ABSTRACT
OBJECTIVE: This study aimed to distinguish tuberculous spondylodiscitis (TS) from pyogenic spondylodiscitis (PS) based on laboratory, magnetic resonance imaging (MRI) and computed tomography (CT) findings. Further, a novel diagnostic model for differential diagnosis was developed. METHODS: We obtained MRI, CT and laboratory data from TS and PS patients. Predictive models were built using binary logistic regression analysis. The receiver operating characteristic curve was analyzed. Both internal and external validation was performed. RESULTS: A total of 81 patients with PS (n = 46) or TS (n = 35) were enrolled. All patients had etiological evidence from the focal lesion. Disc signal or height preservation, skip lesion or multi segment (involved segments ≥ 3) involvement, paravertebral calcification, massive sequestra formation, subligamentous bone destruction, bone erosion with osteosclerotic margin, higher White Blood Cell Count (WBC) and positive result of tuberculosis infection T cell spot test (T-SPOT.TB) were more prevalent in the TS group. A diagnostic model was developed and included four predictors: WBC<7.265 * (10^9/L), skip lesion or involved segments ≥ 3, massive sequestra formation and subligamentous bone destruction. The model showed good sensitivity, specificity, and total accuracy (91.4%, 95.7%, and 93.8%, respectively); the area under the receiver operating characteristic curve (AUC) was 0.981, similar to the results of internal validation using bootstrap resampling (1000 replicates) and external validation set, indicating good clinical predictive ability. CONCLUSIONS: This study develop a good diagnostic model based on both CT and MRI, as well as laboratory findings, which may help clinicians distinguish between TS and PS.
ABSTRACT
We aimed to uncover the pathological mechanism of ischemic stroke (IS) using a combined analysis of untargeted metabolomics and proteomics. The serum samples from a discovery set of 44 IS patients and 44 matched controls were analyzed using a specific detection method. The same method was then used to validate metabolites and proteins in the two validation sets: one with 30 IS patients and 30 matched controls, and the other with 50 IS patients and 50 matched controls. A total of 105 and 221 differentially expressed metabolites or proteins were identified, and the association between the two omics was determined in the discovery set. Enrichment analysis of the top 25 metabolites and 25 proteins in the two-way orthogonal partial least-squares with discriminant analysis, which was employed to identify highly correlated biomarkers, highlighted 15 pathways relevant to the pathological process. One metabolite and seven proteins exhibited differences between groups in the validation set. The binary logistic regression model, which included metabolite 2-hydroxyhippuric acid and proteins APOM_O95445, MASP2_O00187, and PRTN3_D6CHE9, achieved an area under the curve of 0.985 (95% CI: 0.966-1) in the discovery set. This study elucidated alterations and potential coregulatory influences of metabolites and proteins in the blood of IS patients.
ABSTRACT
Aim: A keloid is a fibroproliferative cutaneous disorder secondary to skin injury, caused by an imbalance in fibroblast proliferation and apoptosis. However, the pathogenesis is not fully understood. In this study, candidate genes for keloid were identified and used to construct a diagnostic model. Methods: Three datasets related to keloids were downloaded from NCBI Gene Expression Omnibus. Fibroblast-related genes were screened, and fibroblast scores for the samples were determined. Then, a weighted gene co-expression network analysis (WGCNA) was used to identify modules and genes associated with keloids and the fibroblast score. Differentially expressed genes (DEGs) between keloid and control samples were identified and compared with fibroblast-related genes and genes in the modules. Overlapping genes were evaluated using functional enrichment analyses. Signature genes were further screened, and a diagnostic model was constructed. Finally, correlations between immune cell frequences and signature genes were analyzed. Results: In total, 124 fibroblast-related genes were obtained, and the fibroblast score was an effective indicator of the sample type. WGCNA revealed five modules that were significantly correlated with both the disease state and fibroblast scores, including 1760 genes. Additionally, 589 DEGs were identified, including 16 that overlapped with fibroblast-related genes and genes identified in the WGCNA. These genes were related to cell proliferation and apoptosis and were involved in FoxO, Rap1, p53, Ras, MAPK, and PI3K-Akt pathways. Finally, a six fibroblast-related gene signature (CCNB1, EGFR, E2F8, BTG1, TP63, and IGF1) was identified and used for diagnostic model construction. The proportions of regulatory T cells and macrophages were significantly higher in keloid tissues than in controls. Conclusion: The established model based on CCNB1, EGFR, E2F8, BTG1, TP63, and IGF1 showed good performance and may be useful for keloid diagnosis.
ABSTRACT
Background: Combining machine learning (ML) with gait analysis is widely applicable for diagnosing abnormal gait patterns. Objective: To analyze gait adaptability characteristics in stroke patients, develop ML models to identify individuals with GAD, and select optimal diagnostic models and key classification features. Methods: This study was investigated with 30 stroke patients (mean age 42.69 years, 60% male) and 50 healthy adults (mean age 41.34 years, 58% male). Gait adaptability was assessed using a CMill treadmill on gait adaptation tasks: target stepping, slalom walking, obstacle avoidance, and speed adaptation. The preliminary analysis of variables in both groups was conducted using t-tests and Pearson correlation. Features were extracted from demographics, gait kinematics, and gait adaptability datasets. ML models based on Support Vector Machine, Decision Tree, Multi-layer Perceptron, K-Nearest Neighbors, and AdaCost algorithm were trained to classify individuals with and without GAD. Model performance was evaluated using accuracy (ACC), sensitivity (SEN), F1-score and the area under the receiver operating characteristic (ROC) curve (AUC). Results: The stroke group showed a significantly decreased gait speed (p = 0.000) and step length (SL) (p = 0.000), while the asymmetry of SL (p = 0.000) and ST (p = 0.000) was higher compared to the healthy group. The gait adaptation tasks significantly decreased in slalom walking (p = 0.000), obstacle avoidance (p = 0.000), and speed adaptation (p = 0.000). Gait speed (p = 0.000) and obstacle avoidance (p = 0.000) were significantly correlated with global F-A score in stroke patients. The AdaCost demonstrated better classification performance with an ACC of 0.85, SEN of 0.80, F1-score of 0.77, and ROC-AUC of 0.75. Obstacle avoidance and gait speed were identified as critical features in this model. Conclusion: Stroke patients walk slower with shorter SL and more asymmetry of SL and ST. Their gait adaptability was decreased, particularly in obstacle avoidance and speed adaptation. The faster gait speed and better obstacle avoidance were correlated with better functional mobility. The AdaCost identifies individuals with GAD and facilitates clinical decision-making. This advances the future development of user-friendly interfaces and computer-aided diagnosis systems.
ABSTRACT
INTRODUCTION: This study aimed to screen immune-related marker genes of ischemic stroke (IS). MATERIAL AND METHODS: Two IS-related gene expression datasets were downloaded. The significantly differentially expressed genes (DEGs) and miRNAs (DEMs) between IS and control groups were selected. The differential immune cells were analysed. Weighted gene co-expression network analysis (WGCNA) was applied to analyse immune-related genes, followed by function analysis and interaction network construction. Then, key genes were further screened using optimization algorithm to construct a diagnostic model. Finally, miRNA regulatory network of several key genes was established. RESULTS: In total 321 DEGs and 140 DEMs were obtained. 11 immune cell types were significantly different between IS and control groups. WGCNA identified two key modules, involving 202 differential immune genes. The greenyellow module was enriched in biological processes and pathways associated with T cells, while the midnightblue module was mainly associated with apoptosis, and inflammatory response-related functions and pathways. Protein interaction network identified 10 hub nodes, such as CD8A, ITGAM and TLR4. LASSO regression selected 8 key feature genes, and a risk score model was established. Key model genes were enriched in 63 GO biological processes, such as microglial cell activation, and B cell apoptotic process, and 3 KEGG pathways, such as negative regulation of nuclear cell cycle DNA replication, and hematopoietic cell lineage. Finally, a total of 25 miRNA-target relationship pairs were obtained. CONCLUSIONS: This study identified some immune-related marker genes and constructed a diagnostic model based on 8 immune-related genes in IS.
Subject(s)
Gene Regulatory Networks , Ischemic Stroke , Humans , Ischemic Stroke/genetics , Ischemic Stroke/immunology , Gene Regulatory Networks/genetics , MicroRNAs/genetics , Gene Expression Profiling/methods , Protein Interaction Maps/geneticsABSTRACT
BACKGROUND: Early detection and treatment of colorectal cancer (CRC) is crucial for improving patient survival rates. This study aims to identify signature molecules associated with CRC, which can serve as valuable indicators for clinical hematological screening. METHOD: We have systematically searched the Human Protein Atlas database and the relevant literature for blood protein-coding genes. The CRC dataset from TCGA was used to compare the acquired genes and identify differentially expressed molecules (DEMs). Weighted Gene Co-expression Network Analysis (WGCNA) was employed to identify modules of co-expressed molecules and key molecules within the DEMs. Signature molecules of CRC were then identified from the key molecules using machine learning. These findings were further validated in clinical samples. Finally, Logistic regression was used to create a predictive model that calculated the likelihood of CRC in both healthy individuals and CRC patients. We evaluated the model's sensitivity and specificity using the ROC curve. RESULT: By utilizing the CRC dataset, WGCNA analysis, and machine learning, we successfully identified seven signature molecules associated with CRC from 1478 blood protein-coding genes. These markers include S100A11, INHBA, QSOX2, MET, TGFBI, VEGFA and CD44. Analyzing the CRC dataset showed its potential to effectively discriminate between CRC and normal individuals. The up-regulated expression of these markers suggests the existence of an immune evasion mechanism in CRC patients and is strongly correlated with poor prognosis. CONCLUSION: The combined detection of the seven signature molecules in CRC can significantly enhance diagnostic efficiency and serve as a novel index for hematological screening of CRC.
ABSTRACT
Background: Schizophrenia is a devastating mental disease with high heritability. A growing number of susceptibility genes associated with schizophrenia, as well as their corresponding SNPs loci, have been revealed by genome-wide association studies. However, using SNPs as predictors of disease and diagnosis remains difficult. Here, we aimed to uncover susceptibility SNPs in a Chinese population and to construct a prediction model for schizophrenia. Methods: A total of 210 participants, including 70 patients with schizophrenia, 70 patients with bipolar disorder, and 70 healthy controls, were enrolled in this study. We estimated 14 SNPs using published risk loci of schizophrenia, and used these SNPs to build a model for predicting schizophrenia via comparison of genotype frequencies and regression. We evaluated the efficacy of the diagnostic model in schizophrenia and control patients using ROC curves and then used the 70 patients with bipolar disorder to evaluate the model's differential diagnostic efficacy. Results: 5 SNPs were selected to construct the model: rs148415900, rs71428218, rs4666990, rs112222723 and rs1716180. Correlation analysis results suggested that, compared with the risk SNP of 0, the risk SNP of 3 was associated with an increased risk of schizophrenia (OR = 13.00, 95% CI: 2.35-71.84, p = 0.003). The ROC-AUC of this prediction model for schizophrenia was 0.719 (95% CI: 0.634-0.804), with the greatest sensitivity and specificity being 60% and 80%, respectively. The ROC-AUC of the model in distinguishing between schizophrenia and bipolar disorder was 0.591 (95% CI: 0.497-0.686), with the greatest sensitivity and specificity being 60% and 55.7%, respectively. Conclusion: The SNP risk score prediction model had good performance in predicting schizophrenia. To the best of our knowledge, previous studies have not applied SNP-based models to differentiate between cases of schizophrenia and other mental illnesses. It could have several potential clinical applications, including shaping disease diagnosis, treatment, and outcomes.
ABSTRACT
PURPOSE: Nonalcoholic steatohepatitis (NASH) has been an increasingly significant contributor to hepatocellular carcinoma (HCC). Understanding the progression from NASH to HCC is critical to early diagnosis and elucidating the underlying mechanisms. RESULTS: 5 significant prognostic genes related to NASH-HCC transformation were identified through algorithm selection, which were ME1, TP53I3, SOCS2, GADD45G and CYP7A1. A diagnostic model for NASH prediction was established (AUC=0.988). TP53I3 and SOCS2 were selected as potential critical genes in the progression of NASH-HCC by external dataset validation and in vitro experiments on NASH and HCC cell lines. Immune infiltration analysis illustrated the correlation between 5 significant prognostic genes and immune cells. Single-cell analysis identified hepatocytes related to NASH-HCC transformation markers, revealing their promoting role in the transformation from NASH to HCC. CONCLUSION: With bulk-seq analysis and single-cell analysis, 5 significant prognostic genes related to NASH-HCC transformation were identified and validated at both dataset and in vitro experiment level. Among them, TP53I3 and SOCS2 might be potential critical genes in NASH-HCC progression. Single-cell analysis identified and revealed the critical role that NASH-HCC related hepatocytes play in NASH-HCC tansformation. Our research may introduce a new perspective to the diagnosis, treatment of NASH-related HCC.
ABSTRACT
Ischemia-reperfusion injury (IRI) is a cumulation of pathophysiological processes that involves cell and organelle damage upon blood flow constraint and subsequent restoration. However, studies on overall immune infiltration and ferroptosis in liver ischemia-reperfusion injury (LIRI) are limited. This study explored immune cell infiltration and ferroptosis in LIRI using bioinformatics and experimental validation. The GSE151648 dataset, including 40 matched pairs of pre- and post- transplant liver samples was downloaded for bioinformatic analysis. Eleven hub genes were identified by overlapping differentially expressed genes (DEGs), iron genes, and genes identified through weighted gene co-expression network analysis (WGCNA). Subsequently, the pathway enrichment, transcription factor-target, microRNA-mRNA and protein-protein interaction networks were investigated. The diagnostic model was established by logistic regression, which was validated in the GSE23649 and GSE100155 datasets and verified using cytological experiments. Moreover, several drugs targeting these genes were found in DrugBank, providing a more effective treatment for LIRI. In addition, the expression of 11 hub genes was validated using quantitative real-time polymerase chain reaction (qRT-PCR) in liver transplantation samples and animal models. The expression of the 11 hub genes increased in LIRI compared with the control. Five genes were significantly enriched in six biological process terms, six genes showed high enrichment for LIRI-related signaling pathways. There were 56 relevant transcriptional factors and two central modules in the protein-protein interaction network. Further immune infiltration analysis indicated that immune cells including neutrophils and natural killer cells were differentially accumulated in the pre- and post-transplant groups, and this was accompanied by changes in immune-related factors. Finally, 10 targeted drugs were screened. Through bioinformatics and further experimental verification, we identified hub genes related to ferroptosis that could be used as potential targets to alleviate LIRI.
ABSTRACT
BACKGROUND: Early childhood caries (ECC) is a challenge for pediatric dentists all over the world, and dietary factor is an important factor affecting the occurrence of ECC. Currently, there is limited research on the impact of dietary nutrient intake from Chinese diets on ECC. The purpose of this study is to explore the correlation of dietary nutrients intake with ECC and caries activity (CA) among children aged 3-5 years, and to provide dietary guidance to slow down the occurrence and development of ECC. METHODS: A cross-sectional study was conducted in 2022. A total of 155 children were divided into three groups: caries-free group, ECC group and Severe early childhood caries (SECC) group according to the caries statues. And according to the caries activity test (CAT) value, they were also divided into three group: low CA group (L-CA), middle CA group (M-CA) and high CA group (H-CA). The 24-hour dietary intake information was collected by mobile phone application (APP). The intake of children's daily dietary nutrients were calculated referring to "China Food Composition Tables". RESULTS: In this study, 17, 39,and 99 children were diagnosed with caries-free, ECC, and SECC. There were 33, 36, and 86 children diagnosed with L-CA, M-CA, and H-CA. The risk of ECC was increased with the intake of cholesterol(OR = 1.005) and magnesium (OR = 1.026) and decreased with the intake of iron (OR = 0.770). The risk of SECC was increased with the intake of cholesterol (OR = 1.003). The risk of high CA was increased with the intake of cholesterol (OR = 1.002). The combined application of dietary total calories, carbohydrate, cholesterol, sodium, magnesium and selenium in the diagnosis of ECC had an area under ROC curve of 0.741. CONCLUSIONS: The increased dietary cholesterol intake may be a common risk factor for ECC and high CA in children aged 3-5. The combined application of dietary intake of total calories, carbohydrate, cholesterol, sodium, magnesium and selenium has a higher predictive value for the occurrence of ECC.
Subject(s)
Dental Caries , Humans , Cross-Sectional Studies , Child, Preschool , Dental Caries/epidemiology , Dental Caries/etiology , Dental Caries/prevention & control , Male , Female , China/epidemiology , Diet , Nutrients/administration & dosage , Energy IntakeABSTRACT
Background: With the widespread use of computed tomography (CT), the detection rate of pulmonary nodules in children has gradually increased. Due to the lack of epidemiological evidence and clinical guideline on pulmonary nodule treatment in children, we aimed to provide a reference for the clinical diagnosis and management of pediatirc pulmonary nodules. Methods: This retrospective study collected consecutive cases from April 2012 to July 2021 in the Shanghai Children's Medical Center. The sample included children with pulmonary nodules on chest CT scans and met the inclusion criteria. All patients were categorized into tumor and non-tumor groups by pre-CT clinical diagnosis. Nodule characteristics between groups were analyzed. To establish a clinical assessment model for the benign versus malignant pulmonary nodules, patients who have been followed-up for three months were detected and a decision tree model for nodule malignancy prediction was constructed and validated. Results: The sample comprised 1341 patients with an average age of 7.2 ± 4.6 years. More than half of them (51.7%) were diagnosed with malignancies before CT scan. 48.3% were diagnosed with non-tumor diseases or healthy. Compared to non-tumor group, children with tumor were more likely to have multiple nodules in both lungs, with larger size and often be accompanied by osteolytic or mass lesions. Based on the decision tree model, patients' history of malignancies, nodules diameter size≥5mm, and specific nodule distribution (multiple in both lungs, multiple in the right lung or solitary in the upper or middle right lobe) were important potential predictors for malignity. In the validation set, sensitivity, specificity and AUC were 0.855, 0.833 and 0.828 (95%CI: 0.712-0.909), respectively. Conclusion: This study conducted a clinical assessment model to differentiate benignity and malignancy of pediatric pulmonary nodules. We suggested that a nodule's diameter, distribution and patient's history of malignancies are predictable factors in benign or malignant determination.
ABSTRACT
Background: Progress in research on expression profiles in osteoarthritis (OA) has been limited to individual tissues within the joint, such as the synovium, cartilage, or meniscus. This study aimed to comprehensively analyze the common gene expression characteristics of various structures in OA and construct a diagnostic model. Methods: Three datasets were selected: synovium, meniscus, and knee joint cartilage. Modular clustering and differential analysis of genes were used for further functional analyses and the construction of protein networks. Signature genes with the highest diagnostic potential were identified and verified using external gene datasets. The expression of these genes was validated in clinical samples by Real-time (RT)-qPCR and immunohistochemistry (IHC) staining. This study investigated the status of immune cells in OA by examining their infiltration. Results: The merged OA dataset included 438 DEGs clustered into seven modules using WGCNA. The intersection of these DEGs with WGCNA modules identified 190 genes. Using Least Absolute Shrinkage and Selection Operator (LASSO) and Random Forest algorithms, nine signature genes were identified (CDADC1, PPFIBP1, ENO2, NOM1, SLC25A14, METTL2A, LINC01089, L3HYPDH, NPHP3), each demonstrating substantial diagnostic potential (areas under the curve from 0.701 to 0.925). Furthermore, dysregulation of various immune cells has also been observed. Conclusion: CDADC1, PPFIBP1, ENO2, NOM1, SLC25A14, METTL2A, LINC01089, L3HYPDH, NPHP3 demonstrated significant diagnostic efficacy in OA and are involved in immune cell infiltration.
ABSTRACT
OBJECTIVE: The diagnosis of symptomatic urinary stones during pregnancy is challenging; ultrasonography has a low specificity and sensitivity for diagnosing urinary stones. This study aimed to develop a clinical diagnostic model to assist clinicians in distinguishing symptomatic urinary stones in pregnant women. METHODS: In this retrospective cohort study, we consecutively collected clinical data from pregnant women who presented with acute abdominal, lumbar, and lumbar and abdominal pain at the emergency department of our hospital between January 1, 2017, and December 31, 2019. To distinguish patients with urinary calculi from those without, we reviewed the follow-up records within 2 weeks post-consultation, ultrasonography results within 2 weeks, or self-reports of stone passage within 2 weeks. We selected risk factors from the baseline clinical and laboratory data of patients to establish a diagnostic model. RESULTS: Of the total patients included in the study, 105 patients were diagnosed as having symptomatic urinary stones and 126 were determined to have abdominal pain for reasons other than urinary stones. The initial model had an area under the curve (AUC) of 0.9966. The No-Lab Model had an AUC of 0.9856. The Lab Model had an AUC of 0.832. The Stone Model had an AUC of 0.9952. The simplified Stone Model did not show a decrease in discriminative ability. CONCLUSION: Of the four diagnostic models that we established for preliminary diagnosis of symptomatic urinary tract stones in pregnant women, the simplified Stone Model demonstrated excellent performance. Users can scan quick response codes to access web-based diagnostic model interfaces, facilitating easy clinical operation.
ABSTRACT
The clinical diagnosis of biliary atresia (BA) poses challenges, particularly in distinguishing it from cholestasis (CS). Moreover, the prognosis for BA is unfavorable and there is a dearth of effective non-invasive diagnostic models for detection. Therefore, the aim of this study is to elucidate the metabolic disparities among children with BA, CS, and normal controls (NC) without any hepatic abnormalities through comprehensive metabolomics analysis. Additionally, our objective is to develop an advanced diagnostic model that enables identification of BA. The plasma samples from 90 children with BA, 48 children with CS, and 47 NC without any liver abnormalities children were subjected to metabolomics analysis, revealing significant differences in metabolite profiles among the 3 groups, particularly between BA and CS. A total of 238 differential metabolites were identified in the positive mode, while 89 differential metabolites were detected in the negative mode. Enrichment analysis revealed 10 distinct metabolic pathways that differed, such as lysine degradation, bile acid biosynthesis. A total of 18 biomarkers were identified through biomarker analysis, and in combination with the exploration of 3 additional biomarkers (LysoPC(18:2(9Z,12Z)), PC (22:5(7Z,10Z,13Z,16Z,19Z)/14:0), and Biliverdin-IX-α), a diagnostic model for BA was constructed using logistic regression analysis. The resulting ROC area under the curve was determined to be 0.968. This study presents an innovative and pioneering approach that utilizes metabolomics analysis to develop a diagnostic model for BA, thereby reducing the need for unnecessary invasive examinations and contributing to advancements in diagnosis and prognosis for patients with BA.
Subject(s)
Biliary Atresia , Biomarkers , Cholestasis , Metabolic Networks and Pathways , Metabolomics , Biliary Atresia/blood , Biliary Atresia/diagnosis , Biliary Atresia/metabolism , Humans , Metabolomics/methods , Cholestasis/blood , Cholestasis/diagnosis , Cholestasis/metabolism , Female , Male , Biomarkers/blood , Infant , Child, Preschool , Diagnosis, Differential , ROC Curve , Metabolome , Case-Control Studies , ChildABSTRACT
Background: Cervical cancer is among the most prevalent malignancies worldwide. This study explores the relationships between angiogenesis-related genes (ARGs) and immune infiltration, and assesses their implications for the prognosis and treatment of cervical cancer. Additionally, it develops a diagnostic model based on angiogenesis-related differentially expressed genes (ARDEGs). Methods: We systematically evaluated 15 ARDEGs using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Gene Set Enrichment Analysis (GSEA), and Gene Set Variation Analysis (GSVA). Immune cell infiltration was assessed using a single-sample gene-set enrichment analysis (ssGSEA) algorithm. We then constructed a diagnostic model for ARDEGs using Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis and evaluated the diagnostic value of this model and the hub genes in predicting clinical outcomes and immunotherapy responses in cervical cancer. Results: A set of ARDEGs was identified from the Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), and UCSC Xena database. We performed KEGG, GO, and GSEA analyses on these genes, revealing significant involvement in cell proliferation, differentiation, and apoptosis. The ARDEGs diagnostic model, constructed using LASSO regression analysis, showed high predictive accuracy in cervical cancer patients. We developed a reliable nomogram and decision curve analysis to evaluate the clinical utility of the ARDEG diagnostic model. The 15 ARDEGs in the model were associated with clinicopathological features, prognosis, and immune cell infiltration. Notably, ITGA5 expression and the abundance of immune cell infiltration (specifically mast cell activation) were highly correlated. Conclusion: This study identifies the prognostic characteristics of ARGs in cervical cancer patients, elucidating aspects of the tumor microenvironment. It enhances the predictive accuracy of immunotherapy outcomes and establishes new strategies for immunotherapeutic interventions.
ABSTRACT
OBJECTIVE: The aim of this study was to reveal the biological functionalities associated with endoplasmic reticulum stress (ERS)-related genes (ERSGs) in the context of diabetic retinopathy (DR). METHODS: Differentially expressed genes (DEGs) within the DR group and the Control group were identified and then integrated with ERSGs. Gene Ontology (GO) and Gene Set Enrichment Analysis (GSEA) methodologies were used to investigate potential biological mechanisms. A diagnostic model for ERS and a nomogram were formulated based on biomarkers selected through the Least Absolute Shrinkage and Selection Operator method. The diagnostic efficacy of this model was thoroughly evaluated. ERS-associated subtypes were identified, and the Single-Sample GSEA (ssGSEA) and CIBERSORT algorithms were used to assess immune infiltration. RESULTS: We identified 10 ERS-related DEGs (ERSRDEGs) within the DR Group. Subsequently, a diagnostic model was constructed based on 5 ERS genes, namely CCND1, IGFBP2, TLR4, TXNIP, and VIM. The validation analysis demonstrated the commendable diagnostic performance of the model. Analysis of the ssGSEA immune characteristics revealed a positive correlation in the DR group between myeloid-derived suppressor cells (MDSC), regulatory T cells (Tregs), and CCND1 TXNIP. Furthermore, a significant negative correlation was observed between central memory CD4 T cells and CCND1. In the context of CIBERSORT, the results indicated a positive correlation between macrophages and IGFBP2, as well as Tregs and IGFBP2 in the DR group. Notably, a conspicuous negative correlation was identified between resting mast cells and IGFBP2. CONCLUSION: The present study provides novel diagnostic biomarkers for DR from an ERS perspective.
ABSTRACT
Cervical cancer (CCa) is the fourth most prevalent and common cancer affecting women worldwide, with increasing incidence and mortality rates. Hence, early detection of CCa plays a crucial role in improving outcomes. Non-invasive imaging procedures with good diagnostic performance are desirable and have the potential to lessen the degree of intervention associated with the gold standard, biopsy. Recently, artificial intelligence-based diagnostic models such as Vision Transformers (ViT) have shown promising performance in image classification tasks, rivaling or surpassing traditional convolutional neural networks (CNNs). This paper studies the effect of applying a ViT to predict CCa using different image benchmark datasets. A newly developed approach (ViT-PSO-SVM) was presented for boosting the results of the ViT based on integrating the ViT with particle swarm optimization (PSO), and support vector machine (SVM). First, the proposed framework extracts features from the Vision Transformer. Then, PSO is used to reduce the complexity of extracted features and optimize feature representation. Finally, a softmax classification layer is replaced with an SVM classification model to precisely predict CCa. The models are evaluated using two benchmark cervical cell image datasets, namely SipakMed and Herlev, with different classification scenarios: two, three, and five classes. The proposed approach achieved 99.112% accuracy and 99.113% F1-score for SipakMed with two classes and achieved 97.778% accuracy and 97.805% F1-score for Herlev with two classes outperforming other Vision Transformers, CNN models, and pre-trained models. Finally, GradCAM is used as an explainable artificial intelligence (XAI) tool to visualize and understand the regions of a given image that are important for a model's prediction. The obtained experimental results demonstrate the feasibility and efficacy of the developed ViT-PSO-SVM approach and hold the promise of providing a robust, reliable, accurate, and non-invasive diagnostic tool that will lead to improved healthcare outcomes worldwide.
ABSTRACT
OBJECTIVE: To develop a diagnostic model for predicting occult uterine sarcoma in patients with presumed uterine fibroids. MATERIALS AND METHODS: We retrospectively reviewed 41631 patients with presumed uterine fibroids who presented for HIFU treatment in 13 hospitals between November 2008 and October 2023. Of these patients, 27 with occult uterine sarcoma and 54 with uterine fibroids were enrolled. Univariate analysis and multivariate logistics regression analysis were used to determine the independent risk factors for the diagnosis of occult uterine sarcoma. A prediction model was constructed based on the coefficients of the risk factors. RESULTS: The multivariate analysis revealed abnormal vaginal bleeding, ill-defined boundary of tumor, hyperintensity on T2WI, and central unenhanced areas as independent risk factors. A scoring system was created to assess for occult uterine sarcoma risk. The score for abnormal vaginal bleeding was 56. The score for ill-defined lesion boundary was 90. The scores for lesions with hypointensity, isointensity signal/heterogeneous signal intensity, and hyperintensity on T2WI were 0, 42, and 93, respectively. The scores for lesions without enhancement on the mass margin, uniform enhancement of tumor, and no enhancement in the center of tumor were 0, 20, and 100, respectively. Patients with a higher total score implied a higher likelihood of a diagnosis of occult uterine sarcoma than that of patients with a lower score. The established model showed good predictive efficacy. CONCLUSIONS: Our results demonstrated that the diagnostic prediction model can be used to evaluate the risk of uterine sarcoma in patients with presumed uterine fibroids.
Subject(s)
High-Intensity Focused Ultrasound Ablation , Leiomyoma , Sarcoma , Uterine Neoplasms , Humans , Female , Leiomyoma/diagnostic imaging , Leiomyoma/therapy , Sarcoma/diagnostic imaging , Sarcoma/therapy , Middle Aged , Adult , Uterine Neoplasms/therapy , Risk Assessment , Retrospective Studies , High-Intensity Focused Ultrasound Ablation/methodsABSTRACT
Primary glomerular disease (PGD) is an idiopathic cause of renal glomerular lesions that is characterized by proteinuria or hematuria and is the leading cause of chronic kidney disease (CKD). The identification of circulating biomarkers for the diagnosis of PGD requires a thorough understanding of the metabolic defects involved. In this study, ultra-high performance liquid chromatography-tandem mass spectrometry was performed to characterize the amino acid (AA) profiles of patients with pathologically diagnosed PGD, including minimal change disease (MCD), focal segmental glomerular sclerosis (FSGS), membranous nephropathy, and immunoglobulin A nephropathy. The plasma concentrations of asparagine and ornithine were low, and that of aspartic acid was high, in patients with all the pathologic types of PGD, compared to healthy controls. Two distinct diagnostic models were generated using the differential plasma AA profiles using logistic regression and receiver operating characteristic analyses, with areas under the curves of 1.000 and accuracies up to 100.0% in patients with MCD and FSGS. In conclusion, the progression of PGD is associated with alterations in AA profiles, The present findings provide a theoretical basis for the use of AAs as a non-invasive, real-time, rapid, and simple biomarker for the diagnosis of various pathologic types of PGD.
Subject(s)
Amino Acids , Biomarkers , Metabolomics , Humans , Female , Male , Amino Acids/blood , Adult , Metabolomics/methods , Middle Aged , Biomarkers/blood , Glomerulosclerosis, Focal Segmental/blood , Glomerulosclerosis, Focal Segmental/diagnosis , Nephrosis, Lipoid/blood , Nephrosis, Lipoid/diagnosis , Glomerulonephritis, Membranous/blood , Glomerulonephritis, Membranous/diagnosis , Tandem Mass Spectrometry , Chromatography, High Pressure Liquid , Glomerulonephritis, IGA/blood , Glomerulonephritis, IGA/diagnosis , Kidney Glomerulus/metabolism , Kidney Glomerulus/pathologyABSTRACT
OBJECTIVE: To develop a machine learning diagnostic model based on MMP7 and other serological testing indicators for early and efficient diagnosis of biliary atresia (BA). METHODS: A retrospective analysis was conducted on patient information from those hospitalized for pathological jaundice at Beijing Children's Hospital between January 1, 2019, and December 31, 2023. Patients with serum MMP7, liver stiffness measurements, and other routine serological tests were included in the study. Six machine learning models were constructed, including logistic regression (LR), random forest (RF), decision tree (DET), support vector machine classifier (SVC), neural network (MLP), and extreme gradient boosting (XGBoost), to diagnose BA. The area under the receiver operating characteristic curve was used to evaluate the diagnostic efficacy of the various models. RESULTS: A total of 98 patients were included in the study, comprising 64 BA patients and 34 patients with other cholestatic liver diseases. Among the six machine learning models, the XGBoost algorithm model and RF algorithm model achieved the best predictive performance, with an AUROC of nearly 100% in both the training and validation sets. In the training set, these two algorithm models achieved an accuracy, precision, recall, F1 score, and AUROC of 1. Through model interpretation analysis, serum MMP7 levels, serum GGT levels, and acholic stools were identified as the most important indicators for diagnosing BA. The nomogram constructed based on the XGBoost algorithm model also demonstrated convenient and efficient diagnostic efficacy. CONCLUSION: Machine learning models, especially the XGBoost algorithm and RF algorithm models, constructed based on preoperative serum MMP7 and serological tests can diagnose BA more efficiently and accurately. The most important influencing factors for diagnosis are serum MMP7, serum GGT, and acholic stools.