RSC Adv. 2025 Apr 17;15(16):12331-12341. doi: 10.1039/d5ra01341k. eCollection 2025 Apr 16.

ABSTRACT

Atypical teratoid/rhabdoid tumor (AT/RT) is a rare and aggressive tumor of the primary central nervous system primarily affecting children. It typically originates in the cerebellum and brain stem and is associated with a low survival rate. While standard chemotherapy has been used as a primary treatment for AT/RTs, its success rate is unsatisfactory, and patients often experience severe side effects. Therefore, there is an urgent need to develop new and effective treatment strategies. One promising approach for identifying new therapies is drug repurposing. Although many FDA-approved drugs have been repurposed for various cancers, there have been no reports of such applications for AT/RTs. In this study, a library of 2130 FDA-approved drugs was screened using a high-throughput screening system against 2D traditional cultures and 3D spheroid cultures of AT/RT cell lines (BT-12 and BT-16). From this screening, colchicine, a non-chemotherapeutic agent, was identified as a promising candidate. It exhibited IC50 values of 0.016 and 0.056 μM against 2D BT-12 and 2D BT-16 cells, respectively, and IC50 values of 0.004 and 0.023 μM against 3D BT-12 and BT-16 spheroid cultures. Additionally, the cytotoxic effects of colchicine on human brain endothelial cells and human astrocytes were evaluated, and CC50 > 20 μM was observed, which is over two orders of magnitude higher than its effective concentrations in AT/RT cells, indicating considerably lower toxicity to normal brain cells and brain endothelial cells. In conclusion, colchicine shows significant potential to be repurposed as a treatment for AT/RTs, providing a safer and more effective therapeutic option for this rare and challenging disease.

PMID:40248220 | PMC:PMC12004362 | DOI:10.1039/d5ra01341k

KDD. 2025 Aug;2025(v1):2347-2358. doi: 10.1145/3690624.3709418. Epub 2025 Jul 20.

ABSTRACT

Drug repurposing identifies new therapeutic uses for existing drugs, reducing the time and costs compared to traditional de novo drug discovery. Most existing drug repurposing studies using real-world patient data often treat the entire population as homogeneous, ignoring the heterogeneity of treatment responses across patient subgroups. This approach may overlook promising drugs that benefit specific subgroups but lack notable treatment effects across the entire population, potentially limiting the number of repurposable candidates identified. To address this, we introduce STEDR, a novel drug repurposing framework that integrates subgroup analysis with treatment effect estimation. Our approach first identifies repurposing candidates by emulating multiple clinical trials on real-world patient data and then characterizes patient subgroups by learning subgroup-specific treatment effects. We deploy STEDR to Alzheimer's Disease (AD), a condition with few approved drugs and known heterogeneity in treatment responses. We emulate trials for over one thousand medications on a large-scale real-world database covering over 8 million patients, identifying 14 drug candidates with beneficial effects to AD in characterized subgroups. Experiments demonstrate STEDR's superior capability in identifying repurposing candidates compared to existing approaches. Additionally, our method can characterize clinically relevant patient subgroups associated with important AD-related risk factors, paving the way for precision drug repurposing.

PMID:40248108 | PMC:PMC12001032 | DOI:10.1145/3690624.3709418

Biomed Rep. 2025 Apr 9;22(6):95. doi: 10.3892/br.2025.1973. eCollection 2025 Jun.

ABSTRACT

Biofilm formation is a key factor in the persistence and recurrence of otorhinolaryngology (ORL) infections, driving antimicrobial resistance and treatment failure. Chronic conditions, such as rhinosinusitis, otitis media and tonsillitis, are linked to biofilm-producing pathogens, forming protective extracellular matrices that shield bacteria from immune defenses and antibiotics. The present review explores emerging pharmacological strategies to disrupt biofilm integrity and improve treatment outcomes. Strategies such as quorum sensing inhibitors, antibiofilm peptides, enzymatic dispersal agents, and drug repurposing can potentially disrupt biofilms and counter-resistance mechanisms. Furthermore, novel therapies (including nanotechnology-based drug delivery systems, phage therapy and immunomodulation) offer innovative alternatives for managing biofilm-associated infections. However, clinical implementation remains challenging. Future research should prioritize optimizing drug formulations, refining delivery techniques, and exploring synergistic combinations to enhance biofilm eradication. Implementing these innovative strategies can improve the management of chronic ORL infections, reducing recurrence rates and enhancing patient outcomes.

PMID:40247931 | PMC:PMC12001231 | DOI:10.3892/br.2025.1973

Am J Med Genet C Semin Med Genet. 2025 Apr 18:e32142. doi: 10.1002/ajmg.c.32142. Online ahead of print.

ABSTRACT

Usher syndrome, the most common form of deaf-blindness, displays extensive genetic, allelic, and phenotypic heterogeneity. The dual sensory impairment associated with this autosomal recessive disorder makes Usher syndrome an important target for gene therapy, with dozens of published preclinical studies targeting multiple Usher syndrome genes and using multiple gene therapy strategies. Nine genes have been conclusively linked to Usher syndrome; however, data on the prevalence and contribution of specific genetic variants is lacking. Such information is essential to choosing a favorable target gene or therapeutic approach during clinical trial design. Here, we used large genomic databases to systematically evaluate the genomics of Usher syndrome. We ascertained pathogenic Usher syndrome variants from three clinical databases and determined the occurrence of these pathogenic Usher syndrome variants within: (1) a publicly available dataset including worldwide populations (GnomAD), (2) a cohort of 3888 children without hearing loss, and (3) 637 children with hearing loss. Results show significant variability in the frequency of Usher syndrome variants by gene and genetic ancestry. 1% of control subjects carry a pathogenic USH variant. Pathogenic variants in USH2A are the most prevalent, at 1 in 150 individuals (0.0062). Calculated general population prevalence for all Usher syndrome subtypes is 1 in ~29,000, indicating that 30,405 individuals in the United States and 721,769 individuals worldwide are affected. We estimate that 324 babies in the United States and 12,090 worldwide are born with Usher syndrome each year. We identify key targets for genetic therapy based on population-level prevalence including a focus on alternatives to gene replacement therapies, specifically for USH2A.

PMID:40248902 | DOI:10.1002/ajmg.c.32142

Curr Pharm Biotechnol. 2025 Apr 16. doi: 10.2174/0113892010353450250408114725. Online ahead of print.

ABSTRACT

BACKGROUND: Overdose involving opioids is the black heart of the addiction crisis. "Pre-addiction," as an encouraging concept by NIDA and NIAAA, seems best captured with the construct of dopamine dysregulation. Referring to the abundant publications on "Reward Deficiency Syndrome" (RDS), Genetic Addiction Risk Score (GARS) test, RDSQ29, and KB220, Pre-addiction can be referred to as "reward dysregulation" as a suitable suggestion. The hypothesis is that the true phenotype is RDS, and other behavioral disorders are endophenotypes where the genetic variants play important roles, specifically in the Brain Reward Cascade (BRC).

METHODS: This study tested the pharmacogenomics of the GARS panel by a multi-model in silico investigation in four layers: 1) Protein-Protein Interactions (PPIs); 2) Gene Regulatory Networks (GRNs); 3) Disease, drugs and chemicals (DDCs); and 4) Gene Coexpression Networks (GCNs).

RESULTS: All in silico findings were combined in an Enrichment Analysis for 59 refined genes, which represented highly significant associations of dopamine pathways in the BRC and supported our hypothesis.

CONCLUSION: This paper provides scientific evidence for the importance of incorporating GARS as a predictive test to identify Pre-addiction, introduce unique therapeutic targets assisting in the treatment of pain, drug dosing of prescription pharmaceuticals, and identify the risk for subsequent addiction early in -life.

PMID:40247807 | DOI:10.2174/0113892010353450250408114725

J Chemother. 2025 Apr 17:1-7. doi: 10.1080/1120009X.2025.2489837. Online ahead of print.

ABSTRACT

Fluoropyrimidines (FPs) are antineoplastic agents used for the treatment of various solid tumors, especially gastrointestinal cancers. Patients with variations in dihydropyrimidine dehydrogenase gene (DPYD), which can determine the partial or complete deficiency of the dihydropyrimidine dehydrogenase enzyme (DPD), are at an increased risk of developing severe and potentially life-threatening toxicity. Worldwide the introduction of pharmacogenetic testing into clinical practice has been a slow process and in our center the analysis of the DPYD gene has been adopted since April 2020. We evaluated the clinical application of routine DPYD screening and its ability to prevent early-onset of fluoropyrimidine-related toxicity in patients treated at the Oncology Reference Center of Basilicata (IRCCS-CROB), a recognized cancer centre in Southern Italy. From April 2020 to November 2022, 300 patients (male 137; female 163) diagnosed with various types of cancer were subjected to DPYD genotyping, before starting treatment with FPs. In accordance with the current European Medicines Agency (EMA) and the Italian Association of Medical Oncology (AIOM) guidelines patients were tested for four DPYD variants that are associated with reduced DPD activity. FPs dose adjustments in DPYD variant carriers were made following the previously mentioned guidelines. Three hundred patients underwent DPYD testing and thirteen (4.3%) patients were found to be heterozygous variant carriers; ten out of thirteen patients received FP dose reduction as indicated by the guidelines, one out of thirteen patients received alternative treatment, two of the thirteen patients received no treatment at all. The main toxicities observed in patients who received a DPYD genotype-based dose reduction were anemia, neutropenia, nausea and mucositis but events were primarily grade 1 or 2. Our experience confirms the technical feasibility and the usefulness of DPYD genotyping to reduce the risk of severe FPs toxicities.

PMID:40247645 | DOI:10.1080/1120009X.2025.2489837

Clin Pharmacol Ther. 2025 Apr 17. doi: 10.1002/cpt.3681. Online ahead of print.

ABSTRACT

The ESR1 gene is relevant in breast cancer treatments in the pharmacogenetics context. However, Native, African, and mixed populations are known to be underrepresented in genomic studies. This is particularly important given that the difference in variants' frequencies among different populations can lead to population-specific clinical implications. Therefore, this study aims to infer the genomic subcontinental ancestry and allele frequencies of the ESR1 gene variants in 2,427 individuals from 26 populations worldwide from the 1000 Genomes Project and 125 Natives from Peru, whose genomes have not yet been analyzed in the literature regarding this gene. Linear regression with Bonferroni correction analyses was conducted based on ancestry inference and frequencies. Our findings demonstrate subcontinental differentiation of African, Asian, European, and Native populations. Overall, 102 associations (P < 0.01) were found for 68 clinically relevant variants. Particularly, subcontinental associations were observed for variants associated with the Native, Asian, European, and African components. We highlight the findings for the rs9349799 and rs2234693 variants, previously associated with altered responses to breast cancer treatments. rs9349799 was positively associated with the South-Asian component, while rs2234693 was negatively associated with the Coast/Amazonian Native and positively associated with the East-African component. Nearly half of the variants are intronic, highlighting the importance of studying whole genomes rather than just exomes. These results emphasize subcontinental differences' relevance for designing pharmacogenetic panels. Including neglected populations in genomic and pharmacogenomic studies is essential for democratic access to scientific advances and for more egalitarian and effective pharmacogenetic implementation, tailored to each population's specificities.

PMID:40247433 | DOI:10.1002/cpt.3681

Mhealth. 2025 Mar 21;11:20. doi: 10.21037/mhealth-24-66. eCollection 2025.

ABSTRACT

BACKGROUND AND OBJECTIVE: With the increased adoption of digital health solutions, such as telehealth, there is a need to consider current practices and considerations towards digital health Literacy. The objective of this review is to explore what digital health literacy considerations have been detailed in cystic fibrosis telehealth papers.

METHODS: The found papers published from a recent systematic review exploring telehealth within cystic fibrosis care were taken and analysed. These papers were obtained from PubMed, Web of Science, and Scopus databases and included any paper written in English up to May 2021. Data pertaining to Health Literacy, Digital Literacy/Competency, Digital Health Literacy, Training, Readiness Assessments, and Sustained Use were extrapolated using Elicit AI Research Assistant 2024.

KEY CONTENT AND FINDINGS: From the 26 papers, the data of interest was sparse and mostly unavailable for this review. This may be due to several reasons; however the implication of this mitigation is discussed with reference to the digital divide, health in-equalities, and safety.

CONCLUSIONS: This review highlights that a structured approach to assess digital health literacy of care teams and people with cystic fibrosis is critical to the future success of safe telehealth use, and other digital health solutions.

PMID:40248749 | PMC:PMC12004305 | DOI:10.21037/mhealth-24-66

J Clin Transl Endocrinol. 2025 Apr 2;40:100391. doi: 10.1016/j.jcte.2025.100391. eCollection 2025 Jun.

ABSTRACT

BACKGROUND/AIMS: People with CF (PwCF) have low total, high, and low density lipoprotein cholesterol (TC, HDL-C and LDL-C) and historically have had low prevalence of cardiovascular disease. More recently, cases of acute myocardial infarction are reported in PwCF. The impact of elexacaftor-tezacaftor-ivacaftor (ETI) on cholesterol and triglyceride (TG) concentrations, traditional cardiometabolic risk factors, is unknown.

METHODS/RESULTS: TC, LDL-C, HDL-C, and TG concentrations were analyzed from participants enrolled in the observational PROMISE study of clinically prescribed ETI prior to and 12-18 months after initiation. Pre-ETI and follow-up concentrations were compared, and relationships between TC, LDL-C, HDL-C and TG and clinical factors were tested using linear mixed-effect models.Fasting samples were available for 51 participants (25 M/26F, median age 17.4 y) with pancreatic exocrine insufficiency at baseline and 12-18 months after ETI initiation. TC and HDL-C were higher after 12-18 mo ETI in an unadjusted model, but with adjustment for BMI-Z, only HDL-C remained significantly higher at follow up (p < 0.05). Low HDL-C was the most common abnormality (>50 %), but prevalence of participants meeting criteria for low HDL-C did not differ between timepoints.

CONCLUSIONS: In a population of youth and young adults with CF, TC and HDL-C were higher after 12-18 months of ETI, but differences in TC were attenuated with adjustment for BMI-Z. Prevalence of low HDL-C was high at both timepoints.

PMID:40248170 | PMC:PMC12005328 | DOI:10.1016/j.jcte.2025.100391

Learn Health Syst. 2024 Sep 14;9(2):e10453. doi: 10.1002/lrh2.10453. eCollection 2025 Apr.

ABSTRACT

BACKGROUND: Medication adherence data are an important quality indicator in cystic fibrosis (CF) care, yet real-time objective data are not routinely available. An online application (CFHealthHub) has been designed to deliver these data to people with CF and their clinical team. Adoption of this innovation is the focus of an National Health Service England-funded learning health system and Quality Improvement Collaborative (QIC). This study applies the capability, opportunity, and motivation model of behavior change to assess whether the QIC had supported healthcare professionals' uptake of accessing patient adherence data.

METHOD: This was a mixed-method study, treating each multidisciplinary team as an individual case. Click analytic data from CFHealthHub were collected between January 1, 2018, and September 22, 2019. Thirteen healthcare practitioners participated in semi-structured interviews, before and after establishing the QIC. Qualitative data were analyzed using the behavior change model.

RESULTS: The cases showed varied improvement trajectories. While two cases reported reduced barriers, one faced persistent challenges. Participation in the QIC led to enhanced confidence in the platform's utility. Reduced capability, opportunity, and motivation barriers correlated with increased uptake, demonstrating value in integrating behavior change theory into QICs.

CONCLUSION: QICs can successfully reduce barriers and enable uptake of e-health innovations such as adherence monitoring technology. However, ongoing multi-level strategies are needed to embed changes. Further research should explore sustainability mechanisms and their impact on patient outcomes.

PMID:40247895 | PMC:PMC12000760 | DOI:10.1002/lrh2.10453

Curr Cardiol Rev. 2025 Apr 17. doi: 10.2174/011573403X351048250329170744. Online ahead of print.

ABSTRACT

Cardiovascular disease (CVD) Continues to be the leading cause of mortality worldwide, underscoring the critical need for effective prevention and management strategies. The ability to predict cardiovascular risk accurately and cost-effectively is central to improving patient outcomes and reducing the global burden of CVD. While useful, traditional tools used for risk assessment are often limited in their scope and fail to adequately account for atypical presentations and complex patient profiles. These limitations highlight the necessity for more advanced approaches, particularly integrating artificial intelligence (AI) into cardiovascular risk prediction. Our review explores the transformative role of AI in enhancing the accuracy, efficiency, and accessibility of cardiovascular risk prediction models. The implementation of AI-driven risk assessment tools has shown promising results, not only in improving CVD mortality rates but also in enhancing quality of life (QOL) markers and reducing healthcare costs. Machine learning (ML) algorithms predicted 2-year survival rates after MI with improved accuracy compared to traditional models. Deep Learning (DL) forecasted hypertension risk with a 91.7% accuracy based on electronic health records. Furthermore, AI-driven ECG (Electrocardiography) analysis has demonstrated high precision in identifying left ventricular systolic dysfunction, even with noisy single-lead data from wearable devices. These tools enable more personalized treatment strategies, foster greater patient engagement, and support informed decision-making by healthcare providers. Unfortunately, the widespread adoption of AI in CVD risk assessment remains a challenge, largely due to a lack of education and acceptance among healthcare professionals. To overcome these barriers, it is crucial to promote broader education on the benefits and applications of AI in cardiovascular risk prediction. By fostering a greater understanding and acceptance of these technologies, we can accelerate their integration into clinical practice, ultimately aiming to mitigate the global impact of CVD.

PMID:40248921 | DOI:10.2174/011573403X351048250329170744

J Anaesthesiol Clin Pharmacol. 2025 Apr-Jun;41(2):351-356. doi: 10.4103/joacp.joacp_151_24. Epub 2025 Mar 24.

ABSTRACT

BACKGROUND AND AIMS: The field of anaesthesiology and perioperative medicine has explored advancements in science and technology, ensuring precision and personalized anesthesia plans. The surge in the usage of chat-generative pretrained transformer (Chat GPT) in medicine has evoked interest among anesthesiologists to assess its performance in the operating room. However, there is concern about accuracy, patient privacy and ethics. Our objective in this study assess whether Chat GPT can provide assistance in clinical decisions and compare them with those of resident anesthesiologists.

MATERIAL AND METHODS: In this cross-sectional study conducted at a teaching hospital, a set of 30 hypothetical clinical scenarios in the operating room were presented to resident anesthesiologists and Chat-GPT 4. The first five scenarios out of 30 were typed with three additional prompts in the same chat to determine if there was any detailing of answers. The responses were labeled and assessed by three reviewers not involved in the study.

RESULTS: The interclass coefficient (ICC) values show variation in the level of agreement between Chat GPT and anesthesiologists. For instance, the ICC of 0.41 between A1 and Chat GPT indicates a moderate level of agreement, whereas the ICC of 0.06 between A2 and Chat GPT suggests a comparatively weaker level of agreement.

CONCLUSIONS: In this study, it was found that there were variations in the level of agreement between Chat GPT and resident anesthesiologists' response in terms of accuracy and comprehensiveness of responses in solving intraoperative scenarios. The use of prompts improved the agreement of Chat GPT with anesthesiologists.

PMID:40248774 | PMC:PMC12002681 | DOI:10.4103/joacp.joacp_151_24

J Med Imaging (Bellingham). 2025 Mar;12(2):024005. doi: 10.1117/1.JMI.12.2.024005. Epub 2025 Apr 16.

ABSTRACT

PURPOSE: Breast cancer, the most common cancer type among women worldwide, requires early detection and accurate diagnosis for improved treatment outcomes. Segmenting fat and fibroglandular tissue (FGT) in magnetic resonance imaging (MRI) is essential for creating volumetric models, enhancing surgical workflow, and improving clinical outcomes. Manual segmentation is time-consuming and subjective, prompting the development of automated deep-learning algorithms to perform this task. However, configuring these algorithms for 3D medical images is challenging due to variations in image features and preprocessing distortions. Automated machine learning (AutoML) frameworks automate model selection, hyperparameter tuning, and architecture optimization, offering a promising solution by reducing reliance on manual intervention and expert knowledge.

APPROACH: We compare nnU-Net and Auto3Dseg, two AutoML frameworks, in segmenting fat and FGT on T1-weighted MRI images from the Duke breast MRI dataset (100 patients). We used threefold cross-validation, employing the Dice similarity coefficient (DSC) and Hausdorff distance (HD) metrics for evaluation. The F -test and Tukey honestly significant difference analysis were used to assess statistical differences across methods.

RESULTS: nnU-Net achieved DSC scores of 0.946 ± 0.026 (fat) and 0.872 ± 0.070 (FGT), whereas Auto3DSeg achieved 0.940 ± 0.026 (fat) and 0.871 ± 0.074 (FGT). Significant differences in fat HD ( F = 6.3020 , p < 0.001 ) originated from the full resolution and the 3D cascade U-Net. No evidence of significant differences was found in FGT HD or DSC metrics.

CONCLUSIONS: Ensemble approaches of Auto3Dseg and nnU-Net demonstrated comparable performance in segmenting fat and FGT on breast MRI. The significant differences in fat HD underscore the importance of boundary-focused metrics in evaluating segmentation methods.

PMID:40248763 | PMC:PMC12003052 | DOI:10.1117/1.JMI.12.2.024005

IJCAI (U S). 2024 Aug;2024:8272-8280. doi: 10.24963/ijcai.2024/914.

ABSTRACT

The development of electronic health records (EHR) systems has enabled the collection of a vast amount of digitized patient data. However, utilizing EHR data for predictive modeling presents several challenges due to its unique characteristics. With the advancements in machine learning techniques, deep learning has demonstrated its superiority in various applications, including healthcare. This survey systematically reviews recent advances in deep learning-based predictive models using EHR data. Specifically, we introduce the background of EHR data and provide a mathematical definition of the predictive modeling task. We then categorize and summarize predictive deep models from multiple perspectives. Furthermore, we present benchmarks and toolkits relevant to predictive modeling in healthcare. Finally, we conclude this survey by discussing open challenges and suggesting promising directions for future research.

PMID:40248670 | PMC:PMC12005588 | DOI:10.24963/ijcai.2024/914

IISE Trans Healthc Syst Eng. 2025;15(1):1-14. doi: 10.1080/24725579.2024.2398592. Epub 2024 Sep 7.

ABSTRACT

Advanced sensing and imaging provide unprecedented opportunities to collect data from diverse sources for increasing information visibility in spatiotemporal dynamical systems. Furthermore, the fundamental physics of the dynamical system is commonly elucidated through a set of partial differential equations (PDEs), which plays a critical role in delineating the manner in which the sensing signals can be modeled. Reliable predictive modeling of such spatiotemporal dynamical systems calls upon the effective fusion of fundamental physics knowledge and multi-source sensing data. This paper proposes a multi-source data and knowledge fusion framework via deep learning for dynamical systems with applications to spatiotemporal cardiac modeling. This framework not only achieves effective data fusion through capturing the physics-based information flow between different domains, but also incorporates the geometric information of a 3D system through a graph Laplacian for robust spatiotemporal predictive modeling. We implement the proposed framework to model cardiac electrodynamics under both healthy and diseased heart conditions. Numerical experiments demonstrate the superior performance of our framework compared with traditional approaches that lack the capability for effective data fusion or geometric information incorporation.

PMID:40248641 | PMC:PMC12002414 | DOI:10.1080/24725579.2024.2398592

Transpl Int. 2025 Apr 3;38:14377. doi: 10.3389/ti.2025.14377. eCollection 2025.

ABSTRACT

This study developed a predictive model for Post-Transplant Diabetes Mellitus (PTDM) by integrating clinical and radiological data to identify at-risk kidney transplant recipients. In a retrospective analysis across three Mayo Clinic sites, clinical metrics were combined with deep learning analysis of pre-transplant CT images, focusing on body composition parameters like adipose tissue and muscle mass instead of BMI or other biomarkers. Among 2,005 nondiabetic kidney recipients, 335 (16.7%) developed PTDM within the first year. PTDM patients were older, had higher BMIs, elevated triglycerides, and were more likely to be male and non-White. They exhibited lower skeletal muscle area, greater visceral adipose tissue (VAT), more intermuscular fat, and higher subcutaneous fat (all p < 0.001). Multivariable analysis identified age (OR: 1.05, 95% CI: 1.03-1.08, p < 0.0001), family diabetes history (OR: 1.55, CI: 1.14-2.09, p = 0.0061), White race (OR: 0.43, CI: 0.28-0.66, p < 0.0001), and VAT area (OR: 1.37, CI: 1.14-1.64, p = 0.0009) as predictors. The combined model achieved C-statistic of 0.724 (CI: 0.692-0.757), outperforming the clinical-only model (C-statistic 0.68). Patients with PTDM in the first year had higher mortality than those without PTDM. This model improves predictive precision, enabling accurate identification and intervention for at risk patients.

PMID:40248509 | PMC:PMC12003133 | DOI:10.3389/ti.2025.14377

World J Gastroenterol. 2025 Apr 14;31(14):104280. doi: 10.3748/wjg.v31.i14.104280.

ABSTRACT

BACKGROUND: Esophageal squamous cell carcinoma is a major histological subtype of esophageal cancer. Many molecular genetic changes are associated with its occurrence. Raman spectroscopy has become a new method for the early diagnosis of tumors because it can reflect the structures of substances and their changes at the molecular level.

AIM: To detect alterations in Raman spectral information across different stages of esophageal neoplasia.

METHODS: Different grades of esophageal lesions were collected, and a total of 360 groups of Raman spectrum data were collected. A 1D-transformer network model was proposed to handle the task of classifying the spectral data of esophageal squamous cell carcinoma. In addition, a deep learning model was applied to visualize the Raman spectral data and interpret their molecular characteristics.

RESULTS: A comparison among Raman spectral data with different pathological grades and a visual analysis revealed that the Raman peaks with significant differences were concentrated mainly at 1095 cm-1 (DNA, symmetric PO, and stretching vibration), 1132 cm-1 (cytochrome c), 1171 cm-1 (acetoacetate), 1216 cm-1 (amide III), and 1315 cm-1 (glycerol). A comparison among the training results of different models revealed that the 1D-transformer network performed best. A 93.30% accuracy value, a 96.65% specificity value, a 93.30% sensitivity value, and a 93.17% F1 score were achieved.

CONCLUSION: Raman spectroscopy revealed significantly different waveforms for the different stages of esophageal neoplasia. The combination of Raman spectroscopy and deep learning methods could significantly improve the accuracy of classification.

PMID:40248385 | PMC:PMC12001190 | DOI:10.3748/wjg.v31.i14.104280

Front Cardiovasc Med. 2025 Apr 3;12:1525966. doi: 10.3389/fcvm.2025.1525966. eCollection 2025.

ABSTRACT

Neurocardiology is an evolving field focusing on the interplay between the nervous system and cardiovascular system that can be used to describe and understand many pathologies. Acute ischemic stroke can be understood through this framework of an interconnected, reciprocal relationship such that ischemic stroke occurs secondary to cardiac pathology (the Heart-Brain axis), and cardiac injury secondary to various neurological disease processes (the Brain-Heart axis). The timely assessment, diagnosis, and subsequent management of cerebrovascular and cardiac diseases is an essential part of bettering patient outcomes and the progression of medicine. Artificial intelligence (AI) and machine learning (ML) are robust areas of research that can aid diagnostic accuracy and clinical decision making to better understand and manage the disease of neurocardiology. In this review, we identify some of the widely utilized and upcoming AI/ML algorithms for some of the most common cardiac sources of stroke, strokes of undetermined etiology, and cardiac disease secondary to stroke. We found numerous highly accurate and efficient AI/ML products that, when integrated, provided improved efficacy for disease prediction, identification, prognosis, and management within the sphere of stroke and neurocardiology. In the focus of cryptogenic strokes, there is promising research elucidating likely underlying cardiac causes and thus, improved treatment options and secondary stroke prevention. While many algorithms still require a larger knowledge base or manual algorithmic training, AI/ML in neurocardiology has the potential to provide more comprehensive healthcare treatment, increase access to equitable healthcare, and improve patient outcomes. Our review shows an evident interest and exciting new frontier for neurocardiology with artificial intelligence and machine learning.

PMID:40248254 | PMC:PMC12003416 | DOI:10.3389/fcvm.2025.1525966

Proc SPIE Int Soc Opt Eng. 2023 Feb;12465:124650O. doi: 10.1117/12.2655263. Epub 2023 Apr 7.

ABSTRACT

Intravenous contrast enhancement phase information is important for computer-aided diagnosis of CT scans because the visual appearance of the scans varies substantially among the different phases. Although phase information could help to refine training data curation for downstream tasks, it is seldom included in the process of data augmentation for training a deep learning model. Unfortunately, in the current clinical settings, phase information is either unavailable or unreliable in most PACS systems. This motivates us to develop a method to automatically classify multiphase CT scans. In this study, a residual network (ResNet34) was utilized to classify five CT phases commonly used in the clinical environment: non-contrast, arterial, portal venous, nephrographic, and delayed contrast phases. A dataset of 395 multiphase CT scans was weakly labeled using keywords. The weakly-labeled dataset was split into 316 training, and 79 test CT scans. We compared the ResNet34 with two other popular classification models, VGG19 and DenseNet121. ResNet34 achieved the highest accuracy of 99%, while the accuracy of VGG19 and DenseNet121 were 97% and 95%, respectively. In addition, ResNet34 had fewer parameters to train in comparison with two other models, which could reduce the inference time to 35 seconds per scan and enhance generalizability of the model. High accuracy of multiphase classification suggests a potential way to improve data curation based on CT contrast enhancement phase. This would be useful to improve deep learning models by enhancing dataset curation and providing more realistic augmented data.

PMID:40248190 | PMC:PMC12004730 | DOI:10.1117/12.2655263

iScience. 2025 Mar 17;28(4):112224. doi: 10.1016/j.isci.2025.112224. eCollection 2025 Apr 18.

ABSTRACT

Noninvasive methods for liver fibrosis staging are urgently needed due to its significance in predicting significant morbidity and mortality. In this study, we developed an automated DL-based segmentation and classification model (Model-C). Test-time adaptation was used to address data distribution shifts. We then established a deep learning-radiologist complementarity decision system (DRCDS) via a decision model determining whether to adopt Model-C's diagnosis or defer to radiologists. Model-C (AUCs of 0.89-0.92) outperformed models based on liver (AUCs: 0.84-0.90) or spleen (AUCs: 0.69-0.70). With test-time adaptation, the Obuchowski index values of Model-C in three external sets improved from 0.81, 0.73, and 0.73 to 0.85, 0.85, and 0.81. DRCDS performed slightly better than Model-C or senior radiologists, with 73.7%-92.0% of cases adopting Model-C's diagnosis. In conclusion, DRCDS could diagnose liver fibrosis with high accuracy. Additionally, we provided solutions to model generalization and human-machine complementarity issues in multi-classification problems.

PMID:40248124 | PMC:PMC12005311 | DOI:10.1016/j.isci.2025.112224