Hounsfield Units in Plain Language: What CT Numbers Actually Mean
Hounsfield units form the numeric scale that appears on every CT scan. The scale measures how much an x-ray beam is weakened as it passes through tissue. Water sits at zero. Air registers near minus one thousand. These numbers let doctors compare densities on the same image.
Radiologists view CT slices on screens that translate Hounsfield units into shades of gray. Higher numbers appear brighter. Lower numbers appear darker. A fresh bleed inside the brain usually measures between fifty and eighty Hounsfield units, which places it brighter than brain tissue on non-contrast head scans.
How the Scale Was Created
Sir Godfrey Hounsfield designed the scale in the 1970s so each material would receive a reproducible number. The formula sets water at exactly zero and air at minus one thousand. All other tissues fall between or beyond these anchors. Bone often exceeds several hundred Hounsfield units, while fat sits near minus one hundred.
Why Blood Looks Bright on Head CT
Fresh blood contains dense proteins and iron that absorb x-rays more than surrounding brain tissue. On a non-contrast scan this difference produces Hounsfield values of fifty to eighty. The higher density makes the clot appear white against the gray brain. Over days the clot loses density and the brightness fades.
Window Settings Change What You See
Radiologists adjust two numbers called window width and window level. These settings decide which Hounsfield range appears as shades of gray on the monitor. A brain window might center around thirty-five Hounsfield units with a width of eighty. The same image shown in a bone window will hide the bleed but reveal skull fractures. Eyeball brightness alone can mislead without proper windows.
Common Hounsfield Values in Brain Imaging
| Tissue or Material | Typical Hounsfield Units |
|---|---|
| Air | -1000 |
| Fat | -100 to -50 |
| Water or CSF | 0 to 15 |
| Brain gray matter | 30 to 45 |
| Fresh blood clot | 50 to 80 |
| Bone | 300 to 2000 |
One Number Never Makes a Diagnosis
A Hounsfield measurement describes density only. The same value can arise from different causes. A reading of sixty might represent acute blood in one patient and a calcified mass in another. Location, shape, and clinical history always matter. The radiology team combines the numbers with visual pattern and patient story before reaching any conclusion.
ABC2-SPH Score Uses Volume, Not Density
Prognostic tools such as the ABC2-SPH mortality score estimate thirty-day risk after spontaneous intracerebral hemorrhage. The score draws on patient age, Glasgow Coma Scale, hematoma volume calculated by the ABC over two method, presence of intraventricular blood, and whether the bleed is deep or lobar. It does not rely on exact Hounsfield numbers. Volume in millilitres remains the key imaging variable. Decisions about care belong to the treating neurology and neurosurgery team.
Questions to Ask the Care Team
- What does the density of this area suggest on my scan?
- Which window setting best shows the finding?
- How does the volume of blood affect the ABC2-SPH estimate?
- Will follow-up scans use the same window settings?
- Can changes in Hounsfield units over time help track progress?
- What other imaging or lab tests will clarify the picture?
CT scanners assign a Hounsfield unit to every tiny block of tissue, called a voxel. Modern machines can distinguish differences as small as five units under ideal conditions. In daily practice radiologists look at patterns rather than single numbers. A bleed that measures seventy on day one may drop to forty by day seven as the blood breaks down.
Contrast dye raises the Hounsfield value of vessels and some tumors by two hundred units or more. Non-contrast scans therefore remain the first step when doctors look for fresh hemorrhage. The bright appearance of blood on these scans comes directly from its natural density, not from any added substance.
Artifacts can alter measured values. Metal implants create streaks that push nearby numbers hundreds of units higher or lower. Beam hardening inside thick bone can make deep brain tissue look artificially dense. Technologists and radiologists correct for these effects when they interpret the study.
Students learning CT often practice measuring sample regions of interest. A circular cursor placed inside a hemorrhage gives an average Hounsfield unit and a standard deviation. The deviation shows how uniform the material is. Fresh clot usually shows low variation while mixed blood and brain tissue shows higher spread.
Families sometimes see the numbers displayed on the radiology report. These figures support the written description but do not replace it. A report might state that a left basal ganglia hematoma measures sixty millilitres with Hounsfield units averaging sixty-five, consistent with acute hemorrhage. The ABC2-SPH components are then listed separately so the team can discuss prognosis openly.
Each scanner brand calibrates the scale slightly differently, yet the differences stay small for soft tissue. Daily quality checks keep the water phantom reading within two units of zero. This consistency lets doctors compare scans taken on different machines or at different hospitals.
Research continues to explore whether texture analysis of Hounsfield distributions inside a clot can predict expansion risk. These advanced methods remain investigational. Standard care still relies on the simple density difference that makes acute blood bright on non-contrast head CT.