"Should this patient get a CT or an MRI?" It's the most common imaging decision in modern medicine, and the answer is rarely straightforward. Both are cross-sectional modalities that revolutionized diagnostic imaging — but they work on completely different physical principles and excel at different things.
CT = fast, great for bone, acute bleeding, and trauma. Uses ionizing radiation.
MRI = detailed soft tissue contrast, no radiation. Slower, more expensive, more contraindications.
A CT scanner rotates an X-ray tube and detector array around the patient, acquiring hundreds of projections from different angles. A computer reconstructs these into cross-sectional (axial) slices using filtered back projection or iterative reconstruction. Each voxel in the image represents a Hounsfield unit — a measure of X-ray attenuation relative to water (0 HU). Air = -1000 HU, water = 0 HU, bone = +400 to +1000 HU, metal = +2000+ HU.
Key traits: Excellent spatial resolution, fast acquisition (seconds), excellent for bone and acute hemorrhage. Exposes the patient to ionizing radiation (effective doses typically 2-10 mSv for a CT, compared to 0.1 mSv for a chest X-ray).
MRI uses powerful magnetic fields and radio waves — no ionizing radiation involved. (typically 1.5T or 3T) and radiofrequency pulses to excite hydrogen protons in the body. When the RF pulse stops, protons relax back to their equilibrium state, emitting signals that are spatially encoded by gradient coils and reconstructed into images. The key parameters — T1 relaxation, T2 relaxation, proton density, and diffusion — produce different tissue contrasts.
Key traits: Superior soft tissue contrast, no ionizing radiation, multiplanar capability without repositioning the patient. But: slow (20-60+ minutes), expensive, loud, and contraindicated for many implants.
| Clinical Question | Best Modality | Why |
|---|---|---|
| Acute head trauma | CT | Fast, excellent for skull fractures and acute hemorrhage (blood is hyperdense on CT) |
| Suspected stroke (acute) | CT | CT perfusion + CT angiography rule out hemorrhage and show salvageable tissue. "Time is brain." |
| Brain tumor characterization | MRI | Superior soft tissue contrast, gadolinium enhancement patterns, MR spectroscopy |
| Multiple sclerosis | MRI | FLAIR sequences show demyelinating plaques exquisitely. CT is nearly useless for MS. |
| Spinal trauma / fracture | CT | Bone detail is superior. CT is the first-line for C-spine clearance in trauma. |
| Spinal cord compression | MRI | Visualizes the cord, disc herniation, epidural abscess, and metastases. CT myelography is an alternative. |
| Pulmonary embolism | CT | CT pulmonary angiography (CTPA) is the gold standard. Fast, widely available. |
| Liver lesion characterization | MRI | Multi-phase contrast MRI with hepatobiliary agents is superior. CT is a good screening tool. |
| Kidney stones | CT | Non-contrast CT (KUB CT) has >95% sensitivity. Stones are radiopaque. MRI can't see stones well. |
| Knee internal derangement | MRI | Meniscal tears, ACL/PCL injuries, and cartilage defects. CT arthrography is a second-line option. |
| Shoulder rotator cuff | MRI | Excellent soft tissue contrast. Ultrasound is also good. CT is not appropriate. |
In an ideal world, every patient gets the optimal modality. In the real world:
This is why CT is often the "good enough" choice in emergency settings. An MRI might be the better test, but if the patient needs an answer now, CT wins.
CT and MRI are complementary, not competitive. A trauma patient with a suspected epidural hematoma goes to CT first — speed saves lives. That same patient, a week later with persistent neurological deficits, goes to MRI to characterize the underlying brain injury. Understanding when to use each is what separates a technician from a technologist.