Role Definition
| Field | Value |
|---|---|
| Job Title | Radiation Therapist |
| Seniority Level | Mid-Level (3-7 years) |
| Primary Function | Administers prescribed radiation treatments to cancer patients. Positions and immobilizes patients on treatment tables using custom devices (masks, molds), operates linear accelerators (linacs) to deliver precise radiation doses, performs image-guided verification before each fraction, supports treatment planning by contouring organs at risk, conducts daily machine quality assurance, monitors patients for side effects across multi-week courses, and provides emotional support throughout treatment. Works in hospital radiation oncology departments and freestanding cancer centres. |
| What This Role Is NOT | Not a Radiation Oncologist (physician who prescribes treatment and bears ultimate clinical accountability). Not a Medical Physicist/Dosimetrist (who designs treatment plans and calculates dose distributions). Not a Radiologic Technologist (who performs diagnostic imaging, not therapeutic radiation). Not a Nuclear Medicine Technologist (different modality entirely). |
| Typical Experience | 3-7 years. Bachelor's degree in radiation therapy (associate's accepted in some states). ARRT(T) certification required. State licensure required in most states. CPR/BLS certified. Median salary $101,990 (BLS, May 2024). ~19,200 employed nationally. |
Seniority note: Entry-level radiation therapists perform the same hands-on tasks under closer supervision and would score similarly. Senior/lead therapists with supervisory, training, and protocol development responsibilities would score slightly higher Green.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 2 | Daily hands-on patient positioning on treatment tables, applying immobilization devices (thermoplastic masks, vacuum bags), aligning patients with lasers and imaging systems. Physical work is significant but occurs in a structured clinical environment with standardised equipment — not the unstructured variability of a skilled trade. |
| Deep Interpersonal Connection | 2 | Radiation therapists see the same patients daily for 4-8 weeks of treatment. They develop genuine therapeutic relationships — calming anxious patients, monitoring emotional wellbeing, providing reassurance during an inherently frightening process. Trust is core to the patient experience. |
| Goal-Setting & Moral Judgment | 1 | Exercises clinical judgment on patient setup accuracy, when to halt treatment for safety, and when to escalate concerns to the radiation oncologist. Operates within physician-prescribed protocols rather than setting treatment direction. |
| Protective Total | 5/9 | |
| AI Growth Correlation | 0 | Demand driven by cancer incidence (aging population) and advances in radiation therapy indications. AI adoption neither creates nor destroys demand for radiation therapists — it changes how they work, not whether they are needed. |
Quick screen result: Protective 5/9 with neutral correlation — likely Green or upper Yellow. Proceed to task analysis.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| Patient positioning & immobilization (setup, align with lasers/CBCT, apply immobilization devices) | 25% | 1 | 0.25 | NOT INVOLVED | Entirely physical — every patient has different anatomy, body habitus, and comfort needs. Custom immobilization, manual adjustments, and real-time patient interaction. No robotic pathway viable for the dexterity and patient trust required. |
| Treatment delivery (operate linac, monitor treatment, manage beam-on) | 25% | 2 | 0.50 | AUGMENTATION | AI-integrated linacs (Varian TrueBeam, Elekta Unity) automate beam sequencing and motion management. The therapist operates the machine, monitors the patient via camera, pauses for patient movement, and manages real-time safety. Human remains in the loop for every fraction. |
| Patient assessment & monitoring (track side effects, assess skin reactions, manage comfort, report to oncologist) | 15% | 2 | 0.30 | AUGMENTATION | AI symptom-tracking tools and predictive toxicity models are emerging. The therapist physically assesses the patient daily — inspecting skin reactions, evaluating pain, and deciding when to escalate. Clinical observation and patient communication remain human-led. |
| Treatment planning support (contouring OARs, plan verification, adaptive re-planning) | 15% | 3 | 0.45 | AUGMENTATION | AI auto-contouring (Varian, RayStation, Limbus AI, MVision) dramatically reduces manual contouring time. The therapist reviews and edits AI-generated contours rather than drawing them from scratch. Human oversight remains mandatory — AI generates, human validates. |
| Quality assurance (daily machine QA, dosimetry spot checks, safety interlocks) | 10% | 3 | 0.30 | AUGMENTATION | AI-assisted QA tools automate some dosimetry checks and log analysis. Physical machine checks (laser alignment, mechanical tests, safety interlocks) and professional judgment on pass/fail criteria remain human responsibilities. |
| Documentation & EHR (treatment records, billing, regulatory compliance) | 5% | 4 | 0.20 | DISPLACEMENT | Record-and-verify systems (ARIA, MOSAIQ) increasingly automate treatment logging. AI handles much of the documentation workflow. Human reviews and signs off. |
| Patient education & emotional support (explain treatment, manage anxiety, family communication) | 5% | 1 | 0.05 | NOT INVOLVED | Patients undergoing weeks of radiation therapy need genuine human support. Explaining side effects, managing treatment fatigue, and providing emotional continuity across a 6-8 week course is irreducibly human work. |
| Total | 100% | 2.05 |
Task Resistance Score: 6.00 - 2.05 = 3.95/5.0
Displacement/Augmentation split: 5% displacement, 65% augmentation, 30% not involved.
Reinstatement check (Acemoglu): AI creates new tasks — reviewing and editing AI-generated contours, interpreting adaptive re-planning recommendations, validating AI-suggested treatment modifications, and managing AI QA analytics. The role is shifting from manual drafting to AI-output validation, which requires deeper clinical understanding, not less.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | 0 | BLS projects 2% growth 2024-2034, slower than the 3% average for all occupations. About 900 openings per year, mostly replacement. Small occupation (19,200) with stable but not growing demand. |
| Company Actions | 0 | No hospitals or cancer centres cutting radiation therapist positions citing AI. No expansion signals either. AI auto-contouring adopted widely but affects workflow efficiency, not headcount. ASRT convened a consensus committee on staffing shortages — not layoffs. |
| Wage Trends | 1 | Median $101,990 (May 2024) — well above the $49,500 national median. Strong compensation reflecting specialised skills and licensing requirements. Wages stable to modestly growing. |
| AI Tool Maturity | 0 | AI auto-contouring (Varian, RayStation DL segmentation, Limbus AI, MVision) is in production and reduces contouring time significantly. AI-assisted treatment planning creates plans within minutes. However, these tools augment the therapist — none operate linacs or position patients autonomously. Net effect is workflow transformation, not displacement. |
| Expert Consensus | 1 | Academic consensus: role transforms rather than disappears. PMC (2024): "automation will not just remove tasks but create new opportunities." 96% of radiation therapists acknowledge AI will affect their role, but expert literature unanimously frames this as augmentation. ASTRO and ASRT position AI as complementary to human expertise. |
| Total | 2 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 2 | ARRT(T) certification mandatory. State licensure required in most states. Accredited degree programme prerequisite. No regulatory pathway for AI systems to independently deliver radiation therapy. FDA/NRC regulations mandate human oversight of therapeutic radiation. |
| Physical Presence | 2 | Must physically be with the patient — positioning on the treatment table, applying immobilization devices, aligning with imaging systems, and responding to patient distress or movement during treatment. On-site, hands-on, every fraction. |
| Union/Collective Bargaining | 0 | Minimal union representation among radiation therapists. No collective bargaining protections specific to the role. |
| Liability/Accountability | 2 | Radiation therapy errors can cause severe patient harm (radiation burns, damage to healthy tissue, death in extreme cases). Someone must bear personal liability for correct patient setup, dose delivery, and safety. Criminal prosecution has occurred for radiation therapy errors. AI has no legal personhood. |
| Cultural/Ethical | 1 | Cancer patients undergoing radiation therapy are vulnerable and expect human caregivers. Moderate cultural resistance to autonomous radiation delivery. Not as visceral as surgery or end-of-life care, but patients and families want a human controlling the machine that delivers radiation. |
| Total | 7/10 |
AI Growth Correlation Check
Confirmed 0 (Neutral). AI adoption in radiation oncology does not create or destroy demand for radiation therapists. Demand is driven by cancer incidence rates, aging population demographics, and expanding indications for radiotherapy (e.g., hypofractionation, SBRT, proton therapy). AI auto-contouring and automated planning make existing therapists more efficient but do not eliminate the need for the human who positions the patient and operates the machine. This is Green (Transforming) — not Accelerated, because there is no recursive AI dependency.
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 3.95/5.0 |
| Evidence Modifier | 1.0 + (2 × 0.04) = 1.08 |
| Barrier Modifier | 1.0 + (7 × 0.02) = 1.14 |
| Growth Modifier | 1.0 + (0 × 0.05) = 1.00 |
Raw: 3.95 × 1.08 × 1.14 × 1.00 = 4.8632
JobZone Score: (4.8632 - 0.54) / 7.93 × 100 = 54.5/100
Zone: GREEN (Green ≥48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 30% |
| AI Growth Correlation | 0 |
| Sub-label | Green (Transforming) — ≥20% task time at 3+, Growth Correlation ≠ 2 |
Assessor override: None — formula score accepted.
Assessor Commentary
Score vs Reality Check
The 54.5 AIJRI score sits 6.5 points above the Green Zone boundary — moderately above but not borderline. The score aligns well with Radiologic Technologist (56.5), which shares the same physical + imaging + licensing profile but faces less AI-driven workflow change in treatment planning. The slightly lower score reflects that radiation therapists spend 30% of task time on functions scoring 3+ (contouring, QA) where AI is actively transforming workflows — more than the 25% for radiologic technologists. The assessment is not barrier-dependent: removing barriers entirely would still yield a 3.95 task resistance with positive evidence, keeping the role in upper Yellow or low Green.
What the Numbers Don't Capture
- Hypofractionation trend. Advanced techniques (SBRT, SRS) deliver higher doses in fewer fractions, reducing the total number of treatment sessions. This could suppress headcount growth even as cancer incidence rises — the BLS 2% growth projection already reflects this compression.
- Treatment planning role expansion vs contraction. AI auto-contouring removes the manual drafting work but creates new validation and adaptive re-planning tasks. Whether this is net-positive or net-negative for the therapist role depends on departmental structure — some centres may reassign planning tasks to dosimetrists while therapists focus purely on delivery.
- Small occupation size. With only 19,200 workers nationally, small changes in cancer centre operations have outsized effects on the occupation. The BLS projecting only 900 annual openings means the labour market is thin.
Who Should Worry (and Who Shouldn't)
Radiation therapists who are strong in hands-on patient care — patient positioning, immobilization, bedside manner, and real-time clinical assessment during treatment — are deeply protected. The physical, interpersonal core of the role is untouchable by current AI. Radiation therapists who have specialised primarily in contouring and treatment planning support should pay close attention — AI auto-contouring tools are already reducing the time required for these tasks by 50-90%, and departments may consolidate this work. The single biggest separator is whether your daily value comes from being at the treatment machine with patients or from sitting at a planning workstation. If you are at the linac, you are safe. If you are primarily at the planning terminal, your workflow is being compressed.
What This Means
The role in 2028: Radiation therapists will use AI auto-contouring to review and edit AI-generated structures rather than drawing them manually. Treatment planning will be increasingly automated, with therapists validating AI-generated plans rather than creating them from scratch. The core job — positioning patients, operating the linac, monitoring treatment delivery, and providing daily patient care across multi-week courses — remains entirely human. Adaptive radiotherapy will expand the therapist's role in real-time plan adjustments.
Survival strategy:
- Master the linac-side role — patient positioning, image-guided verification, motion management, and real-time clinical assessment during treatment are the most AI-resistant skills in the profession
- Learn to work with AI contouring and planning tools — become the clinician who validates and improves AI outputs rather than competing with AI on speed; facilities will value therapists who can critically evaluate AI-generated contours
- Pursue advanced technique certifications — SBRT, SRS, proton therapy, and adaptive radiotherapy specialisations anchor you in the highest-complexity settings where physical presence and clinical judgment are non-negotiable
Timeline: 5-10+ years. Driven by the fundamental requirement for physical patient contact, linac operation, and human oversight of therapeutic radiation delivery. Workflow transformation in planning tasks is happening now, but the delivery-side role has no viable automation pathway.
