Visualising statistical models using dynamic nomograms

Jalali, Amirhossein; Alvarez‐Iglesias, Alberto; Roshan, Davood; Newell, John

doi:10.1371/journal.pone.0225253

Cited by 103 publications

(84 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, using the selected predictors from the LASSO regression, a nomogram was developed using the “rms,” “survival,” “foreign,” and “openxlsx” R packages ( 15 – 18 ). A dynamic web-based calculator was constructed using “DynNom” package ( 19 ).…”

Section: Methodsmentioning

confidence: 99%

A Simple-to-Use Web-Based Calculator for Survival Prediction in Acute Respiratory Distress Syndrome

Liu

Huang

2021

Front. Med.

View full text Add to dashboard Cite

Background: The aim of this study was to construct and validate a simple-to-use model to predict the survival of patients with acute respiratory distress syndrome.Methods: A total of 197 patients with acute respiratory distress syndrome were selected from the Dryad Digital Repository. All eligible individuals were randomly stratified into the training set (n=133) and the validation set (n=64) as 2: 1 ratio. LASSO regression analysis was used to select the optimal predictors, and receiver operating characteristic and calibration curves were used to evaluate accuracy and discrimination of the model. Clinical usefulness of the model was also assessed using decision curve analysis and Kaplan-Meier analysis.Results: Age, albumin, platelet count, PaO2/FiO2, lactate dehydrogenase, high-resolution computed tomography score, and etiology were identified as independent prognostic factors based on LASSO regression analysis; these factors were integrated for the construction of the nomogram. Results of calibration plots, decision curve analysis, and receiver operating characteristic analysis showed that this model has good predictive ability of patient survival in acute respiratory distress syndrome. Moreover, a significant difference in the 28-day survival was shown between the patients stratified into different risk groups (P < 0.001). For convenient application, we also established a web-based calculator (https://huangl.shinyapps.io/ARDSprognosis/).Conclusions: We satisfactorily constructed a simple-to-use model based on seven relevant factors to predict survival and prognosis of patients with acute respiratory distress syndrome. This model can aid personalized treatment and clinical decision-making.

show abstract

Section: Methodsmentioning

confidence: 99%

A Simple-to-Use Web-Based Calculator for Survival Prediction in Acute Respiratory Distress Syndrome

Liu

Huang

2021

Front. Med.

View full text Add to dashboard Cite

show abstract

“…Points were assigned to values of age (20 to 90 years in 5‐year steps), sex, genotype, and tau (with cut points 250; 500; 1000; 2500; 5000; 10,000; 25,000; 50,000 pg/mL) based on the Cox regression coefficients. The points measure the contribution of the values of the predictor variables on the “linear predictor” scale 26 . For each possible sum of points, the survival probability can be derived from S(t) = S 0 (t) exp(total points) where S(t) denotes the survival function and S 0 (t) denotes the baseline survival function for an individual whose covariates are equal to zero 27 .…”

Section: Methodsmentioning

confidence: 99%

A prognostic model for overall survival in sporadic Creutzfeldt‐Jakob disease

Llorens

Rübsamen

Hermann³

et al. 2020

Alzheimer's & Dementia

View full text Add to dashboard Cite

Introduction: We developed a prognostic model for overall survival after diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD) using data from a German surveillance study. Methods: We included 1226 sCJD cases (median age 66 years, range 19-89 years; 56.8% women with information on age, sex, codon 129 genotype, 14-3-3 in the cerebrospinal fluid (CSF), and CSF tau concentrations. The prognostic accuracy for overall survival was measured by the c statistics of multivariable Cox proportional hazard models. A score chart was derived to predict 6-month survival and median survival time. Results: A model containing age, sex, codon 129 genotype, and CSF tau (with two-way interactions) was selected as the model with the highest c statistic (0.686, 95% confidence interval: 0.665-0.707) in a cross-validation approach. Discussion: We developed the first prognostic model for overall survival of sCJD patients based on readily available information only. The developed score chart serves as a hands-on prediction tool for clinical practice.

show abstract

“…All statistical analyses were performed with R software, version 3.6.1 (The R Foundation). Lastly, we have developed a dynamic nomogram online clinical tool to estimate the likelihood of myocardial fibrosis based on clinical parameters and T‐wave data (Jalali et al., 2019). The data that support the findings of this study are available from the corresponding author upon reasonable request.…”

Section: Methodsmentioning

confidence: 99%

T‐wave and its association with myocardial fibrosis on cardiovascular magnetic resonance examination

Zaręba

Truong

Mazur

et al. 2020

Noninvasive Electrocardiol

View full text Add to dashboard Cite

Background Risk stratification in non‐ischemic myocardial disease poses a challenge. While cardiovascular magnetic resonance (CMR) is a comprehensive tool, the electrocardiogram (ECG) provides quick impactful clinical information. Studying the relationships between CMR and ECG can provide much‐needed risk stratification. We evaluated the electrocardiographic signature of myocardial fibrosis defined as presence of late gadolinium enhancement (LGE) or extracellular volume fraction (ECV) ≥29%. Methods We evaluated 240 consecutive patients (51% female, 47.1 ± 16.6 years) referred for a clinical CMR who underwent 12‐lead ECGs within 90 days. ECG parameters studied to determine association with myocardial fibrosis included heart rate, QRS amplitude/duration, T‐wave amplitude, corrected QT and QT peak, and Tpeak‐Tend. Abnormal T‐wave was defined as low T‐wave amplitude ≤200 µV or a negative T wave, both in leads II and V5. Results Of the 147 (61.3%) patients with myocardial fibrosis, 67 (28.2%) had ECV ≥ 29%, and 132 (54.6%) had non‐ischemic LGE. An abnormal T‐wave was more prevalent in patients with versus without myocardial fibrosis (66% versus 42%, p < .001). Multivariable analysis demonstrated that abnormal T‐wave (OR 1.95, 95% CI 1.09–3.49, p = .03) was associated with myocardial fibrosis (ECV ≥ 29% or LGE) after adjustment for clinical covariates (age, gender, history of hypertension, and heart failure). Dynamic nomogram for predicting myocardial fibrosis using clinical parameters and the T‐wave was developed: https://normogram.shinyapps.io/CMR_Fibrosis/. Conclusion Low T‐wave amplitude ≤ 200 µV or negative T‐waves are independently associated with myocardial fibrosis. Prospective evaluation of T‐wave amplitude may identify patients with a high probability of myocardial fibrosis and guide further indication for CMR.

show abstract

Visualising statistical models using dynamic nomograms

Cited by 103 publications

References 29 publications

A Simple-to-Use Web-Based Calculator for Survival Prediction in Acute Respiratory Distress Syndrome

A Simple-to-Use Web-Based Calculator for Survival Prediction in Acute Respiratory Distress Syndrome

A prognostic model for overall survival in sporadic Creutzfeldt‐Jakob disease

T‐wave and its association with myocardial fibrosis on cardiovascular magnetic resonance examination

Contact Info

Product

Resources

About