Medical Physics: Imaging, Radiation, and Treatment

Medical physics sits at the intersection of fundamental science and direct patient care — where the behavior of photons, protons, and magnetic fields gets translated into diagnoses and treatment plans. This page covers how medical physicists apply the principles of radiation, electromagnetism, and nuclear science to clinical imaging and cancer therapy, what distinguishes different modalities, and where the boundaries of each approach fall.

Definition and scope

A linear accelerator producing 6-megavolt X-rays. A superconducting magnet operating at 1.5 or 3 tesla. A positron-emitting isotope with a half-life of 110 minutes. These aren't abstractions from a physics lecture — they're the daily instruments of clinical medical physics, a field formally recognized by the American Board of Radiology (ABR) and the American Board of Medical Physics (ABMP) as a credentialed specialty.

Medical physics spans four primary subfields: diagnostic imaging physics (X-ray, CT, fluoroscopy), nuclear medicine physics (PET, SPECT), radiation oncology physics (linear accelerators, brachytherapy), and MRI physics. Practitioners calibrate equipment, design radiation treatment plans, verify dose delivery, and ensure regulatory compliance under standards set by the Nuclear Regulatory Commission (NRC) and the International Atomic Energy Agency (IAEA).

The physical principles underlying these applications — wave-particle duality, radioactive decay, electromagnetic induction — are part of the broader conceptual framework described across physics as a discipline.

How it works

The core mechanism differs by modality, but the common thread is controlled energy deposition: delivering enough energy to reveal or destroy a target, without collateral damage that outweighs the benefit.

X-ray and CT imaging work through differential attenuation. Dense tissue (bone) absorbs more photons; soft tissue absorbs fewer. A detector array records the pattern. In computed tomography, the X-ray source rotates around the patient while detectors capture projections from hundreds of angles; reconstruction algorithms (typically filtered back-projection or iterative reconstruction) reassemble those projections into a 3D volume. A modern 64-slice CT scanner completes a full chest acquisition in under 5 seconds.

MRI works through nuclear magnetic resonance. Hydrogen nuclei (protons) in water molecules precess when placed in a strong magnetic field. A radiofrequency pulse tips those protons out of alignment; when the pulse stops, they release energy as they realign. Gradient coils encode spatial position by varying the field across three axes. No ionizing radiation is involved — a distinction with real clinical significance for pediatric and repeat imaging.

Nuclear medicine inverts the geometry. Instead of an external beam passing through the patient, a radiotracer — typically a molecule labeled with a gamma- or positron-emitting isotope — is administered internally. PET scanners detect the paired 511 keV photons produced when a positron annihilates with an electron. FDG (fluorodeoxyglucose), labeled with fluorine-18, accumulates in metabolically active tissue, making it useful for oncologic staging.

Radiation therapy uses ionizing radiation to damage the DNA of tumor cells beyond their repair capacity. The standard delivery platform is the linear accelerator (LINAC), which accelerates electrons to near-light speed and converts them into therapeutic X-ray beams. Intensity-modulated radiation therapy (IMRT) shapes dose distribution using computer-controlled multileaf collimators, allowing dose to conform tightly to tumor volumes while sparing adjacent organs. Proton therapy achieves even sharper dose falloff through the Bragg peak — the point at which protons deposit maximum energy before stopping, leaving minimal exit dose. For a deeper look at how energy transfer and wave behavior ground these applications, how science builds explanatory models provides useful context.

Common scenarios

Medical physics appears across clinical settings in ways that aren't always visible:

1. Treatment planning verification — Before a patient receives radiation, a medical physicist performs independent dose calculations and phantom measurements to confirm the treatment plan matches the delivered dose within tolerance, typically ±3% (AAPM Task Group 40).
2. CT dose optimization — Radiologists and physicists work together to set protocol parameters (kVp, mAs, pitch) that minimize patient dose while maintaining diagnostic image quality, guided by ALARA (As Low As Reasonably Achievable) principles.
3. Brachytherapy source calibration — Implanted radioactive seeds (iodine-125 for prostate cancer, iridium-192 for cervical cancer) require precise activity measurements before placement, with traceability to NIST standards.
4. MRI safety screening — Metal implant compatibility is assessed using the classification system established by ASTM International (F2503), distinguishing MR Safe, MR Conditional, and MR Unsafe devices.
5. Radiation protection surveys — Following equipment installation or repair, physicists measure leakage radiation to verify shielding adequacy under NRC or state regulatory requirements.

Decision boundaries

The choice between imaging modalities isn't arbitrary — it follows well-defined clinical and physical logic.

CT versus MRI comes down to speed, availability, and tissue contrast. CT resolves bone and acute hemorrhage faster and more reliably; MRI produces superior soft-tissue contrast, particularly for brain, spine, and joint pathology, but requires 20–60 minutes per scan and cannot be used safely near ferromagnetic implants or in patients with certain pacemakers.

Proton therapy versus IMRT is largely a question of anatomy and age. Protons offer meaningful dose-sparing advantages for tumors adjacent to critical structures — brainstem, spinal cord, optic apparatus — and for pediatric patients, where minimizing integral dose reduces the risk of secondary malignancy over decades. The National Cancer Institute notes that proton facilities remain less widely available than photon-based systems, and clinical evidence comparing outcomes is still accumulating for several tumor sites.

PET versus SPECT turns on sensitivity and resolution. PET's coincidence detection produces higher spatial resolution (4–6 mm) compared to SPECT's collimated single-photon approach (8–12 mm), but SPECT isotopes have longer half-lives, making them practical for procedures requiring delayed imaging.