scicalc

scicalc provides common calculations for PK/PD dataset preparation and analysis, including renal function, body measurements, organ function classification, and reproducible data I/O with hash verification.

Installation

You can install the development version of scicalc from GitHub with:

pak::pkg_install("a2-ai/scicalc")

Features

Renal Function

• eGFR calculation with multiple methods: CKD-EPI 2009, CKD-EPI 2021, CKD-EPI 2021 with cystatin C, MDRD, and Schwartz (pediatric)
• Absolute eGFR conversion from relative (mL/min/1.73m²) to absolute (mL/min)
• Creatinine clearance using Cockcroft-Gault equation
• Renal function classification supporting both regulatory (FDA) and clinical (KDIGO) categories

Body Measurements

• BMI calculation and categorization (underweight, normal, overweight, obese classes)
• Body surface area using DuBois or Mosteller methods
• Ideal body weight and adjusted ideal body weight

Organ Function Classification

• Hepatic function classification using NCI-ODWG criteria (bilirubin and AST-based)
• Renal function classification with regulatory and clinical standards

Age Classification

• Age categories: neonate, infant, child, adolescent, adult, elderly

Data Helpers

• Demographic conversions: is_female(), is_black(), is_asian(), etc. for use in clinical equations
• Unit conversions: creatinine (mg/dL to µmol/L), bilirubin, albumin
• Geometric statistics: geometric mean, SD, and CV

Reproducible Data I/O

• Read/write with hash verification: Automatically compute and display file hashes when reading or writing data
• Multiple formats: CSV, Parquet, SAS (sas7bdat), Excel, XPT, PZFX
• Hash validation: Read files only if they match an expected hash

Quick Example

library(scicalc)
library(dplyr)

# Sample patient data
patients <- data.frame(
  ID = 1:4,
  SEX = c("Female", "Male", "Female", "Male"),
  RACE = c("White", "Black", "Asian", "White"),
  AGE = c(45, 62, 38, 71),
  WEIGHT = c(68, 85, 55, 78),
  HEIGHT = c(165, 178, 158, 172),
  CREAT = c(0.9, 1.4, 0.7, 1.8)
)

# Calculate derived values
patients <- patients %>%
  mutate(
    BSA = bsa(WEIGHT, HEIGHT),
    EGFR = egfr(is_female(SEX), is_black(RACE), AGE, CREAT),
    AEGFR = aegfr(EGFR, BSA),
    RFC = rfc(AEGFR)
  )

patients
#>   ID    SEX  RACE AGE WEIGHT HEIGHT CREAT      BSA      EGFR     AEGFR RFC
#> 1  1 Female White  45     68    165   0.9 1.749277  80.34339  81.23862   2
#> 2  2   Male Black  62     85    178   1.4 2.032001  56.82811  66.74842   2
#> 3  3 Female Asian  38     55    158   0.7 1.548984 113.45681 101.58541   1
#> 4  4   Male White  71     78    172   1.8 1.911017  39.74485  43.90352   3

Documentation

See the package website for full documentation.