ClinicoPath

---

Abstract

The ClinicoPath Ecosystem: A Comprehensive Open-Source Toolkit for Clinicopathological Research

Background: Clinicopathological research is fundamental to advancing evidence-based medicine, biomarker validation, and precision oncology. However, it requires complex and specialized statistical methods ranging from survival modeling and decision analysis to inter-rater reliability and rich statistical visualization. The technical barrier of programming-based statistical software can limit clinicians and pathology researchers from performing these analyses reproducibly. To address this gap, we developed ClinicoPath, an open-source umbrella toolkit built for the jamovi statistical platform and R.

Architecture: ClinicoPath coordinates 62 production analyses distributed across 5 focused, specialized submodules available directly in the jamovi library, alongside an active development and staging infrastructure:

1. ClinicoPathDescriptives (14 analyses): Baseline characteristics (Table 1), cross-tabulation with significance tests, descriptive summaries, demographic age pyramids, treatment pathway alluvial flows, Venn diagrams, variable hierarchy trees, and robust data quality/outlier screening.
2. jsurvival (9 analyses): Comprehensive time-to-event analysis, Kaplan-Meier curves with risk tables, Cox proportional hazards regression, continuous biomarker threshold detection, odds ratios, high-dimensional LASSO-Cox regularized regression, and clinical date/time interval tools.
3. meddecide (15 analyses): Diagnostic test accuracy evaluation, Fagan nomogram decision calculators, test combination/co-testing/sequential algorithms, Decision Curve Analysis (DCA), ROC curve modeling, inter-rater reliability (Cohen’s/Fleiss’ Kappa), regularized prediction (LASSO logistic), and precision/power sample size planning.
4. jjstatsplot (19 analyses): Publication-ready statistical visualizations integrating ggstatsplot and modern plotting tools into jamovi (between-group and within-subject box-violin plots, scatter plots with marginals, correlation matrices, raincloud and ridgeline distributions, waffle charts, arc networks, and automated intelligent plot selection).
5. OncoPath (5 analyses): Specialized oncology and pathology research tools including swimmer plots for patient timelines, waterfall and spider plots for tumor burden response (adapted RECIST v1.1 thresholds), diagnostic test meta-analysis for pathology & AI validation, quantitative IHC marker heterogeneity, and multi-rater pathology diagnostic agreement.

Conclusion: ClinicoPath provides a powerful, accessible, and free-to-use toolkit that empowers medical researchers to conduct sophisticated statistical analyses without requiring programming expertise. By integrating these essential functions into the intuitive jamovi graphical interface while generating reproducible R code, the module lowers barriers to rigorous biomedical data analysis, enhances transparency, and accelerates translation of clinical findings.

---

📚 Documentation & Submodule Websites

All submodule documentation is hosted on dedicated pkgdown websites:

• 🌐 ClinicoPath Umbrella: https://www.serdarbalci.com/ClinicoPathJamoviModule/
• 📊 ClinicoPathDescriptives: https://www.serdarbalci.com/ClinicoPathDescriptives/
• ⏱️ jsurvival: https://www.serdarbalci.com/jsurvival/
• 🏥 meddecide: https://www.serdarbalci.com/meddecide/
• 📈 jjstatsplot: https://www.serdarbalci.com/jjstatsplot/
• 🔬 OncoPath: https://www.serdarbalci.com/OncoPath/

---

📊 Test Data & Learning Resources

Comprehensive test datasets and downloadable .omv example analysis files:

• Test Data Catalog - Curated overview with download links and sample analysis files.
  • Featured: Kappa sample size planning (kappasizeci, kappasizefixedn, kappasizepower), Decision Curve Analysis (decisioncurve), and Waterfall plots (waterfall).
  • Access in R: vignette("test-data-catalog", package = "ClinicoPath")
• Complete Test Data Catalog - Complete inventory of example data files.
• Function Reference Guide - Complete reference across all module functions.

---

🎓 Tutorial Series

Comprehensive step-by-step tutorials for clinical and translational researchers:

• Tutorial Series Home - Learning paths for clinical trials, diagnostic pathology, and advanced modeling.
• Quick Start: Getting Started with ClinicoPath
• Clinical Trials: Table One for Baseline Characteristics
• Survival Analysis: Kaplan-Meier & Cox Regression
• Diagnostic Testing: ROC Analysis & Optimal Cutpoints
• Advanced Modeling: Decision Curve Analysis
• Reproducibility: Automated Reports & Version Control

---

💻 Installation in jamovi

Method 1: Install Submodules from the jamovi Library (Recommended)

Each specialized submodule can be installed directly inside jamovi:

1. Open jamovi (version >= 2.6).
2. Click the Modules button (+) in the top right corner.
3. Select jamovi library.
4. Search for the module name or browse categories:
   • ClinicoPathDescriptives (under Exploration)
   • jsurvival (under Survival)
   • meddecide (under meddecide)
   • jjstatsplot (under jjstatsplot)
   • OncoPath (under OncoPath)
5. Click Install.

Method 2: Sideload .jmo Package Files

Pre-compiled .jmo files are available on GitHub Releases:

1. Download the .jmo file for your platform from ClinicoPath Releases.
2. In jamovi, click Modules (+) → Sideload (folder icon).
3. Select the downloaded .jmo file.

---

💻 Installation in R

You can install the development version of the umbrella package or individual submodules from GitHub:

# Install remotes if needed
if (!requireNamespace("remotes", quietly = TRUE)) install.packages("remotes")

# Umbrella package (contains all functions)
remotes::install_github("sbalci/ClinicoPathJamoviModule")

# Or individual submodules
remotes::install_github("sbalci/ClinicoPathDescriptives")
remotes::install_github("sbalci/jsurvival")
remotes::install_github("sbalci/meddecide")
remotes::install_github("sbalci/jjstatsplot")
remotes::install_github("sbalci/OncoPath")

---

🔬 Feature Overview by Submodule

1. ClinicoPathDescriptives (14 Analyses)

Menu: Exploration

• Table One (tableone): Publication-ready baseline patient summary tables with automatic variable type detection, significance tests, standardized mean differences (SMD), and missing value reporting.
• Cross Tables (crosstable): Multi-way contingency tables with Pearson Chi-Square, Fisher’s exact test, and multiple comparison corrections.
• Continuous Summaries (summarydata): Automated descriptive statistics for continuous variables with natural language interpretations.
• Categorical Summaries (reportcat): Frequency distributions and percentages for categorical variables.
• Age Pyramid (agepyramid): Demographic population pyramid plots split by gender or disease subgroups.
• Alluvial Diagrams (alluvial): Categorical flow diagrams visualizing patient therapy transitions and clinical trajectories.
• Venn Diagrams (venn): Set relationship diagrams supporting 2 to 7 sets with statistical overlap counts.
• Variable Tree (vartree): Hierarchical tree visualizations for cohort stratification and inclusion/exclusion pathways.
• Data Quality Assessment (dataquality): Multi-variable data health dashboard summarizing missingness patterns, variable types, and distributions.
• Single Variable Check (checkdata): Interactive data validation tool for screening individual variables.
• Outlier Detection (outlierdetection): Multi-method outlier detection leveraging IQR, Z-scores, robust covariance, and DBSCAN.
• Benford Analysis (benford): Digital data integrity screening using first-digit Benford distribution conformity.
• Chi-Square Post-Hoc (chisqposttest): Pairwise post-hoc proportion comparisons following significant chi-square tests.
• Categorize Variables (categorize): Binning and recoding of continuous variables into clinically meaningful ordinal categories.

---

2. jsurvival (9 Analyses)

Menu: Survival

• Survival Analysis (survival): Kaplan-Meier curves, log-rank tests, Cox proportional hazards regression, median survival with CIs, 1-, 3-, and 5-year survival rates, and risk tables.
• Single Arm Survival (singlearm): Cohort survival analysis for single-arm clinical trials or registry cohorts without an explanatory group.
• Multivariable Survival (multisurvival): Multivariable Cox proportional hazards modeling with covariate adjustment, hazard ratio forest plots, and adjusted survival curves.
• Continuous Survival (survivalcont): Survival analysis for continuous variables with optimal cut-point detection (maxstat) and quantile splits.
• Odds Ratio Analysis (oddsratio): Binary outcome evaluation with 2x2 contingency tables, odds ratio calculations, and forest plots.
• LASSO-Cox Regression (lassocox): L1-penalized Cox regression via glmnet for high-dimensional feature selection, cross-validation tuning, and coefficient path plots.
• Time Interval Calculator (timeinterval): Follow-up and survival duration calculation from clinical dates with date order validation.
• DateTime Converter (datetimeconverter): Parsing, standardization, and conversion of clinical timestamps into analysis-ready formats.
• Outcome Organizer (outcomeorganizer): Clinical endpoint mapping and standardization for time-to-event indicators (OS, DFS, PFS).

---

3. meddecide (15 Analyses)

Menu: meddecide

• Medical Decision (decision): Diagnostic test accuracy metrics: Sensitivity, Specificity, PPV, NPV, Positive/Negative Likelihood Ratios, DOR, and accuracy with 95% CIs.
• Decision Calculator (decisioncalculator): Interactive clinical calculator for pre- and post-test probabilities with Fagan nomograms.
• Compare Tests (decisioncompare): Direct statistical comparison of two or more diagnostic tests against a reference standard.
• Combine Tests (decisioncombine): Combinatorial evaluation of 2 to 3 diagnostic tests to find optimal panel algorithms; includes decision heatmaps.
• Co-Testing Analysis (cotest): Simultaneous (parallel) testing strategies to quantify sensitivity gains and specificity trade-offs.
• Sequential Testing (sequentialtests): Two-stage serial testing algorithms (screening followed by confirmatory testing).
• No Gold Standard (nogoldstandard): Diagnostic accuracy estimation when reference standards are imperfect, using latent class analysis and Bayesian Hui-Walter estimation.
• Decision Curve Analysis (DCA) (decisioncurve): Clinical net benefit evaluation across threshold probabilities, comparing “treat all”, “treat none”, and model-guided strategies; calculates interventions avoided.
• Clinical ROC Analysis (enhancedROC): Publication-ready ROC curves, empirical and smooth AUC with DeLong/bootstrap CIs, and optimal cutoff detection.
• Advanced ROC Analysis (psychopdaROC): In-depth ROC coordinate evaluation with threshold tables and cost-weighted cutoff optimization.
• Interrater Reliability (agreement): Cohen’s Kappa, Fleiss’ Kappa, weighted kappa for ordinal scales, Gwet’s AC1, and percentage agreement.
• LASSO Logistic Regression (lassologistic): L1-penalized logistic regression for regularized binary outcome prediction and sparse biomarker selection.
• Kappa Sample Size (CI) (kappaSizeCI): Precision-based sample size calculator determining required subjects for a target confidence interval half-width.
• Kappa Fixed N Analysis (kappaSizeFixedN): Calculates lowest expected Kappa and lower confidence bound for a fixed sample size.
• Kappa Power Analysis (kappaSizePower): Hypothesis testing power-based sample size calculator for inter-observer agreement.

---

4. jjstatsplot (19 Analyses)

Menu: jjstatsplot

• Histograms (jjhistostats): Distribution visualization with Shapiro-Wilk normality testing and central tendency overlays.
• Scatter Plot (jjscatterstats): Pairwise continuous association with correlation coefficients, regression fits, and marginal plots.
• Correlation Matrix (jjcorrmat): Multi-variable correlation matrices with significance markers and clustering.
• Hull Plot (hullplot): Bivariate scatter with convex polygonal hull boundaries for distinct clinical clusters.
• Between-Groups Box-Violin (jjbetweenstats): Group comparison with violin plots, boxplots, raw jittered points, ANOVA / Kruskal-Wallis, and effect sizes.
• Within-Subjects Box-Violin (jjwithinstats): Repeated measures comparison with repeated measures ANOVA or Friedman tests.
• Horizontal Dot Plot (jjdotplotstats): Horizontal box-violin comparison across categorical factors with detailed effect sizes.
• Dot Chart (jjdotchart): Cleveland-style dot charts comparing observed group summaries against reference values.
• Bar Charts (jjbarstats): Categorical frequency comparisons with Chi-square, Fisher’s exact test, and Cramer’s V.
• Pie Charts (jjpiestats): Proportion visualization with chi-square goodness-of-fit testing.
• Segmented Total Bar (jjsegmentedtotalbar): Stacked proportion bars reporting both segment-level and aggregate statistics.
• Waffle Charts (jwaffle): Square icon waffle charts for intuitive patient proportion visualization.
• Raincloud Plot (raincloud): Combined raw jittered points, boxplot summary, and half-density distribution cloud.
• Advanced Raincloud (advancedraincloud): Enhanced raincloud plot supporting longitudinal tracking and multi-group stratification.
• Ridgeline Plot (jjridges): Multi-group density ridges (joyplots) for comparing biomarker distribution shifts across stages.
• Arc Diagram (jjarcdiagram): Network arc diagrams displaying connections between pathological entities.
• Line Chart (linechart): Longitudinal trends and trajectories with error bars and confidence intervals.
• Lollipop Chart (lollipop): High-data-to-ink ratio lollipop plots for comparing ranked numerical values across categories.
• Automatic Plot Selection (statsplot2): Intelligent plotting engine that automatically inspects input variable types and renders the optimal statistical plot.

---

5. OncoPath (5 Analyses)

Menu: OncoPath

• Swimmer Plot (swimmerplot): Patient timeline visualization using enhanced ggswim; displays disease duration, clinical milestones, discrete events, adverse event flags, and response durations for oncology trial reporting.
• Treatment Response: Waterfall & Spider (waterfall): Patient-level tumor burden analysis; generates publication-ready waterfall and spider plots; measures progression against nadir; categorizes response (CR, PR, SD, PD) using adapted RECIST v1.1 thresholds; reports ORR, DCR, and person-time metrics.
• Diagnostic Test Meta-Analysis (diagnosticmeta): Meta-analysis of diagnostic accuracy studies in pathology and AI/ML algorithm validation; implements bivariate random-effects modeling (Reitsma method), proportional-hazards SROC (Holling model), meta-regression, and publication-ready forest/SROC plots.
• IHC Heterogeneity Analysis (ihcheterogeneity): Statistical analysis of immunohistochemical biomarker heterogeneity; evaluates intratumoral expression variance, multi-marker profiles, spatial distribution indices, and clinical-pathological correlates.
• Pathology Agreement (pathagreement): Multi-rater agreement analysis tailored to histopathology; computes Cohen’s Kappa, Fleiss’ Kappa, Krippendorff’s alpha, diagnostic consensus determinations (majority, super-majority), and rater concordance matrices.

---

🛠️ System Requirements

• jamovi: Version >= 2.6 (or higher)
• R: Version >= 4.1.0
• Operating Systems: macOS (Apple Silicon & Intel), Windows (64-bit), Linux

---

Acknowledgements

ClinicoPath is made possible thanks to the outstanding open-source contributions of the R and jamovi communities, including:

• jamovi developers: Jonathon Love, Ravi Selker, Damian Dropmann
• finalfit developer: Ewen Harrison
• ggstatsplot developer: Indrajeet Patil
• ggswim developers
• tangram developer: Shawn Garbett
• easystats and report developers
• tableone developer: Kazuki Yoshida
• survival developer: Terry Therneau
• survminer developer: Alboukadel Kassambara
• vtree developer: Nick Barrowman
• easyalluvial developer: Björn Oettinghaus
• mada and metafor developers
• The entire R and biostatistics community

---

Citation

If you use ClinicoPath or any of its submodules in your research or publications, please cite:

@manual{balci2026clinicopath,
  title  = {ClinicoPath: jamovi Module for Clinicopathological Research},
  author = {Serdar Balci},
  year   = {2026},
  url    = {https://www.serdarbalci.com/ClinicoPathJamoviModule/},
  doi    = {10.5281/zenodo.3997188}
}

License

GPL (>= 2) — see the LICENSE file for details.