ABR Medical Physics Boards — Exam Survival Guide

Window: Residency / CAMPEP pathway | Exams: Part 1 (General + Clinical), Part 2 (Therapeutic / Diagnostic / Nuclear), Part 3 Oral

This is the exam that decides whether you’re trusted to sit in the room where dose, image quality, and patient safety live. Treat it that way.

Must-do topics

Think in three blocks:

1) Part 1 – General + Clinical (CB exams, two days)

General medical physics foundations

Radiation interactions, dosimetry basics, attenuation, scatter, buildup.

Imaging physics: x-ray production, CT, MR basics, nuclear medicine, ultrasound.

Radiation biology + risk: dose–response models, stochastic vs deterministic effects.

Clinical physics across modalities

Radiotherapy: linac basics, beam quality, depth-dose curves, SSD/SAD, wedges, MLCs.

Diagnostic: detectors, image quality metrics (SNR, CNR, MTF, DQE), QC tests.

Nuclear: radiopharmaceuticals, detectors, counting statistics, QC, regulatory basics.

Goal: show that your physics instincts won’t hurt a patient.

2) Part 2 – Discipline-specific (Therapy / Diagnostic / Nuclear)

Therapeutic medical physics

3D-CRT, IMRT, VMAT, SBRT, SRS: planning logic, trade-offs, QA.

Patient-specific QA, commissioning philosophy, uncertainty budgets.

TG-51/TG-142 style thinking: how you prove the machine is behaving.

Diagnostic medical physics

CT protocols and dose optimisation, iterative reconstruction.

DR/Mammo/Fluoro: AEC, grids, scatter, contrast optimisation, QC frequencies.

MR safety, gradients, SAR, artifacts; ultrasound performance metrics.

Nuclear medical physics

SPECT/PET system performance; energy windows, corrections, sensitivity vs resolution.

Radiopharmacy basics, dosimetry, regulations, shielding and contamination control.

3) Part 3 – Oral (4-ish hours remote oral)

Clinical scenarios:

“Linac output off by 2%—walk through diagnosis, actions, documentation.”

“New CT protocol looks noisy—how do you evaluate and fix it?”

They’re testing: judgement, communication, and safety, not derivations.

Top traps (avoid)

Studying as if this is a pure physics exam and under-preparing the clinical judgement side.

Memorising TG numbers and forgetting the logic behind the recommendations.

Underestimating statistics/uncertainty and their impact on clinical decisions.

For Part 3: rambling into the weeds instead of giving a structured, safe answer.

Time split (macro)

Very rough prep rhythm across the track (not residency schedule, just how to think):

Part 1:

Heavy bookwork + problem-solving; lots of short derivations and sanity-check calcs.

Part 2:

60–70% case-style questions, 30–40% formulas/details; heavy use of past questions, local QA data, TG reports.

Part 3:

Mock orals > anything else; repeated drilling of “walk me through how you would…” style responses.

Exam timing itself (current structure):

Part 1: two CB sessions on different days (General ~5 h, Clinical ~3 h).

Part 2: single ~5 h CB exam, discipline-specific.

Part 3: ~4 h remote oral exam over multiple stations.

Last-48h checklist

Part 1/2 style days:

One timed mixed block (2–3 h) of past/representative questions.

Quick sweeps of:

Interaction types, beam quality metrics, common QC tests and tolerances.

“Typical numbers” you should know (order-of-magnitude dose, HVLs, half-value layers, CTDI vs DLP meaning).

No new TG reports. Only summary tables + your own crib notes.

Part 3:

5–8 mock cases the night before. For each:

State the problem.

Lay out 3–5 ordered steps (investigate → measure → decide → document → communicate).

End with how you’d document & follow-up.

Quick frames

For every question / case, ask:

What’s the patient-level risk here? (over/under-dose, mis-positioning, missed diagnosis, unnecessary dose).

What measurements or checks actually move the needle?

What is the safest, simplest intervention that restores control?

Who else needs to know? (radiation oncologist, radiologist, techs, management, regulators).

If you answer like a clinical colleague rather than a back-room mathematician, you’re in the right zone.

Speed tactics (MCQ parts)

Kill answers that:

Violate basic safety (treating without verifying, ignoring out-of-tolerance machines).

Add complexity without improving dose or image quality in a measurable way.

Sanity-check numbers:

If a quick back-of-the-envelope calc says “this dose is clearly crazy,” trust that instinct.

If two answers are technically plausible:

Prefer the one that reduces uncertainty and adds QC instead of trusting assumptions.

Day-of mini-plan

Light warm-up: a handful of short problems (SSD/SAD calc, simple image quality question) to get your brain into “physics with units” mode.

During exam:

First pass: mark anything that would take more than ~90 seconds; move on.

Second pass: attack only the marked items; don’t re-read what you’re already sure of.

For orals:

Answer in headings out loud: “First, I’d verify X. Second, I’d measure Y. Third, I’d…”.

If you realise mid-answer that you’d change approach, say so clearly and pivot. That reads as maturity, not failure.