Pub Date : 2025-11-29DOI: 10.1016/j.slast.2025.100372
Yirui Niu , Quan Dai , Min Li
Skeletal Class II malocclusion is a common dentofacial deformity often associated with dysregulation in the growth and remodeling of periodontal tissues. Understanding the cellular heterogeneity and molecular pathways of periodontal ligament (PDL) cells is crucial for identifying novel therapeutic targets. However, traditional bulk sequencing methods lack the resolution to distinguish cell-type-specific gene expression and epigenetic regulation, limiting insights into the pathogenesis of this condition. To address these limitations, we propose an integrated framework utilizing 5200 high-quality PDL Single-Cell RNA combined with Assay for Transposase-Accessible Chromatin using sequencing (scRNA-ATAC-seq) to perform a multi-omics dissection of PDL cells in Class II malocclusion. This approach enables the simultaneous profiling of transcriptomic and chromatin accessibility landscapes at single-cell resolution, uncovering cell-specific regulatory networks. Using this scRNA-ATAC-seq method, we successfully identified 12 distinct PDL cell subpopulations and revealed key transcription factors and signaling pathways involved in aberrant skeletal development. Data integration achieved a 0.94 accuracy score, enabling confident regulatory mapping. The findings provide critical insights into the gene regulatory architecture underlying Class II malocclusion and highlight potential cell-specific therapeutic targets for clinical intervention.
{"title":"Single cell and multi omics dissection of periodontal ligament cells identifies regulatory networks and therapeutic targets in skeletal class II malocclusion","authors":"Yirui Niu , Quan Dai , Min Li","doi":"10.1016/j.slast.2025.100372","DOIUrl":"10.1016/j.slast.2025.100372","url":null,"abstract":"<div><div>Skeletal Class II malocclusion is a common dentofacial deformity often associated with dysregulation in the growth and remodeling of periodontal tissues. Understanding the cellular heterogeneity and molecular pathways of periodontal ligament (PDL) cells is crucial for identifying novel therapeutic targets. However, traditional bulk sequencing methods lack the resolution to distinguish cell-type-specific gene expression and epigenetic regulation, limiting insights into the pathogenesis of this condition. To address these limitations, we propose an integrated framework utilizing 5200 high-quality PDL Single-Cell RNA combined with Assay for Transposase-Accessible Chromatin using sequencing (scRNA-ATAC-seq) to perform a multi-omics dissection of PDL cells in Class II malocclusion. This approach enables the simultaneous profiling of transcriptomic and chromatin accessibility landscapes at single-cell resolution, uncovering cell-specific regulatory networks. Using this scRNA-ATAC-seq method, we successfully identified 12 distinct PDL cell subpopulations and revealed key transcription factors and signaling pathways involved in aberrant skeletal development. Data integration achieved a 0.94 accuracy score, enabling confident regulatory mapping. The findings provide critical insights into the gene regulatory architecture underlying Class II malocclusion and highlight potential cell-specific therapeutic targets for clinical intervention.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"36 ","pages":"Article 100372"},"PeriodicalIF":3.7,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145650051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-14DOI: 10.1016/j.slast.2025.100366
{"title":"Life sciences and anxiety - Between enlightenment and uncertainty.","authors":"","doi":"10.1016/j.slast.2025.100366","DOIUrl":"10.1016/j.slast.2025.100366","url":null,"abstract":"","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":" ","pages":"100366"},"PeriodicalIF":3.7,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145534900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-13DOI: 10.1016/j.slast.2025.100365
Boyang Ma , Haiyan Hu , Yu Lin , Zhiheng Wang , Qingyu Song
Although mechanism-to-intervention processes are becoming possible because to the convergence of single-cell technologies with RNA treatment methods, combined host-microbiome analysis with systematic target identification for colorectal adenoma is still fragmented. Here, we created a repeatable computational pipeline that combines MaAsLin2 for host-microbiome association modeling, QIIME2/DADA2 for microbiome processing, and Seurat/Harmony for single-cell analysis. Under strict statistical control (FDR < 0.05), three single-cell RNA sequencing datasets (GSE117875, GSE178341, and GSE144735; totaling 426,425 cells) were combined with parallel microbiome datasets (PRJNA397906, PRJNA541510, and PRJNA672605; 975 samples). In adenoma-associated microbiomes, we measured a 26.8 % decrease in Shannon diversity (4.21→3.08), with a notable enrichment of Fusobacterium nucleatum and a depletion of Faecalibacterium prausnitzii. Immune activation patterns, goblet cell malfunction (MUC2 2.4-fold drop), and stem cell expansion (LGR5 3.2-fold increase) were all identified by single-cell analysis. 847 significant host-microbiome interactions were found by integration analysis, and F. nucleatum showed a substantial correlation with markers of inflammatory signaling (NFKB1: β=0.64, FDR<0.001) and stem cell proliferation (LGR5: β=0.72, FDR<0.001). 25 RNA-targetable candidates were identified by systematic prioritizing, including mRNA restoration targets (MUC2, FOXP3) and ASO/siRNA suppression targets (NFKB1, IL1B). By converting host-microbiome interaction networks into systematic RNA therapeutic options, this technology framework creates a template for the translation of transcriptomics into therapeutics and offers a repeatable pipeline for the creation of precision interventions in colorectal disease.
{"title":"Technology-enabled integration of single-cell transcriptomics and microbiome data identifies RNA-targetable host-microbiota networks in colorectal adenoma","authors":"Boyang Ma , Haiyan Hu , Yu Lin , Zhiheng Wang , Qingyu Song","doi":"10.1016/j.slast.2025.100365","DOIUrl":"10.1016/j.slast.2025.100365","url":null,"abstract":"<div><div>Although mechanism-to-intervention processes are becoming possible because to the convergence of single-cell technologies with RNA treatment methods, combined host-microbiome analysis with systematic target identification for colorectal adenoma is still fragmented. Here, we created a repeatable computational pipeline that combines MaAsLin2 for host-microbiome association modeling, QIIME2/DADA2 for microbiome processing, and Seurat/Harmony for single-cell analysis. Under strict statistical control (FDR < 0.05), three single-cell RNA sequencing datasets (GSE117875, GSE178341, and GSE144735; totaling 426,425 cells) were combined with parallel microbiome datasets (PRJNA397906, PRJNA541510, and PRJNA672605; 975 samples). In adenoma-associated microbiomes, we measured a 26.8 % decrease in Shannon diversity (4.21→3.08), with a notable enrichment of Fusobacterium nucleatum and a depletion of Faecalibacterium prausnitzii. Immune activation patterns, goblet cell malfunction (MUC2 2.4-fold drop), and stem cell expansion (LGR5 3.2-fold increase) were all identified by single-cell analysis. 847 significant host-microbiome interactions were found by integration analysis, and F. nucleatum showed a substantial correlation with markers of inflammatory signaling (NFKB1: β=0.64, FDR<0.001) and stem cell proliferation (LGR5: β=0.72, FDR<0.001). 25 RNA-targetable candidates were identified by systematic prioritizing, including mRNA restoration targets (MUC2, FOXP3) and ASO/siRNA suppression targets (NFKB1, IL1B). By converting host-microbiome interaction networks into systematic RNA therapeutic options, this technology framework creates a template for the translation of transcriptomics into therapeutics and offers a repeatable pipeline for the creation of precision interventions in colorectal disease.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"36 ","pages":"Article 100365"},"PeriodicalIF":3.7,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145530929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-12DOI: 10.1016/j.slast.2025.100364
Jamien Lim, Tal Murthy
{"title":"Literature highlights column: From the literature: Life Sciences Discovery and Technology Highlights.","authors":"Jamien Lim, Tal Murthy","doi":"10.1016/j.slast.2025.100364","DOIUrl":"10.1016/j.slast.2025.100364","url":null,"abstract":"","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":" ","pages":"100364"},"PeriodicalIF":3.7,"publicationDate":"2025-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145515029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-06DOI: 10.1016/j.slast.2025.100362
Mohammad Aatif , Mohamed S. AboHoussien , Ahmed T. Elhendawy , Ghazala Muteeb , Eduardo L. Fabella , Doaa S.R. Khafaga
Prostate cancer remains one of the leading causes of cancer-related illness and death in men globally. Despite advancements in diagnostics and traditional therapies, significant challenges such as drug resistance, systemic toxicity, and restricted specificity persist in impeding successful management. In recent years, nanotechnology has emerged as a revolutionary method in cancer treatment, offering targeted drug delivery, enhanced bioavailability, and reduced off-target side effects. Eco-friendly or green-synthesized nanomaterials have gained significant attention due to their biocompatibility, sustainability, and reduced environmental impact. This review delineates the present state of prostate cancer treatment and the constraints of traditional pharmacological approaches. We subsequently investigate the burgeoning role of nanomedicine in addressing these difficulties, focusing specifically on eco-friendly nanomaterials synthesized from plant extracts, microbial systems, and natural polymers. These biosynthesized nanoparticles (NPs) provide dual benefits: medicinal effectiveness and diminished environmental impact, consistent with the tenets of green chemistry and sustainable medicine. Additionally, we examine diverse drug delivery systems employing green NPs for prostate cancer, including liposomes, polymeric NPs, and metal-based systems synthesized through environmentally friendly methods. Recent in vitro and in vivo research is rigorously examined to assess the clinical potential of these methodologies. The review identifies significant translational hurdles, such as large-scale repeatability, regulatory constraints, and stability concerns, while proposing potential future approaches to enhance the therapeutic application of eco-friendly nanomedicine in prostate cancer treatment.
{"title":"Nanomedicine for prostate cancer: Modern therapies based on green synthesis of nanoparticles","authors":"Mohammad Aatif , Mohamed S. AboHoussien , Ahmed T. Elhendawy , Ghazala Muteeb , Eduardo L. Fabella , Doaa S.R. Khafaga","doi":"10.1016/j.slast.2025.100362","DOIUrl":"10.1016/j.slast.2025.100362","url":null,"abstract":"<div><div>Prostate cancer remains one of the leading causes of cancer-related illness and death in men globally. Despite advancements in diagnostics and traditional therapies, significant challenges such as drug resistance, systemic toxicity, and restricted specificity persist in impeding successful management. In recent years, nanotechnology has emerged as a revolutionary method in cancer treatment, offering targeted drug delivery, enhanced bioavailability, and reduced off-target side effects. Eco-friendly or green-synthesized nanomaterials have gained significant attention due to their biocompatibility, sustainability, and reduced environmental impact. This review delineates the present state of prostate cancer treatment and the constraints of traditional pharmacological approaches. We subsequently investigate the burgeoning role of nanomedicine in addressing these difficulties, focusing specifically on eco-friendly nanomaterials synthesized from plant extracts, microbial systems, and natural polymers. These biosynthesized nanoparticles (NPs) provide dual benefits: medicinal effectiveness and diminished environmental impact, consistent with the tenets of green chemistry and sustainable medicine. Additionally, we examine diverse drug delivery systems employing green NPs for prostate cancer, including liposomes, polymeric NPs, and metal-based systems synthesized through environmentally friendly methods. Recent in vitro and in vivo research is rigorously examined to assess the clinical potential of these methodologies. The review identifies significant translational hurdles, such as large-scale repeatability, regulatory constraints, and stability concerns, while proposing potential future approaches to enhance the therapeutic application of eco-friendly nanomedicine in prostate cancer treatment.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"35 ","pages":"Article 100362"},"PeriodicalIF":3.7,"publicationDate":"2025-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145477298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-03DOI: 10.1016/j.slast.2025.100363
Jizhao Yao , Mingfeng Ye , Taosheng Miao , Lusheng Miao
Background
Hidradenitis suppurativa (HS) represents a prevalent inflammatory dermatosis that not only triggers persistent local inflammation but also imposes significant psychosocial burdens. Given that the pathogenesis of HS remains incompletely elucidated, there is an urgent need necessitating the development of precision intervention strategies grounded in patients' immune profiles.
Methods
To investigate the pathogenesis of hidradenitis suppurativa, we employed an integrative approach combining bulk transcriptomics, Mendelian randomization (MR), and single-cell transcriptomic profiling. Specifically, transcriptomic analysis utilized Weighted Gene Co-Expression Network Analysis(WGCNA) for co-expression network construction and Least Absolute Shrinkage and Selection Operator(LASSO) regression for feature selection. Mendelian randomization applied IVW, weighted median, simple mode, weighted mode, and Bayesian MR approaches to ensure robust causal inference, with mediation analysis identifying potential metabolites. Molecular docking simulations were conducted to validate drug candidates targeting core genes. For single-cell transcriptomic analysis, we leveraged Gene Expression Omnibus(GEO) datasets followed by dispersion reduction, clustering, and cell-cell communication analysis to elucidate underlying cellular mechanisms.
Result
Our transcriptomic analysis identified Granzyme A(GZMA) as a central therapeutic target (Discovery cohort ROC: 0.945; Validation cohort ROC: 0.819). Furthermore, Mendelian randomization analysis indicated a heightened causal association between GZMA and HS risk: IVW (OR = 1.62, 95 %CI:1.24–2.11; p = 0.0003) and BWMR (OR = 1.62, 95 %CI:1.26–2.08; p = 0.0002). Critically,mediation analysis established N-Acetylputrescine as a potential mediating metabolite. Shifting to the cellular level, single-cell sequencing revealed prominent GZMA expression specifically within NK cells. Analysis of cell-cell interactions revealed communication between NK cells and both T cells and B cells, and highlighted differences in the communication dynamics between GZMA-positive and GZMA-negative subpopulations.
Conclusion
Integrative analysis of transcriptomic, MR, and scRNA-seq data strongly implicates GZMA as a potential therapeutic target and highlights the crucial role of NK cells in HS pathogenesis. These findings provide novel insights into HS immunopathology and pave the way for targeted therapeutic development.
{"title":"Integrated multi-omics identifies GZMA targeting NK cells as a novel therapeutic strategy for hidradenitis suppurativa","authors":"Jizhao Yao , Mingfeng Ye , Taosheng Miao , Lusheng Miao","doi":"10.1016/j.slast.2025.100363","DOIUrl":"10.1016/j.slast.2025.100363","url":null,"abstract":"<div><h3>Background</h3><div>Hidradenitis suppurativa (HS) represents a prevalent inflammatory dermatosis that not only triggers persistent local inflammation but also imposes significant psychosocial burdens. Given that the pathogenesis of HS remains incompletely elucidated, there is an urgent need necessitating the development of precision intervention strategies grounded in patients' immune profiles.</div></div><div><h3>Methods</h3><div>To investigate the pathogenesis of hidradenitis suppurativa, we employed an integrative approach combining bulk transcriptomics, Mendelian randomization (MR), and single-cell transcriptomic profiling. Specifically, transcriptomic analysis utilized Weighted Gene Co-Expression Network Analysis(WGCNA) for co-expression network construction and Least Absolute Shrinkage and Selection Operator(LASSO) regression for feature selection. Mendelian randomization applied IVW, weighted median, simple mode, weighted mode, and Bayesian MR approaches to ensure robust causal inference, with mediation analysis identifying potential metabolites. Molecular docking simulations were conducted to validate drug candidates targeting core genes. For single-cell transcriptomic analysis, we leveraged Gene Expression Omnibus(GEO) datasets followed by dispersion reduction, clustering, and cell-cell communication analysis to elucidate underlying cellular mechanisms.</div></div><div><h3>Result</h3><div>Our transcriptomic analysis identified Granzyme A(GZMA) as a central therapeutic target (Discovery cohort ROC: 0.945; Validation cohort ROC: 0.819). Furthermore, Mendelian randomization analysis indicated a heightened causal association between GZMA and HS risk: IVW (OR = 1.62, 95 %CI:1.24–2.11; <em>p</em> = 0.0003) and BWMR (OR = 1.62, 95 %CI:1.26–2.08; <em>p</em> = 0.0002). Critically,mediation analysis established N-Acetylputrescine as a potential mediating metabolite. Shifting to the cellular level, single-cell sequencing revealed prominent GZMA expression specifically within NK cells. Analysis of cell-cell interactions revealed communication between NK cells and both T cells and B cells, and highlighted differences in the communication dynamics between GZMA-positive and GZMA-negative subpopulations.</div></div><div><h3>Conclusion</h3><div>Integrative analysis of transcriptomic, MR, and scRNA-seq data strongly implicates GZMA as a potential therapeutic target and highlights the crucial role of NK cells in HS pathogenesis. These findings provide novel insights into HS immunopathology and pave the way for targeted therapeutic development.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"35 ","pages":"Article 100363"},"PeriodicalIF":3.7,"publicationDate":"2025-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145454025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-30DOI: 10.1016/j.slast.2025.100359
Mickey Hawelitschek , Felix Lenk , Florian-David Lange , Peter Schmidt , Constantin Rahm , Thomas Walther
In an era of increasingly stringent regulatory requirements and growing demands on laboratories in terms of data quality, integrity and reproducibility, combined with increasing sample throughput, digitization of the laboratory environment is an important tool to meet these growing demands. Unfortunately, many challenges stand in the way. Not only are these barriers difficult to overcome, but they also cut across the different professions that are essential to any digitization effort. For most laboratories, however, the simple question remains: How to transform an existing laboratory? This article proposes a structured method to realize the SmartLab and to verify its feasibility in practice, using a real R&D process. Consequently, on a qualitative level, not only has data integrity and sample traceability been enhanced, but standardization has also been increased. Quantitatively, the walk-away time was tripled and the efficiency of individual process steps was increased by over 20%. The project’s success can be attributed to the implementation of a systematic methodology for digital transformation, which functioned as a comprehensive guide, facilitating the execution of the project in a step-by-step manner.
{"title":"Digital transformation of a real-life R&D process using a structured approach based on the Internet of Things","authors":"Mickey Hawelitschek , Felix Lenk , Florian-David Lange , Peter Schmidt , Constantin Rahm , Thomas Walther","doi":"10.1016/j.slast.2025.100359","DOIUrl":"10.1016/j.slast.2025.100359","url":null,"abstract":"<div><div>In an era of increasingly stringent regulatory requirements and growing demands on laboratories in terms of data quality, integrity and reproducibility, combined with increasing sample throughput, digitization of the laboratory environment is an important tool to meet these growing demands. Unfortunately, many challenges stand in the way. Not only are these barriers difficult to overcome, but they also cut across the different professions that are essential to any digitization effort. For most laboratories, however, the simple question remains: How to transform an existing laboratory? This article proposes a structured method to realize the SmartLab and to verify its feasibility in practice, using a real R&D process. Consequently, on a qualitative level, not only has data integrity and sample traceability been enhanced, but standardization has also been increased. Quantitatively, the walk-away time was tripled and the efficiency of individual process steps was increased by over 20%. The project’s success can be attributed to the implementation of a systematic methodology for digital transformation, which functioned as a comprehensive guide, facilitating the execution of the project in a step-by-step manner.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"35 ","pages":"Article 100359"},"PeriodicalIF":3.7,"publicationDate":"2025-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145427173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-29DOI: 10.1016/j.slast.2025.100361
Ping Wei , Yuanyuan Yang , Yanhong Yan , Chunhong Wei , Pinjing Hui
Objective
To establish and validate the radiomics model of carotid vulnerable plaque.
Methods
182 patients who underwent carotid endarterectomy in the First Affiliated Hospital of Soochow University from January 2019 to June 2022 were retrospectively analyzed. Plaque ultrasound images were acquired and segmented, and features were extracted using the R package “EBImage”. Six machine learning algorithms were used to construct a plaque vulnerability classifier model. The relationship between clinical biomarkers and outcomes was evaluated using linear regression and the False Discovery Rate (FDR) correction method.
Results
Diabetes, neutrophils, monocytes, high-sensitivity C-reactive protein, high-density lipoprotein, and stenosis rate were found to have a strong association with plaque vulnerability. The Naive Bayes algorithm performed well in the training set using image features alone, with an AUC of 0.840, and an AUC of 0.762 in the test set. The Decision Tree algorithm had certain performance in the training set using image features alone, with an AUC of 0.609, and an AUC of 0.626 in the test set. The Naive Bayes algorithm achieved excellent performance in the training set using both plaque ultrasound imaging features and clinical laboratory indicators, with an AUC of 0.922, and an AUC of 0.928 in the test set.
Conclusion
The above radiomics model can be used to predict the vulnerable carotid plaque before surgery. By combining plaque ultrasound imaging with clinical blood biomarkers, a more comprehensive and accurate assessment of plaque vulnerability can be achieved, thereby reducing the risk of cardiovascular events.
{"title":"Development and validation of a multimodal machine learning model integrating ultrasound imaging and serum biomarkers for vulnerable carotid plaque prediction","authors":"Ping Wei , Yuanyuan Yang , Yanhong Yan , Chunhong Wei , Pinjing Hui","doi":"10.1016/j.slast.2025.100361","DOIUrl":"10.1016/j.slast.2025.100361","url":null,"abstract":"<div><h3>Objective</h3><div>To establish and validate the radiomics model of carotid vulnerable plaque.</div></div><div><h3>Methods</h3><div>182 patients who underwent carotid endarterectomy in the First Affiliated Hospital of Soochow University from January 2019 to June 2022 were retrospectively analyzed. Plaque ultrasound images were acquired and segmented, and features were extracted using the R package “EBImage”. Six machine learning algorithms were used to construct a plaque vulnerability classifier model. The relationship between clinical biomarkers and outcomes was evaluated using linear regression and the False Discovery Rate (FDR) correction method.</div></div><div><h3>Results</h3><div>Diabetes, neutrophils, monocytes, high-sensitivity C-reactive protein, high-density lipoprotein, and stenosis rate were found to have a strong association with plaque vulnerability. The Naive Bayes algorithm performed well in the training set using image features alone, with an AUC of 0.840, and an AUC of 0.762 in the test set. The Decision Tree algorithm had certain performance in the training set using image features alone, with an AUC of 0.609, and an AUC of 0.626 in the test set. The Naive Bayes algorithm achieved excellent performance in the training set using both plaque ultrasound imaging features and clinical laboratory indicators, with an AUC of 0.922, and an AUC of 0.928 in the test set.</div></div><div><h3>Conclusion</h3><div>The above radiomics model can be used to predict the vulnerable carotid plaque before surgery. By combining plaque ultrasound imaging with clinical blood biomarkers, a more comprehensive and accurate assessment of plaque vulnerability can be achieved, thereby reducing the risk of cardiovascular events.</div></div>","PeriodicalId":54248,"journal":{"name":"SLAS Technology","volume":"35 ","pages":"Article 100361"},"PeriodicalIF":3.7,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145423446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}