Least Absolute Shrinkage And Selection Operator Algorithm Research Articles

Noninvasive Imaging Evaluation Based on Computed Tomography of the Efficacy of Initial Transarterial Chemoembolization to Predict Outcome in Patients with Hepatocellular Carcinoma.

PurposeThis study aims to develop a new model to more comprehensively and accurately predict the survival of patients with HCC after initial TACE.Patients and MethodsThe whole cohort (n = 102) was randomly divided into a training cohort and a validation cohort in the ratio of 8:2. The optimal radiomics signatures were screened using the least absolute shrinkage and selection operator algorithm (LASSO) regression for constructing the radscore to predict overall survival (OS). The C-index (95% confidence interval, CI), calibration curve, and decision curve analysis (DCA) were used to evaluate the performance of the models. The independent risk factors (hazard ratio, HR) for predicting OS were stratified by Kaplan–Meier (K-M) analysis and the Log rank test.ResultsThe median OS was 439 days (95% CI: 215.795–662.205) in whole cohort, and in the training cohort and validation cohort, the median OS was 552 days (95% CI: 171.172–932.828), 395 days (95% CI: 309.415–480.585), respectively (P = 0.889). After multivariate cox regression, the combined radscore-clinical model was consisted of radscore (HR: 2.065, 95% CI: 1.285–3.316; P = 0.0029) and post-response (HR: 1.880, 95% CI: 1.310–2.697; P = 0.0007), both of which were independent risk factors for the OS. In the validation cohort, the efficacy of both the radscore (C-index: 0.769, 95% CI: 0.496–1.000) and combined model (C-index: 0.770, 95% CI: 0.581–0.806) were higher than that of the clinical model (C-index: 0.655, 95% CI: 0.508–0.802). The calibration curve of the combined model for predicting OS presented good consistency between observations and predictions in both the training cohort and validation cohort.ConclusionNoninvasive imaging has a good prediction performance of survival after initial TACE in patients with HCC. The combined model consisting of post-response and radscore may be able to better predict outcome.

Application of a nomogram to radiomics labels in the treatment prediction scheme for lumbar disc herniation

ObjectiveTo investigate and verify the efficiency and effectiveness of a nomogram based on radiomics labels in predicting the treatment of lumbar disc herniation (LDH).MethodsBy reviewing medical records that were analysed over the past three years, clinical and imaging data of 200 lumbar disc patients at the Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine were obtained. The collected cases were randomly divided into a training group (n = 140) and a testing group (n = 60) at a ratio of 7:3. Two radiologists with experience in reading orthopaedics images independently segmented the ROIs. The whole intervertebral disc with the most obvious protrusion in the sagittal plane T2WI lumbar MRI as a mask (ROI) is sketched. The LASSO (Least Absolute Shrinkage And Selection Operator) algorithm was used to filter the features after extracting the radiomics features. The multivariate logistic regression model was used to construct a quantitative imaging Rad‑Score for the selected features with nonzero coefficients. The radiomics labels and nomogram were evaluated using the receiver operating characteristic curve (ROC) and the area under the curve (AUC). The calibration curve was used to evaluate the consistency between the nomogram prediction and the actual treatment plan. The DCA decision curve was used to evaluate the clinical applicability of the nomogram.ResultFollowing feature extraction, 11 radiomics features were used to construct the radiomics label for predicting the treatment plan of LDH. A nomogram was then constructed. The AUC was 0.93 (95% CI: 0.90–0.97), with a sensitivity of 89%, a specificity of 91%, a positive predictive value of 92.7%, a negative predictive value of 89.4%, and an accuracy of 91%. The calibration curve showed that there was good consistency between the prediction and the actual observation. The DCA decision curve analysis showed that the nomogram of the imaging group has great potential for clinical application when the risk threshold is between 5 and 72%.ConclusionA nomogram based on radiomics labels has good predictive value for the treatment of LDH and can be used as a reference for clinical decision-making.

A Multiparametric MRI-Based Radiomics Nomogram for Preoperative Prediction of Survival Stratification in Glioblastoma Patients With Standard Treatment.

ObjectiveTo construct and validate a radiomics nomogram for preoperative prediction of survival stratification in glioblastoma (GBM) patients with standard treatment according to radiomics features extracted from multiparameter magnetic resonance imaging (MRI), which could facilitate clinical decision-making.MethodsA total of 125 eligible GBM patients (53 in the short and 72 in the long survival group, separated by an overall survival of 12 months) were randomly divided into a training cohort (n = 87) and a validation cohort (n = 38). Radiomics features were extracted from the MRI of each patient. The T-test and the least absolute shrinkage and selection operator algorithm (LASSO) were used for feature selection. Next, three feature classifier models were established based on the selected features and evaluated by the area under curve (AUC). A radiomics score (Radscore) was then constructed by these features for each patient. Combined with clinical features, a radiomics nomogram was constructed with independent risk factors selected by the logistic regression model. The performance of the nomogram was assessed by AUC, calibration, discrimination, and clinical usefulness.ResultsThere were 5,216 radiomics features extracted from each patient, and 5,060 of them were stable features judged by the intraclass correlation coefficients (ICCs). 21 features were included in the construction of the radiomics score. Of three feature classifier models, support vector machines (SVM) had the best classification effect. The radiomics nomogram was constructed in the training cohort and exhibited promising calibration and discrimination with AUCs of 0.877 and 0.919 in the training and validation cohorts, respectively. The favorable decision curve analysis (DCA) indicated the clinical usefulness of the radiomics nomogram.ConclusionsThe presented radiomics nomogram, as a non-invasive tool, achieved satisfactory preoperative prediction of the individualized survival stratification of GBM patients.

TNF Family-Based Signature Predicts Prognosis, Tumor Microenvironment, and Molecular Subtypes in Bladder Carcinoma.

Background: Tumor necrosis factor (TNF) family members play vital roles in cancer development and antitumor immune responses. However, the expression patterns, prognostic values, and immunological characteristics of TNF members in bladder carcinoma (BLCA) remain unclear. Methods: The training cohort, TCGA-BLCA, was downloaded from The Cancer Genome Atlas; another two Gene Expression Omnibus datasets (GSE13507 and GSE32894) and the Xiangya cohort (RNA-sequencing cohort collected from our hospital) were used as the external validation cohort. The least absolute shrinkage and selection operator (LASSO) algorithm and cross-validation were used to screen variables. Cox regression model and random survival forest (RSF) were used to develop the risk score, respectively. Then, we systematically correlated the TNF risk score with the tumor microenvironment (TME) cell infiltration, molecular subtypes of BLCA, and the potential value for predicting the efficacy of immunotherapy. Results: We developed two TNF-based patterns, named TNF cluster 1 and TNF cluster 2. TNF cluster 1 exhibited poorer survival outcome and an inflamed TME characteristic compared with TNF cluster 2. We then filtered out 196 differentially expressed genes between the two TNF clusters and applied the LASSO algorithm and cross-validation to screen out 22 genes to build the risk score. For risk score, we found that RSF exhibited higher efficacy than the Cox regression model, and we chose the risk score developed by RSF for the following analysis. BLCA patients in the higher risk score group showed significantly poorer survival outcomes. Moreover, these results could be validated in the external validation cohorts, including the GSE13507, GSE32894, and Xiangya cohorts. Then, we systematically correlated the risk score with TME cell infiltration and found that it was positively correlated with the infiltration of a majority of immune cells. Also, a higher risk score indicated a basal subtype of BLCA. Notably, the relationship between risk score, TME cell infiltration, and molecular subtypes could be validated in the Xiangya cohort. Conclusion: We developed and validated a robust TNF-based risk score, which could predict prognostic outcomes, TME, and molecular subtypes of BLCA. However, the value of risk score predicting the efficacy of immunotherapy needs further research.

Risk stratification of lung adenocarcinoma using a nomogram combined with ferroptosis-related LncRNAs and subgroup analysis with immune and N6-methyladenosine modification

BackgroundDetermining the prognosis of lung adenocarcinoma (LUAD) is challenging. The present study aimed to identify prognostic ferroptosis-related long noncoding RNAs (FRLs) and construct a prognostic model. Moreover, differential analysis of immune and N6-methyladenosine (m6A)-related genes was systematically conducted.MethodsA total of 504 patients selected from a dataset from The Cancer Genome Atlas were included. The patients with LUAD were randomly divided into a training group and a test group at a ratio of 1:1. Pearson correlation analysis and univariate Cox regression analysis were used to identify the prognostic FRLs. Then, a prognostic model was constructed from the optimized subset of prognostic FRLs based on the least absolute shrinkage and selection operator (LASSO) algorithm. Subsequently, the receiver operating characteristic (ROC) curve and survival analysis were used to evaluate the performance of the model. The risk score based on the prognostic model was analyzed using Cox regression analysis. Moreover, gene set enrichment analysis and differential analysis of immune- and m6A-related genes were conducted.ResultsAfter univariate Cox regression analysis and LASSO algorithm analysis, a total of 19 prognostic FRLs were selected to construct the final model to obtain the risk score. The area under the ROC curve of the prognostic model for 1-year, 3-year, and 5-year overall survival (OS) was 0.763, 0.745, and 0.778 in the training set and 0.716, 0.724, and 0.736 in the validation set, respectively. Moreover, the OS of the high-risk group was significantly worse than that of the low-risk group in the training group (P < 0.001) and in the test group (P < 0.001). After univariate and multivariate Cox regression analysis, the risk score [hazard ratio (HR) = 1.734; P < 0.001] and stage (HR = 1.557; P < 0.001) were both considered significant prognostic factors for LUAD. A nomogram was constructed based on clinical features and risk score. The expression of 34 checkpoint genes and 13 m6A-related genes varied significantly between the two risk groups.ConclusionThis study constructed a prognostic model to effectively predict the OS of patients with LUAD, and these OS-related FRLs might serve as potential therapeutic targets of LUAD.

Identification of Characteristic Genes in Whole Blood of Intervertebral Disc Degeneration Patients by Weighted Gene Coexpression Network Analysis (WGCNA).

Intervertebral disc degeneration (IDD) is a major cause of lower back pain. However, to date, the molecular mechanism of the IDD remains unclear. Gene expression profiles and clinical traits were downloaded from the Gene Expression Omnibus (GEO) database. Firstly, weighted gene coexpression network analysis (WGCNA) was used to screen IDD-related genes. Moreover, least absolute shrinkage and selection operator (LASSO) logistic regression and support vector machine (SVM) algorithms were used to identify characteristic genes. Furthermore, we further investigated the immune landscape by the Cell-type Identification By Estimating Relative Subsets Of RNA Transcripts (CIBERSORT) algorithm and the correlations between key characteristic genes and infiltrating immune cells. Finally, a competing endogenous RNA (ceRNA) network was established to show the regulatory mechanisms of characteristic genes. A total of 2458 genes were identified by WGCNA, and 48 of them were disordered. After overlapping the genes obtained by LASSO and SVM-RFE algorithms, genes including LINC01347, ASAP1-IT1, lnc-SEPT7L-1, B3GNT8, CHRNB3, CLEC4F, LOC102724000, SERINC2, and LOC102723649 were identified as characteristic genes of IDD. Moreover, differential analysis further identified ASAP1-IT1 and SERINC2 as key characteristic genes. Furthermore, we found that the expression of both ASAP1-IT1 and SERINC2 was related to the proportions of T cells gamma delta and Neutrophils. Finally, a ceRNA network was established to show the regulatory mechanisms of ASAP1-IT1 and SERINC2. In conclusion, the present study identified ASAP1-IT1 and SERINC2 as the key characteristic genes of IDD through integrative bioinformatic analyses, which may contribute to the diagnosis and treatment of IDD.

Magnetic resonance imaging-based radiomics signature for preoperative prediction of Ki67 expression in bladder cancer

PurposeThe Ki67 expression is associated with the advanced clinicopathological features and poor prognosis in bladder cancer (BCa). We aimed to develop and validate magnetic resonance imaging (MRI)-based radiomics signatures to preoperatively predict the Ki67 expression status in BCa.Methods and materialsWe retrospectively collected 179 BCa patients with Ki67 expression and preoperative MRI. Radiomics features were extracted from T2-weighted (T2WI) and dynamic contrast-enhancement (DCE) images. The synthetic minority over-sampling technique (SMOTE) was used to balance the minority group (low Ki67 expression group) in the training set. Minimum redundancy maximum relevance was used to identify the best features associated with Ki67 expression. Support vector machine and Least Absolute Shrinkage and Selection Operator algorithms (LASSO) were used to construct radiomics signatures in training and SMOTE-training sets, and diagnostic performance was assessed by the area under the curve (AUC) and accuracy. The decision curve analyses (DCA) and calibration curve and were used to investigate the clinical usefulness and calibration of radiomics signatures, respectively. The Kaplan-Meier test was performed to investigate the prognostic value of radiomics-predicted Ki67 expression status.Results1218 radiomics features were extracted from T2WI and DCE images, respectively. The SMOTE-LASSO model based on nine features achieved the best predictive performance in the SMOTE-training (AUC, 0.859; accuracy, 80.3%) and validation sets (AUC, 0.819; accuracy, 81.5%) with a good calibration performance and clinical usefulness. Immunohistochemistry-based high Ki67 expression and radiomics-predicted high Ki67 expression based on the SMOTE-LASSO model were significantly associated with poor disease-free survival in training and validation sets (all P < 0.05).ConclusionsThe SMOTE-LASSO model could predict the Ki67 expression status and was associated with survival outcomes of the BCa patients, thereby may aid in clinical decision-making.

Analysis and Construction of a Molecular Diagnosis Model of Drug-Resistant Epilepsy Based on Bioinformatics.

Background: Epilepsy is a complex chronic disease of the nervous system which influences the health of approximately 70 million patients worldwide. In the past few decades, despite the development of novel antiepileptic drugs, around one-third of patients with epilepsy have developed drug-resistant epilepsy. We performed a bioinformatic analysis to explore the underlying diagnostic markers and mechanisms of drug-resistant epilepsy. Methods: Weighted correlation network analysis (WGCNA) was applied to genes in epilepsy samples downloaded from the Gene Expression Omnibus database to determine key modules. The least absolute shrinkage and selection operator (LASSO) regression and support vector machine-recursive feature elimination (SVM-RFE) algorithms were used to screen the genes resistant to carbamazepine, phenytoin, and valproate, and sensitivity of the three-class classification SVM model was verified through the receiver operator characteristic (ROC) curve. A protein–protein interaction (PPI) network was utilized to analyze the protein interaction relationship. Finally, ingenuity pathway analysis (IPA) was adopted to conduct disease and function pathway and network analysis. Results: Through WGCNA, 72 genes stood out from the key modules related to drug resistance and were identified as candidate resistance genes. Intersection analysis of the results of the LASSO and SVM-RFE algorithms selected 11, 4, and 5 drug-resistant genes for carbamazepine, phenytoin, and valproate, respectively. Subsequent union analysis obtained 17 hub resistance genes to construct a three-class classification SVM model. ROC showed that the model could accurately predict patient resistance. Expression of 17 hub resistance genes in healthy subjects and patients was significantly different. The PPI showed that there are six resistance genes (CD247, CTSW, IL2RB, MATK, NKG7, and PRF1) that may play a central role in the resistance of epilepsy patients. Finally, IPA revealed that resistance genes (PRKCH and S1PR5) were involved in “CREB signaling in Neurons.” Conclusion: We obtained a three-class SVM model that can accurately predict the drug resistance of patients with epilepsy, which provides a new theoretical basis for research and treatment in the field of drug-resistant epilepsy. Moreover, resistance genes PRKCH and S1PR5 may cooperate with other resistance genes to exhibit resistance effects by regulation of the cAMP-response element-binding protein (CREB) signaling pathway.

Immune Characteristics Analysis and Transcriptional Regulation Prediction Based on Gene Signatures of Chronic Obstructive Pulmonary Disease.

PurposeThe variation in inflammation in chronic obstructive pulmonary disease (COPD) between individuals is genetically determined. This study aimed to identify gene signatures of COPD through bioinformatics analysis based on multiple gene sets and explore their immune characteristics and transcriptional regulation mechanisms.MethodsData from four microarrays were downloaded from the Gene Expression Omnibus database to screen differentially expressed genes (DEGs) between COPD patients and controls. Weighted gene co-expression network analysis was applied to identify trait-related modules and then select key module-related DEGs. The optimized gene set of signatures was obtained using the least absolute shrinkage and selection operator (LASSO) regression analysis. The CIBERSORT algorithm and Pearson correlation test were used to analyze the relationship between gene signatures and immune cells. Finally, public databases were used to predict the transcription factors (TFs) and upstream miRNAs.ResultsA total of 127 DEGs in COPD were identified from the combined dataset. By considering the intersection of DEGs and genes in two trait-related modules, 83 key module-related DEGs were identified, which were mainly enriched in interleukin-related pathways. Seven-gene signatures, including MTHFD2, KANK3, GFPT2, PHLDA1, HS3ST2, FGG, and RPS4Y1, were further selected using the LASSO algorithm. These gene signatures showed the predictive potential for COPD risks and were significantly correlated with 18 types of immune cells. Finally, nine miRNAs and three TFs were predicted to target MTHFD2, GFPT2, PHLDA1, and FGG.ConclusionWe proposed the seven-gene-signature to predict COPD risk and explored its potential immune characteristics and regulatory mechanisms.

Development and Verification of Glutamatergic Synapse-Associated Prognosis Signature for Lower-Grade Gliomas.

Background: Lower-grade glioma (LGG) is the most common histology identified in gliomas, a heterogeneous tumor that may develop into high-grade malignant glioma that seriously shortens patient survival time. Recent studies reported that glutamatergic synapses might play an essential role in the progress of gliomas. However, the role of glutamatergic synapse-related biomarkers in LGG has not been systemically researched yet.Methods: The mRNA expression data of glioma and normal brain tissue were obtained from The Cancer Genome Atlas database and Genotype-Tissue Expression, respectively, and the Chinese Glioma Genome Atlas database was used as a validation set. Difference analysis was performed to evaluate the expression pattern of glutamatergic synapse-related genes (GSRGs) in LGG. The least absolute shrinkage and selection operator (LASSO) Cox regression was applied to construct the glutamatergic synapse-related risk signature (GSRS), and the risk score of each LGG sample was calculated based on the coefficients and expression value of selected GSRGs. Univariate and multivariate Cox regression analyses were used to investigate the prognostic value of risk score. Immunity profile and single-sample gene set enrichment analysis (ssGSEA) were performed to explore the association between risk score and the characters of tumor microenvironment in LGG. Gene set variation analysis (GSVA) was performed to investigate the potential pathways related to GSRS. The HPA database and real-time PCR were used to identify the expression of hub genes identified in GSRS.Results: A total of 22 genes of 39 GSRGs were found differentially expressed among normal and LGG samples. Through the LASSO algorithm, 14-genes GSRS constructed were associated with the prognosis and clinicopathological features of patients with LGG. Furthermore, the risk score level was significantly positively correlated with the infiltrating level of immunosuppressive cells, including M2 macrophages and regulatory T cells. GSVA identified a series of cancer-related pathways related to GSRS, such as P13K-AKT and P53 pathways. Moreover, ATAD1, NLGN2, OXTR, and TNR, hub genes identified in GSRS, were considered as potential prognostic biomarkers in LGG.Conclusion: A 14-genes GSRS was constructed and verified in this study. We provided a novel insight into the role of GSRS in LGG through a series of bioinformatics methods.

Multi-Zone Active Noise Control Strategy for the Scattered Sound Control of an Infinite Rigid Cylinder

Active noise control can be used to reduce the scattered sound of a reflecting object to make it invisible to incident acoustic waves. For the multi-zone active noise control of scattered sound from an infinite rigid cylinder, an active control strategy is proposed that combines the least absolute shrinkage and selection operator (LASSO) algorithm with constraint points and regularized least squares (RLS) algorithm. The proposed control strategy is used to promote control performance through optimizing the secondary loudspeaker placement of an active noise control system. Compared with the RLS algorithm employing the uniformly placed loudspeakers and the traditional LASSO algorithm, the proposed strategy has better reduction performance both in the forward-scattered and backward-scattered sound target areas, and there is less sound amplification in the far field. From 400 Hz–1100 Hz, the proposed strategy provides a 5 dB–16 dB reduction performance advantage in the target area compared to the RLS algorithm employing uniformly placed loudspeakers.

A Five-MicroRNA Signature Predicts the Prognosis in Nasopharyngeal Carcinoma.

BackgroundThere is no effective prognostic signature that could predict the prognosis of nasopharyngeal carcinoma (NPC).MethodsWe constructed a prognostic signature based on five microRNAs using random forest and Least Absolute Shrinkage And Selection Operator (LASSO) algorithm on the GSE32960 cohort (N = 213). We verified its prognostic value using three independent external validation cohorts (GSE36682, N = 62; GSE70970, N = 246; and TCGA-HNSC, N = 523). Through principal component analysis, receiver operating characteristic curve analysis, and C-index calculation, we confirmed the predictive accuracy of this prognostic signature.ResultsWe calculated the risk score based on the LASSO algorithm and divided the patients into high- and low-risk groups according to the calculated optimal cutoff value. The patients in the high-risk group tended to have a worse prognosis outcome and chemotherapy response. The time-dependent receiver operating characteristic curve showed that the 1-year overall survival rate of the five-microRNA signature had an area under the curve of more than 0.83. A functional annotation analysis of the five-microRNA signature showed that the patients in the high-risk group were usually accompanied by activation of DNA repair and MYC-target pathways, while the patients in the low-risk group had higher immune-related pathway signals.ConclusionsWe constructed a five-microRNA prognostic signature, which could accurately predict the prognosis of nasopharyngeal carcinoma, and constructed a nomogram that could conveniently predict the overall survival of patients.

Bioinformatics Analysis Identifies Molecular Markers Regulating Development and Progression of Endometriosis and Potential Therapeutic Drugs.

Endometriosis, a common disease that presents as polymorphism, invasiveness, and extensiveness, with clinical manifestations including dysmenorrhea, infertility, and menstrual abnormalities, seriously affects quality of life in women. To date, its underlying etiological mechanism of action and the associated regulatory genes remain unclear. This study aimed to identify molecular markers and elucidate mechanisms underlying the development and progression of endometriosis. Specifically, we downloaded five microarray expression datasets, namely, GSE11691, GSE23339, GSE25628, GSE7305, and GSE105764, from the Gene Expression Omnibus (GEO) database. These datasets, obtained from endometriosis tissues, alongside normal controls, were subjected to in-depth bioinformatics analysis for identification of differentially expressed genes (DEGs), followed by analysis of their function and pathways via gene ontology (GO) and KEGG pathway enrichment analyses. Moreover, we constructed a protein–protein interaction (PPI) network to explore the hub genes and modules, and then applied machine learning algorithms support vector machine-recursive feature elimination and least absolute shrinkage and selection operator (LASSO) analysis to identify key genes. Furthermore, we adopted the CIBERSORTx algorithm to estimate levels of immune cell infiltration while the connective map (CMAP) database was used to identify potential therapeutic drugs in endometriosis. As a result, a total of 423 DEGs, namely, 233 and 190 upregulated and downregulated, were identified. On the other hand, a total of 1,733 PPIs were obtained from the PPI network. The DEGs were mainly enriched in immune-related mechanisms. Furthermore, machine learning and LASSO algorithms identified three key genes, namely, apelin receptor (APLNR), C–C motif chemokine ligand 21 (CCL21), and Fc fragment of IgG receptor IIa (FCGR2A). Furthermore, 16 small molecular compounds associated with endometriosis treatment were identified, and their mechanism of action was also revealed. Taken together, the findings of this study provide new insights into the molecular factors regulating occurrence and progression of endometriosis and its underlying mechanism of action. The identified therapeutic drugs and molecular markers may have clinical significance in early diagnosis of endometriosis.

Development and Validation of a Practical Prognostic Coagulation Index for Patients with Esophageal Squamous Cell Cancer.

This study aimed to establish an effective and practical prognostic index for esophageal squamous cell cancer (ESCC) based on the coagulation factors. The training cohort of 965 patients with ESCC was retrospectively collected at Sichuan Cancer Hospital from 2012 to 2014, along with clinical characteristics and follow-up information. Risk factors of coagulation status, including 11 blood parameters (platelet [PLT], mean platelet volume [MPV], platelet distribution width [PDW], plateletocrit [PCT], thrombin time [TT], prothrombin time [PT], international normalized ratio [INR], activated partial thromboplastin time [APTT], fibrinogen, D-dimer, and fibrinogen degradation product [FDP]), were studied by least absolute shrinkage and selection operator (LASSO) Cox regression and the Coagulation Index was established. The index was validated in a cohort of 848 patients with ESCC at the same institution, from 2015 to 2016. Three variables of PLT, MPV, and fibrinogen were identified by selecting features with coefficients in the LASSO algorithm, and a Coagulation Index was established as follows: Coagulation Index = 0.0005 × PLT(109/L) - 0.0384 × MPV (fL) + 0.1148 × fibrinogen (g/L). A higher Coagulation Index score was significantly associated with higher pT stage and pN stage (p<0.05). With this prognostic index, patients could be stratified into three risk groups. The 3-year overall survival (OS) rates of the low-, middle- and high-risk groups in the training cohort were 63.5%, 55.5% and 43.1%, respectively (log-rank p<0.001). Similarly, in the validation set, the respective 3-year OS for each risk group was significantly different across the three risk groups. Multivariate analysis indicated that the Coagulation Index remained a significant factor for predicting OS, independently of pathological TNM stage. The Coagulation Index is an independent predictor of survival for patients with ESCC.

Radiomics Nomogram for Identifying Sub-1 cm Benign and Malignant Thyroid Lesions.

PurposeTo develop and validate a radiomics nomogram for identifying sub-1 cm benign and malignant thyroid lesions.MethodA total of 171 eligible patients with sub-1 cm thyroid lesions (56 benign and 115 malignant) who were treated in Yantai Yuhuangding Hospital between January and September 2019 were retrospectively collected and randomly divided into training (n = 136) and validation sets (n = 35). The radiomics features were extracted from unenhanced and arterial contrast-enhanced computed tomography images of each patient. In the training set, one-way analysis of variance and least absolute shrinkage and selection operator (LASSO) logistic regression were used to select the features related to benign and malignant lesions, and the LASSO algorithm was used to construct the radiomics signature. Combined with clinical independent predictive factors, a radiomics nomogram was constructed with a multivariate logistic regression model. The performance of the radiomics nomogram was evaluated by using the receiver operating characteristic (ROC) and calibration curves in the training and validation sets. The clinical usefulness was evaluated by using decision curve analysis (DCA).ResultsThe radiomics signature consisting of 13 selected features achieved favorable prediction efficiency. The radiomics nomogram, which incorporated radiomics signature and clinical independent predictive factors including age and Thyroid Imaging Reporting and Data System category, showed good calibration and discrimination in the training (area under the ROC [AUC]: 0.853; 95% confidence interval [CI]: 0.797, 0.899) and validation sets (AUC: 0.851; 95% CI: 0.735, 0.931). DCA demonstrated that the nomogram was clinically useful.ConclusionAs a noninvasive preoperative prediction tool, the radiomics nomogram incorporating radiomics signature and clinical predictive factors shows favorable predictive efficiency for identifying sub-1 cm benign and malignant thyroid lesions.

Development and Validation of a Predictive Model for Severe COVID-19: A Case-Control Study in China.

Background: Predicting the risk of progression to severe coronavirus disease 2019 (COVID-19) could facilitate personalized diagnosis and treatment options, thus optimizing the use of medical resources.Methods: In this prospective study, 206 patients with COVID-19 were enrolled from regional medical institutions between December 20, 2019, and April 10, 2020. We collated a range of data to derive and validate a predictive model for COVID-19 progression, including demographics, clinical characteristics, laboratory findings, and cytokine levels. Variation analysis, along with the least absolute shrinkage and selection operator (LASSO) and Boruta algorithms, was used for modeling. The performance of the derived models was evaluated by specificity, sensitivity, area under the receiver operating characteristic (ROC) curve (AUC), Akaike information criterion (AIC), calibration plots, decision curve analysis (DCA), and Hosmer–Lemeshow test.Results: We used the LASSO algorithm and logistic regression to develop a model that can accurately predict the risk of progression to severe COVID-19. The model incorporated alanine aminotransferase (ALT), interleukin (IL)-6, expectoration, fatigue, lymphocyte ratio (LYMR), aspartate transaminase (AST), and creatinine (CREA). The model yielded a satisfactory predictive performance with an AUC of 0.9104 and 0.8792 in the derivation and validation cohorts, respectively. The final model was then used to create a nomogram that was packaged into an open-source and predictive calculator for clinical use. The model is freely available online at https://severeconid-19predction.shinyapps.io/SHINY/.Conclusion: In this study, we developed an open-source and free predictive calculator for COVID-19 progression based on ALT, IL-6, expectoration, fatigue, LYMR, AST, and CREA. The validated model can effectively predict progression to severe COVID-19, thus providing an efficient option for early and personalized management and the allocation of appropriate medical resources.

Machine learning algorithm outperforms fibrosis markers in predicting significant fibrosis in biopsy-confirmed NAFLD.

The presence of significant liver fibrosis is a key determinant of long-term prognosis in non-alcoholic fatty liver disease (NAFLD). We aimed to develop a novel machine learning algorithm (MLA) to predict fibrosis severity in NAFLD and compared it with the most widely used non-invasive fibrosis biomarkers. We used a cohort of 553 adults with biopsy-proven NAFLD, who were randomly divided into a training cohort (n=278) for the development of both logistic regression model (LRM) and MLA, and a validation cohort (n=275). Significant fibrosis was defined as fibrosis stage F≥2. MLA and LRM were derived from variables that were selected using a least absolute shrinkage and selection operator (LASSO) logistic regression algorithm. In the training cohort, the variables selected by LASSO algorithm were body mass index, pro-collagen type III, collagen type IV, aspartate aminotransferase and albumin-to-globulin ratio. The diagnostic accuracy of MLA showed the highest values of area under the receiver operator characteristic curve (AUROC: 0.902, 95% CI 0.869-0.904) for identifying fibrosis F≥2. The LRM AUROC was 0.764, 95% CI 0.710-0.816 and significantly better than the AST-to-Platelet ratio (AUROC 0.684, 95% CI 0.605-0.762), FIB-4 score (AUROC 0.594, 95% CI 0.503-0.685) and NAFLD Fibrosis Score (AUROC 0.557, 95% CI 0.470-0.644). In the validation cohort, MLA also showed the highest AUROC (0.893, 95% CI 0.864-0.901). The diagnostic accuracy of MLA outperformed that of LRM in all subgroups considered. Our newly developed MLA algorithm has excellent diagnostic performance for predicting fibrosis F≥2 in patients with biopsy-confirmed NAFLD.

Identification and Validation of Ubiquitin-Specific Proteases as a Novel Prognostic Signature for Hepatocellular Carcinoma.

PurposeUbiquitin-specific proteases (USPs), as a sub-family of deubiquitinating enzymes (DUBs), are responsible for the elimination of ubiquitin-triggered modification. USPs are recently correlated with various malignancies. However, the expression features and clinical significance of USPs have not been systematically investigated in hepatocellular carcinoma (HCC).MethodsGenomic alterations and expression profiles of USPs were investigated in CbioPortal and The Cancer Genome Atlas (TCGA) Liver hepatocellular carcinoma (LIHC) dataset. Cox regression and least absolute shrinkage and selection operator (LASSO) analyses were conducted to establish a risk signature for HCC prognosis in TCGA LIHC cohort. Subsequently, Kaplan-Meier analysis, receiver operating characteristic (ROC) curves and univariate/multivariate analyses were performed to evaluate the prognostic significance of the risk signature in TCGA LIHC and international cancer genome consortium (ICGC) cohorts. Furthermore, we explored the alterations of the signature genes during hepatocarcinogenesis and HCC progression in GSE89377. In addition, the expression feature of USP39 was further explored in HCC tissues by performing western blotting and immunohistochemistry.ResultsGenomic alterations and overexpression of USPs were observed in HCC tissues. The consensus analysis indicated that the USPs-overexpressed sub-Cluster was correlated with aggressive characteristics and poor prognosis. Cox regression with LASSO algorithm identified a risk signature formed by eight USPs for HCC prognosis. High-risk group stratified by the signature score was correlated with advanced tumor stage and poor survival HCC patients in TCGA LIHC cohort. In addition, the 8-USPs based signature could also robustly predict overall survival of HCC patients in ICGC(LIRI-JP) cohort. Furthermore, gene sets enrichment analysis (GSEA) showed that the high-risk score was associated with tumor-related pathways. According to the observation in GSE89377, USP39 expression was dynamically increased with hepatocarcinogenesis and HCC progression. The overexpression of USP39 was further determined in a local HCC cohort and correlated with poor prognosis. The co-concurrence analysis suggested that USP39 might promote HCC by regulating cell-cycle- and proliferation- related genes.ConclusionThe current study provided a USPs-based signature, highlighting its robust prognostic significance and targeted value for HCC treatment.

Evaluating treatment response to neoadjuvant chemoradiotherapy in rectal cancer using various MRI-based radiomics models

BackgroundTo validate and compare various MRI-based radiomics models to evaluate treatment response to neoadjuvant chemoradiotherapy (nCRT) of rectal cancer.MethodsA total of 80 patients with locally advanced rectal cancer (LARC) who underwent surgical resection after nCRT were enrolled retrospectively. Rectal MR images were scanned pre- and post-nCRT. The radiomics features were extracted from T2-weighted images, then reduced separately by least absolute shrinkage and selection operator (LASSO) and principal component analysis (PCA). Four classifiers of Logistic Regression, Random Forest (RF), Decision Tree and K-nearest neighbor (KNN) models were constructed to assess the tumor regression grade (TRG) and pathologic complete response (pCR), respectively. The diagnostic performances of models were determined with leave-one-out cross-validation by generating receiver operating characteristic curves and decision curve analysis.ResultsThree features related to the TRG and 11 features related to the pCR were obtained by LASSO. Top five principal components representing a cumulative contribution of 80% to overall features were selected by PCA. For TRG, the area under the curve (AUC) of RF model was 0.943 for LASSO and 0.930 for PCA, higher than other models (P < 0.05 for both). As for pCR, the AUCs of KNN for LASSO and PCA were 0.945 and 0.712, higher than other models (P < 0.05 for both). The DCA showed that LASSO algorithm was clinically superior to PCA.ConclusionMRI-based radiomics models demonstrated good performance for evaluating the treatment response of LARC after nCRT and LASSO algorithm yielded more clinical benefit.

Finding the best-fit background function for whole-powder-pattern fitting using LASSO combined with tree search

A new linear function for modelling the background in whole-powder-pattern fitting has been derived by applying LASSO (least absolute shrinkage and selection operator) and the technique of tree search. The background function (BGF) consists of terms b n L(2θ/180)−n/2 and b n H(1 − 2θ/180)−n/2 for the low- and high-angle sides, respectively. Some variable parameters of the BGF should be fixed at zero while others should be varied in order to find the best fit for a given data set without inducing overfitting. The LASSO algorithm can automatically select the variables in linear regression analysis. However, it finds the best-fit BGF with a set of adjustable parameters for a given data set while it derives a different set of parameters for a different data set. Thus, LASSO derives multiple solutions depending on the data set used. By regarding the individual solutions from LASSO as nodes of trees, tree structures were constructed from these solutions. The root node has the maximum number of adjustable parameters, P. P decreases with descending levels of the tree one by one, and leaf nodes have just one parameter. By evaluating individual solutions (nodes) by their χ2 index, the best-fit single path from a root node to a leaf node was found. The present BGF can be used simply by varying P in the range 1–10. The BGF thus derived as a final single solution was incorporated into computer programs for Pawley-based whole-powder-pattern decomposition and Rietveld refinement, and the performance of the BGF was tested in comparison with the polynomials currently widely used as the BGF. The present BGF has been demonstrated to be stable and to give an excellent fit, comparable to polynomials but with a smaller number of adjustable parameters and without introducing undulation into the calculated background curve. Basic algorithms used in statistics and machine learning have been demonstrated to be useful in developing an analytical model in X-ray crystallography.