DICOM Window Level Endpoint Tool

Description

Compute DICOM default LINEAR window thresholds c−0.5±(w−1)/2; not LINEAR_EXACT or SIGMOID.

An image grid has array indices as well as physical spacing, timing, and signal conventions. A calculation on supplied metadata describes one part of that grid or signal; it does not inspect an image, validate a complete file, or interpret anatomy.

Compute DICOM default LINEAR window thresholds c−0.5±(w−1)/2; not LINEAR_EXACT or SIGMOID.

When to use this tool

Use it for dicom window level endpoint when the required inputs and conventions are already available.

How to use the tool and how it works

Enter center, width. Use the stated units and boundary conventions.
Run the calculation. Correct any input error before interpreting a result.
Read lower, upper. Inspect any known, defined, inclusion, or boundary flags before interpreting numeric values.

Require width at least one. Return lower=center-0.5-(width-1)/2 and upper=center-0.5+(width-1)/2, the endpoint terms for the DICOM default LINEAR window convention.^[1]

DICOM default LINEAR windowing requires width at least one; width one is a threshold case.^[2]

Input field	Meaning and units	Accepted input
center	Center	Required. Finite number.
width	Width	Required. Finite number; minimum 1.

Output field	Interpretation	Returned shape
lower	Lower	Required. Finite number.
upper	Upper	Required. Finite number.

json

{
  "input": {
    "center": 50,
    "width": 11
  },
  "output": {
    "lower": 44.5,
    "upper": 54.5
  }
}

Limitations and assumptions

At width one both endpoints coincide at center minus one half. The tool computes endpoint terms only; it does not apply a pixel display mapping or implement LINEAR_EXACT or SIGMOID.
The field tables describe structural bounds; the mechanics also state cross-field requirements. Malformed input shapes, disallowed keys, and invalid domains cause input errors. Results must remain finite and within the declared output contract. A successful false flag or zero placeholder is distinct from an error.
Numeric arithmetic uses binary floating point. Exact comparisons have no hidden tolerance or decimal rounding; negative zero is exported as zero. Decimal-text integer operations retain the exactness specifically stated above.

Alternative or Complementary methods

For dicom window level endpoint, check the source records and the supplied rule before comparing the result with another method. Use a full image reader and geometry or signal analysis when the task requires interpreting image data.

References

Calculation and input contract: radiology/dicom-window-level-endpoint — tool.io
DICOM PS3.3, C.11.2: VOI LUT Module — NEMA

Similar or alternative tools

DICOM Window Center Width Domain Bound
DICOM default LINEAR VOI boundaries c-0.5 plus/minus (w-1)/2; width one is a threshold and this tool does not implement LINEAR_EXACT or SIGMOID.
DICOM Frame Increment Time Index Term
Constant Frame Time coordinate for one zero-based frame index; not a variable Frame Time Vector or acquisition datetime parser.