Biostats — AI-Powered Biostatistics Platform

Clinical trial design, survival analysis, and statistical inference powered by AI. Built for biopharma teams who need rigorous, publication-ready results — fast.

🎯 What is Biostats?

Biostats is an open-source biostatistics platform that combines classical statistical methods with AI-powered interpretation. It provides:

• Survival Analysis — Kaplan-Meier, Cox PH, competing risks
• Clinical Trial Design — Sample size, power analysis, adaptive designs
• Bayesian Methods — Posterior estimation, Bayesian adaptive trials
• AI Interpretation — Plain-language explanations of statistical results
• Publication Engine — Auto-generate publication-ready tables and figures

🚀 Quick Start

pip install biostats

# Survival analysis
from biostats import Survival
km = Survival.kaplan_meier(time=[3,6,9,12], event=[1,0,1,1])
km.plot()
km.summary()  # AI-generated plain-language interpretation

# Sample size calculation
from biostats import TrialDesign
design = TrialDesign.two_arm(
    effect_size=0.3, alpha=0.05, power=0.80
)
print(design.sample_size)  # → 176 per arm

# Bayesian analysis
from biostats import Bayesian
posterior = Bayesian.bernoulli(
    successes=45, trials=100, prior_beta=(1, 1)
)
posterior.credible_interval(0.95)  # → (0.353, 0.549)

📊 Core Modules

Module	Description	Status
biostats.survival	KM, Cox PH, Fine-Gray, IPCW	✅ Stable
biostats.trial	Sample size, power, adaptive designs	✅ Stable
biostats.bayesian	MCMC, conjugate priors, model comparison	✅ Stable
biostats.regression	Linear, logistic, Poisson, mixed models	🔨 Beta
biostats.meta	Fixed/random effects meta-analysis	🔨 Beta
biostats.interpret	AI-powered result interpretation	✅ Stable
biostats.publish	Publication-ready tables & figures	🔨 Beta

🧬 Use Cases

• Phase II/III Clinical Trials — Design, interim analysis, sample size re-estimation
• Real-World Evidence — Propensity score matching, IPTW, instrumental variables
• Meta-Analysis — Forest plots, heterogeneity assessment, publication bias
• Health Economics — Cost-effectiveness analysis, QALY estimation
• Epidemiology — Incidence rates, standardized mortality ratios, time-series

📖 Documentation

Full documentation: biostats.readthedocs.io

🤝 Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

📄 License

MIT License — see LICENSE for details.

---

Built with ❤️ by MoKangMedical