of 161 Swedish GC patients from a nationwide population-based cohort who received post-diagnostic PDE5 inhibitors demonstrated lower cancer-specific mortality compared to the controls (HR = 0.66, 95% CI = 0.47-0.92, P = 0.016). Functionally, the PDE5 inhibitor sildenafil exhibited the suppressive ability to prevent oncogenic growth in GC. Mechanistically, sildenafil restrained GC growth by directly activating PKG through PDE5 inhibition for regulating c-MYC expression via its phosphorylation and ubiquitination degradation, thereby suppressing c-MYC stability for IL-6 transcription within the downstream IL-6/JAK/STAT3 signalling pathway. The PDE5 inhibitor sildenafil may serve as a promising adjuvant for GC therapy if further randomized clinical trials confirm its efficacy.

PMID:39827331 | DOI:10.1038/s42003-025-07519-9

Cancer Lett. 2025 Jan 16:217437. doi: 10.1016/j.canlet.2025.217437. Online ahead of print.

NO ABSTRACT

PMID:39826666 | DOI:10.1016/j.canlet.2025.217437

Comput Methods Programs Biomed. 2025 Jan 13;261:108604. doi: 10.1016/j.cmpb.2025.108604. Online ahead of print.

ABSTRACT

BACKGROUND: Alzheimer's disease (AD), the most prevalent form of dementia, remains enigmatic in its origins despite the widely accepted "amyloid hypothesis," which implicates amyloid-beta peptide aggregates in its pathogenesis and progression. Despite advancements in technology and healthcare, the incidence of AD continues to rise. The traditional drug development process remains time-consuming, often taking years to bring an AD treatment to market. Drug repurposing has emerged as a promising strategy for developing cost-effective and efficient therapeutic options by identifying new uses for existing approved drugs, thus accelerating drug development.

OBJECTIVES: This study aimed to examine two key drug repurposing methodologies in general diseases and specifically in AD, which are artificial intelligent (AI) approach and molecular docking approach. In addition, the hybrid approach that integrates AI with molecular docking techniques will be explored too.

METHODOLOGY: This study systematically compiled a comprehensive collection of relevant academic articles, scientific papers, and research studies which were published up until November 2024 (as of the writing of this review paper). The final selection of papers was filtered to include studies related to Alzheimer's disease and general diseases, and then categorized into three groups: AI articles, molecular docking articles, and hybrid articles.

RESULTS: As a result, 331 papers were identified that employed AI for drug repurposing in general diseases, and 58 papers focused specifically in AD. For molecular docking in drug repurposing, 588 papers addressed general diseases, while 46 papers were dedicated to AD. The hybrid approach combining AI and molecular docking in drug repurposing has 52 papers for general diseases and 9 for AD. A comparative review was done across the methods, results, strengths, and limitations in those studies. Challenges of drug repurposing in AD are explored and future prospects are proposed.

DISCUSSION AND CONCLUSION: Drug repurposing emerges as a compelling and effective strategy within AD research. Both AI and molecular docking methods exhibit significant potential in this domain. AI algorithms yield more precise predictions, thus facilitating the exploration of new therapeutic avenues for existing drugs. Similarly, molecular docking techniques revolutionize drug-target interaction modelling, employing refined algorithms to screen extensive drug databases against specific target proteins. This review offers valuable insights for guiding the utilization of AI, molecular docking, or their hybrid in AD drug repurposing endeavors. The hope is to speed up the timeline of drug discovery which could improve the therapeutic approach to AD.

PMID:39826482 | DOI:10.1016/j.cmpb.2025.108604

Sci Rep. 2025 Jan 18;15(1):2390. doi: 10.1038/s41598-025-86211-8.

ABSTRACT

Burkholderia cenocepacia H111 is an obligate aerobic bacterium which has been isolated from a cystic fibrosis (CF) patient. In CF lungs the environment is considered micro-oxic or even oxygen-depleted due to bacterial activities and limited oxygen diffusion in the mucus layer. To adapt to low oxygen concentrations, bacteria possess multiple terminal oxidases. In this study, we identified six terminal oxidases of B. cenocepacia H111 and constructed reporter strains to monitor their expression in different environments. While the heme-copper oxidase aa3 (cta) was constitutively expressed, the bd-1 oxidase (cyd) was induced under oxygen-limited growth conditions. The cyanide-insensitive bd-type terminal oxidase (cio-1) was mainly expressed in cells grown on the surface of solid medium or in liquid cultures in presence of cyanide, which is known to be produced in the CF lung by the often co-residing CF pathogen Pseudomonas aeruginosa. Indeed, a cio-1 insertional mutant was not able to grow in the presence of cyanide confirming the important role of Cio-1 in cyanide resistance. The caa3 oxidase (caa), was only expressed under nutrient limitation when cells were grown on the surface of solid medium. We also investigated the involvement of two regulatory systems, Anr and RoxS/RoxR, in the expression of cio-1 and cyd. Our data suggest, that, given that Cio-1 is only present in prokaryotes and plays an important role in the defense against cyanide-producing P. aeruginosa, it may be a valuable drug target for treatment of polymicrobial infections in CF patients.

PMID:39827173 | DOI:10.1038/s41598-025-86211-8

Respir Res. 2025 Jan 18;26(1):19. doi: 10.1186/s12931-024-03089-2.

ABSTRACT

This retrospective population-based study investigated the impact of elexacaftor/tezacaftor/ivacaftor (ETI) therapy on inhaled medication adherence in people with cystic fibrosis (pwCF). Prescription refill rate (PRR) for several inhaled medications were compared before and after ETI introduction in three major Italian CF centers. We found a significant decrease in PRR for most inhaled antibiotics and dornase-alpha after ETI implementation.This suggests that patients may be reducing their adherence to inhaled medications, potentially due to improved respiratory symptoms and quality of life. The study highlights the challenges of maintaining adherence to chronic inhaled medications in pwCF, even after the introduction of breakthrough therapies like ETI. Monitoring adherence remains crucial for optimizing patient outcomes, and the PRR emerges as a valuable tool for tracking adherence in real-world settings.

PMID:39827103 | DOI:10.1186/s12931-024-03089-2

J Thorac Cardiovasc Surg. 2025 Feb;169(2):484-504. doi: 10.1016/j.jtcvs.2024.08.052.

ABSTRACT

BACKGROUND: Donor lung procurement and preservation is critical for lung transplantation success. Unfortunately, the large variability in techniques impacts organ utilization rates and transplantation outcomes. Compounding this variation, recent developments in cold static preservation and new technological advances with machine perfusion have increased the complexity of the procedure. The objective of the American Association for Thoracic Surgery (AATS) Clinical Practice Standards Committee (CPSC) expert panel was to make evidence-based recommendations for best practices in donor lung procurement and preservation based on review of the existing literature.

METHODS: The AATS CPSC assembled an expert panel of 16 lung transplantation surgeons from 14 centers who developed a consensus document of recommendations. The panel was divided into 7 subgroups covering (1) intraoperative donor assessment, (2) surgical techniques, (3) ex situ static lung preservation methods, (4) hypothermic preservation, (5) normothermic ex vivo lung perfusion (EVLP), (6) donation after circulatory death (DCD) and normothermic regional perfusion, and (7) donor management centers, organ assessment centers, and third-party procurement teams. Following a focused literature review, each subgroup formulated recommendation statements for each subtopic, which were reviewed and further refined using a Delphi process until a 75% consensus was achieved on each final statement by the voting group.

RESULTS: The expert panel achieved consensus on 34 recommendations for current best practices in donor lung procurement and preservation both in brain-dead as well as DCD donation. The use of new methods of cold preservation, the role of EVLP, and DCD with and without concomitant heart donation are described in detail.

CONCLUSIONS: Consistent and best practices in donor lung procurement and preservation are critical to improve both lung transplantation numbers as well as recipient outcomes. The recommendations described here provide guidance for professionals involved in the care of patients with end-stage lung disease considered for transplantation.

PMID:39826938 | DOI:10.1016/j.jtcvs.2024.08.052

Eur J Med Chem. 2025 Feb 15;284:117229. doi: 10.1016/j.ejmech.2024.117229. Epub 2024 Dec 30.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a progressive and fatal interstitial lung disease for which few drugs are available in clinical practice. Here, we identified novel capsaicin analogs by combining in-house chemical library screening and further structural optimization. (E)-1-(3,4-dihydroxyphenyl)-7-phenylhept-1-en-3-one (Compound 14) was found to be the most potent in inhibiting TGF-β-induced collagen accumulation, proliferation and migration in fibroblast cells. Furthermore, compound 14 (IC50 = 0.51 ± 0.06 μM) showed over 100-fold increasing antifibrotic activity compared to capsaicin (IC50 = 53.71 ± 4.78 μM). Notably, compound 14 could target TRPV1, thereby affecting the expression of the fibrosis markers Collagen Ⅰ and α-SMA by inhibiting the TGF-β/Smads and MAPK pathways to exert antifibrotic activity in vitro. Compound 14 significantly inhibited collagen deposition in lung tissues, ameliorated alveolar structures, and increased survival rates in mice with bleomycin-induced pulmonary fibrosis. In addition, compound 14 possessed lower cytotoxicity (compared to nitedanib) and no toxicity in mice. Overall, compound 14 promise as a potential drug candidate for the treatment of IPF.

PMID:39826937 | DOI:10.1016/j.ejmech.2024.117229

Nat Commun. 2025 Jan 18;16(1):825. doi: 10.1038/s41467-025-56077-5.

ABSTRACT

As next-generation sequencing technologies produce deeper genome coverages at lower costs, there is a critical need for reliable computational host DNA removal in metagenomic data. We find that insufficient host filtration using prior human genome references can introduce false sex biases and inadvertently permit flow-through of host-specific DNA during bioinformatic analyses, which could be exploited for individual identification. To address these issues, we introduce and benchmark three host filtration methods of varying throughput, with concomitant applications across low biomass samples such as skin and high microbial biomass datasets including fecal samples. We find that these methods are important for obtaining accurate results in low biomass samples (e.g., tissue, skin). Overall, we demonstrate that rigorous host filtration is a key component of privacy-minded analyses of patient microbiomes and provide computationally efficient pipelines for accomplishing this task on large-scale datasets.

PMID:39827261 | DOI:10.1038/s41467-025-56077-5

Sci Rep. 2025 Jan 18;15(1):2368. doi: 10.1038/s41598-024-83942-y.

ABSTRACT

Antarctic environments are dominated by microorganisms, which are vulnerable to viral infection. Although several studies have investigated the phylogenetic repertoire of bacteria and viruses in these poly-extreme environments with freezing temperatures, high ultra violet irradiation levels, low moisture availability and hyper-oligotrophy, the evolutionary mechanisms governing microbial immunity remain poorly understood. Using genome-resolved metagenomics, we test the hypothesis that Antarctic poly-extreme high-latitude microbiomes harbour diverse adaptive immune systems. Our analysis reveals the prevalence of prophages in bacterial genomes (Bacteroidota and Verrucomicrobiota), suggesting the significance of lysogenic infection strategies in Antarctic soils. Furthermore, we demonstrate the presence of diverse CRISPR-Cas arrays, including Class 1 arrays (Types I-B, I-C, and I-E), alongside systems exhibiting novel gene architecture among their effector cas genes. Notably, a Class 2 system featuring type V variants lacks CRISPR arrays, encodes Cas1 and Cas2 adaptation module genes. Phylogenetic analysis of Cas12 effector proteins hints at divergent evolutionary histories compared to classified type V effectors and indicates that TnpB is likely the ancestor of Cas12 nucleases. Our findings suggest substantial novelty in Antarctic cas sequences, likely driven by strong selective pressures. These results underscore the role of viral infection as a key evolutionary driver shaping polar microbiomes.

PMID:39827180 | DOI:10.1038/s41598-024-83942-y

Cell Rep Methods. 2025 Jan 15:100964. doi: 10.1016/j.crmeth.2024.100964. Online ahead of print.

ABSTRACT

We develop a data harmonization approach for C. elegans volumetric microscopy data, consisting of a standardized format, pre-processing techniques, and human-in-the-loop machine-learning-based analysis tools. Using this approach, we unify a diverse collection of 118 whole-brain neural activity imaging datasets from five labs, storing these and accompanying tools in an online repository WormID (wormid.org). With this repository, we train three existing automated cell-identification algorithms, CPD, StatAtlas, and CRF_ID, to enable accuracy that generalizes across labs, recovering all human-labeled neurons in some cases. We mine this repository to identify factors that influence the developmental positioning of neurons. This growing resource of data, code, apps, and tutorials enables users to (1) study neuroanatomical organization and neural activity across diverse experimental paradigms, (2) develop and benchmark algorithms for automated neuron detection, segmentation, cell identification, tracking, and activity extraction, and (3) share data with the community and comply with data-sharing policies.

PMID:39826553 | DOI:10.1016/j.crmeth.2024.100964

Cell Syst. 2025 Jan 16:S2405-4712(24)00402-2. doi: 10.1016/j.cels.2024.12.008. Online ahead of print.

ABSTRACT

The human gut microbiome contains many bacterial strains of the same species ("strain-level variants") that shape microbiome function. The tremendous scale and molecular resolution at which microbial communities are being interrogated motivates addressing how to describe strain-level variants. We introduce the "Spectral Tree"-an inferred tree of relatedness built from patterns of co-evolutionary constraint between greater than 7,000 diverse bacteria. Using the Spectral Tree to describe over 600 diverse gut commensal strains that we isolated, whole-genome sequenced, and metabolically profiled revealed (1) widespread phylogenetic structure among strain-level variants, (2) the origins of subspecies phylogeny as a shared history of phage infections across humans, and (3) the key role of inter-human strain variation in predicting strain-level metabolic qualities. Overall, our work demonstrates the existence and metabolic importance of structured phylogeny below the level of species for commensal gut bacteria, motivating a redefinition of individual strains according to their evolutionary context. A record of this paper's transparent peer review process is included in the supplemental information.

PMID:39826551 | DOI:10.1016/j.cels.2024.12.008

STAR Protoc. 2025 Jan 16;6(1):103570. doi: 10.1016/j.xpro.2024.103570. Online ahead of print.

ABSTRACT

Analyzing host-microbe interactions is essential for understanding how microbiota changes disrupt host homeostasis. Here, we present a protocol for predicting host-microbe protein-protein interactions and their downstream effects using MicrobioLink. We describe steps for setting up the environment, installing software, and preparing human transcriptomic and bacterial proteomic data. The protocol outlines procedures for predicting protein-protein interactions through domain-motif interactions, integrating multi-omic datasets to map downstream effects, performing network analyses to identify key regulatory pathways, and visualizing multi-layered networks for systems-level data interpretation. For complete details on the use and execution of this protocol, please refer to Gul et al.1 and Poletti et al.2.

PMID:39826120 | DOI:10.1016/j.xpro.2024.103570

J Med Virol. 2025 Jan;97(1):e70180. doi: 10.1002/jmv.70180.

ABSTRACT

An unprecedented global outbreak caused by the monkeypox virus (MPXV) prompted the World Health Organization to declare a public health emergency of international concern on July 23, 2022. Therapeutics and vaccines for MPXV are not widely available, necessitating further studies, particularly in drug repurposing area. To this end, the standardization of in vitro infection systems is essential. The most robust in vitro studies on poxviruses concern the Vaccinia virus, and there are significant gaps in understanding the replicative cycle of MPXV. Herein, we conducted ultrastructural studies using transmission and scanning electron microscopies and 3D reconstruction to describe and elucidate the step-by-step morphogenesis of MPXV. Vero cells, derived from the kidney lineage of Cercopithecus aethiops monkeys, were infected with a strain isolated from an oropharyngeal swab of a patient with suspected Mpox, collected during an observational cohort study conducted between June 12 and August 19, 2022, in Rio de Janeiro, Brazil. Infected Vero cells exhibited several morphological alterations, including cell lysis plaque formation, nuclei with altered chromatin profiles, thickening of the rough endoplasmic reticulum (RER), presence of myelin figures, disorganization of mitochondrial cristae, and the formation of a granular and fibrous matrix (viral factory) surrounded by mitochondria and RER cisternae in a perinuclear space. Viral entry into cells occurred via endocytosis MPXV particles were observed adhering to cytoskeletal filaments, and viral progeny extrusion occurred through exocytosis. This article presents novel data on the morphogenesis of MPXV that have not been previously documented in the literature.

PMID:39825732 | DOI:10.1002/jmv.70180

Mov Disord Clin Pract. 2025 Jan 18. doi: 10.1002/mdc3.14326. Online ahead of print.

ABSTRACT

BACKGROUND: The cerebral Renin-Angiotensin System might have a role in anxiety and depression development.

OBJECTIVE: We explored the effects of Angiotensin II Type 1 receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACE-Is) on anxiety and depression in Parkinson's Disease (PD).

METHODS: Four hundred and twenty-three newly diagnosed drug-naïve PD patients were evaluated using the State-Trait Anxiety Inventory (STAI) and Geriatric Depression Scale (GDS-15) tests and were monitored at baseline and for up to 3 years.

RESULTS: Twelve patients were treated with ARBs and 42 with ACE-Is. ARB-treated patients had lower anxiety STAI scores than those on ACE-Is or drug-free at baseline (17.2 ± 1.3 vs. 21.3 ± 1.3, or 23.8 ± 0.5, respectively, P = 0.021) and during the follow-up (P < 0.01). Depression scores were unaffected by any of the drugs throughout the study.

CONCLUSION: This small sample of ARB-treated PD patients displayed lower levels of anxiety. Randomized clinical trials are warranted.

PMID:39825674 | DOI:10.1002/mdc3.14326

Headache. 2025 Jan 17. doi: 10.1111/head.14903. Online ahead of print.

ABSTRACT

Antibodies targeting either the calcitonin gene-related peptide (CGRP), such as galcanezumab, fremanezumab, and eptinezumab, or the receptor (erenumab) have been approved for the prevention of episodic and chronic migraine. Although widely used and generally effective, a proportion of patients discontinue treatment due to lack of efficacy. In both randomized controlled trials and observational studies, all anti-CGRP monoclonal antibodies (mAbs) have consistently demonstrated comparable efficacy and tolerability, suggesting a pharmacological class effect. However, differences in therapeutic targets, structure, and pharmacokinetic characteristics may influence their efficacy and safety differently. Therefore, in patients not achieving a clinically meaningful response with one anti-CGRP antibody, switching to a different antibody may be a viable option. This review examines the pharmacological characteristics and distinctions among anti-CGRP mAbs, highlighting their mechanisms of action and pharmacokinetic profiles, along with the clinical observational data of switching. Finally, we summarize suggestions from international guidelines.

PMID:39825578 | DOI:10.1111/head.14903

ACS Nano. 2025 Jan 18. doi: 10.1021/acsnano.4c14927. Online ahead of print.

ABSTRACT

The abnormally viscous and thick mucus is a hallmark of cystic fibrosis (CF). How the mutated CF gene causes abnormal mucus remains an unanswered question of paramount interest. Mucus is produced by the hydration of gel-forming mucin macromolecules that are stored in intracellular granules prior to release. Current understanding of mucin/mucus structure before and after secretion remains limited, and contradictory models exist. Here, we used a molecular viscometer and fluorescence lifetime imaging of human bronchoepithelial cells (Normal and CF) to measure nanometer-scale viscosity. We found significantly elevated intraluminal nanoviscosity in a population of CF mucin granules, indicating an intrinsic, presecretory mucin defect. Nanoviscosity influences protein conformational dynamics and function. Its elevation along the protein secretory pathway could arise from molecular overcrowding, impacting mucin's post-translational processing, hydration, and mucus rheology after release. The nanoviscosity of secreted CF mucus was elevated compared to that of non-CF. Interestingly, it was higher after release than in granules. Validation experiments indicate that reduced mobility of water hydrating mucin macromolecules may contribute to the high nanoviscosity in mucus and mucin granules. This suggests that mucins have a weakly ordered state in granules but adopt a highly ordered, nematic crystalline structure when secreted. This challenges the traditional view of mucus as a porous agarose-like gel and suggests an alternative model for mucin organization before and after secretion. Our study also indicates that endoplasmic reticulum stress due to molecular overcrowding could contribute to mucus pathogenesis in CF cells. It encourages the development of therapeutics that target presecretory mechanisms in CF and other muco-obstructive lung diseases.

PMID:39825840 | DOI:10.1021/acsnano.4c14927

Hum Mol Genet. 2025 Jan 17:ddaf007. doi: 10.1093/hmg/ddaf007. Online ahead of print.

ABSTRACT

BACKGROUND: Individuals with cystic fibrosis (CF; a recessive disorder) have an increased risk of colorectal cancer (CRC). Evidence suggests individuals with a single CFTR variant may also have increased CRC risk.

METHODS: Using population-based studies (GECCO, CORECT, CCFR, and ARIC; 53 785 CRC cases and 58 010 controls), we tested for an association between the most common CFTR variant (Phe508del) and CRC risk. For replication, we used whole exome sequencing data from UK Biobank (UKB; 5126 cases and 20 504 controls matched 4:1 based on genetic distance, age, and sex), and extended our analyses to all other heterozygous CFTR variants annotated as CF-causing.

RESULTS: In our meta-analysis of GECCO-CORECT-CCFR-ARIC, the odds ratio (OR) for CRC risk associated with Phe508del was 1.11 (P = 0.010). In our UKB replication, the OR for CRC risk associated with Phe508del was 1.28 (P = 0.002). The sequencing data from UKB also revealed an association between the presence of any other single CF-causing variant (excluding Phe508del) and CRC risk (OR = 1.33; P = 0.030). When stratifying CFTR variants by functional class, class I variants (no protein produced) had a stronger association (OR = 1.77; p = 0.002), while class II variants (misfolding and retention of the protein in the endoplasmic reticulum) other than Phe508del (OR = 1.75; p = 0.107) had similar effect size as Phe508del, and variants in classes III-VI had non-significant ORs less than 1.0 and/or were not present in cases.

CONCLUSIONS: CF-causing heterozygous variants, especially class I variants, are associated with a modest but statistically significant increased CRC risk. More research is needed to explain the biology underlying these associations.

PMID:39825500 | DOI:10.1093/hmg/ddaf007

Brain Struct Funct. 2025 Jan 18;230(2):32. doi: 10.1007/s00429-024-02889-y.

ABSTRACT

The brain undergoes atrophy and cognitive decline with advancing age. The utilization of brain age prediction represents a pioneering methodology in the examination of brain aging. This study aims to develop a deep learning model with high predictive accuracy and interpretability for brain age prediction tasks. The gray matter (GM) density maps obtained from T1 MRI data of 16,377 healthy participants aged 45 to 82 years from the UKB database were included in this study (mean age, 64.27 ± 7.52 , 7811 men). We propose an innovative deep learning architecture for predicting brain age based on GM density maps. The architecture combines a 3D dual-stream fully convolutional residual network (ds-FCRN) with a Transformer-based global-local feature learning paradigm to enhance prediction accuracy. Moreover, we employed Shapley values to elucidate the influence of various brain regions on prediction precision. On a test set of 3,276 healthy subjects (mean age, 64.15 ± 7.45 , 1561 men), our 3D ds-FCRN model achieved a mean absolute error of 2.2 years in brain age prediction, outperforming existing models on the same dataset. The posterior interpretation revealed that the temporal lobe plays the most significant role in the brain age prediction process, while frontal lobe aging is associated with the greatest number of lifestyle factors. Our designed 3D ds-FCRN model achieved high predictive accuracy and high decision transparency. The brain age vectors constructed using Shapley values provided brain region-level insights into life factors associated with abnormal brain aging.

PMID:39826018 | DOI:10.1007/s00429-024-02889-y

Pediatr Cardiol. 2025 Jan 18. doi: 10.1007/s00246-024-03762-9. Online ahead of print.

ABSTRACT

Kawasaki disease (KD) is a febrile vasculitis disorder, with coronary artery lesions (CALs) being the most severe complication. Early detection of CALs is challenging due to limitations in echocardiographic equipment (UCG). This study aimed to develop and validate an artificial intelligence algorithm to distinguish CALs in KD patients and support diagnostic decision-making at admission. A deep learning algorithm named KCPREDICT was developed using 24 features, including basic patient information, five classic KD clinical signs, and 14 laboratory measurements. Data were collected from patients diagnosed with KD between February 2017 and May 2023 at Shanghai Children's Medical Center. Patients were split into training and internal validation cohorts at an 80:20 ratio, and fivefold cross-validation was employed to assess model performance. Among the 1474 KD cases, the decision tree model performed best during the full feature experiment, achieving an accuracy of 95.42%, a precision of 98.83%, a recall of 93.58%, an F1 score of 96.14%, and an area under the receiver operating characteristic curve (AUROC) of 96.00%. The KCPREDICT algorithm can aid frontline clinicians in distinguishing KD patients with and without CALs, facilitating timely treatment and prevention of severe complications. The use of the complete set of 24 diagnostic features is the optimal choice for predicting CALs in children with KD.

PMID:39825907 | DOI:10.1007/s00246-024-03762-9