Our Approach
From imaging problem to sustained improvement.
OraDigit works with healthcare organizations to understand imaging performance, identify priorities, support practical improvement and measure whether change holds over time.
Medical imaging quality improvement is a continuous cycle. People, protocols, equipment, volumes and workflows change. A useful method needs to follow those changes, rather than end with an inspection or a report.
Reassess as conditions change.
For the detailed operating model, see our Quality Improvement Framework. Discover and Baseline expand Assess; Analyze expands Identify, followed by Improve, Measure and Sustain.
Each engagement begins with an agreed problem, scope and review team. Clinical and professional responsibility remains with the healthcare organization and its qualified personnel throughout the process.
01 · Assess
Start with the current state.
Before recommending a solution, establish what is happening and what the team needs to understand. Agree which service, site, scanner, protocol or workflow is in scope, who should participate and what information can appropriately be reviewed.
Depending on the problem and agreed scope, assessment may consider image-quality indicators, repeat or reject information, patient preparation, protocol consistency and acquisition practices. Operational questions may call for workflow mapping, scheduling and handoff review, delays, examination duration, throughput or scanner utilization. Quality-control findings, equipment performance indicators, readiness findings and existing corrective actions may also be relevant.
Data and document review, discussions with staff and observation where appropriate can help establish a baseline. Check what the available data represents, how it was collected and where it is incomplete before interpreting it.
Working output: a defined question, a current-state picture and a usable baseline, with information gaps recorded.
02 · Identify
Understand variation and set priorities.
A poor metric is a starting point, not an explanation. Examine where and when variation appears, which process is involved and whether the concern is isolated or recurring. Staff knowledge helps place a pattern in its clinical and operational context.
Trend review, process mapping and root-cause analysis can help identify likely contributors. Where the data supports it, compare sites, scanners, protocols or workflow steps using consistent definitions. Differences may reflect case mix, equipment, timing or recording practices; an association alone does not prove causality.
Prioritize with the organization: consider patient safety, the scale of the concern, practical constraints and what the team can control. Make uncertainty visible and identify any further information needed before selecting a change.
Working output: agreed priorities and likely contributing factors, with a rationale for what to address first.
03 · Improve
Turn findings into practical action.
Structure improvement around the agreed priorities, with named owners, an appropriate review process and a plan for checking results. OraDigit's role is to help organize and support this work within scope; organizational leadership and qualified personnel approve and own the changes.
Potential work may include protocol standardization under qualified review, patient-preparation changes, workflow redesign, quality-control follow-up, documentation, staff education or communication, and clearer escalation pathways. Equipment findings go to the responsible technical and medical physics personnel. A focused tool may be useful where information or review is the problem.
Where appropriate, use small, reviewed iterations, such as Plan–Do–Study–Act, to learn before expanding a change. Agree the expected effect, safety considerations, responsibilities and measures in advance. OraDigit does not independently dictate clinical protocols.
Working output: an organization-owned action plan with review points, responsibilities and a measurement plan.
04 · Measure
Check whether the change worked.
Implementation is not the endpoint. Compare relevant measures with the baseline using agreed definitions, denominators and review periods. The measure should answer the actual imaging question, rather than reward activity that does not address it.
Depending on the project, measures may include repeat or reject rate, image-quality indicators, protocol consistency, process or turnaround time, delays, utilization, throughput, corrective-action closure or readiness status. Record changes in volume, case mix, equipment or data collection that could affect the comparison.
Review whether performance changed, whether the difference matters to the service and whether there were unintended effects. For example, a shorter process time needs to be considered alongside preparation and image quality. Results may justify continuing, adapting or stopping a change; improvement is not guaranteed.
Working output: a documented comparison, its limitations and an agreed next decision.
05 · Sustain
Keep improvement under review.
Depending on scope, follow-up may monitor agreed indicators, review recurring variation and keep open actions visible. Set review intervals and escalation points with the organization so responsibilities remain clear when performance changes.
Compare periods or sites only where the context and definitions make the comparison useful. Look for regression, unresolved findings and new concerns. Changes in staffing, protocols, equipment or workload may call for a new baseline or renewed assessment.
The aim is a managed improvement process, with priorities revisited as conditions change. Continuous quality improvement management is OraDigit's business direction; the duration, monitoring arrangements and support for any ongoing engagement require agreement. This framework does not imply an existing managed-service contract.
Working output: a follow-up plan that connects monitoring, open actions and reassessment.
Use technology where it improves the process.
Data organization and analytics may support assessment; visualization and comparison may help identify patterns. Structured workflow tools may support improvement. Dashboards and longitudinal tracking may support measurement, while monitoring and alerts may support follow-up. These are possible uses, not a claim that every system currently exists.
Software, automation and AI require a defined purpose, appropriate data governance and professional review. Choose technology when it helps answer the question or carry out the process; evaluate it before relying on its outputs.
Current examples include Order Helper (Demonstration) for structured order information; PET Quant (Research prototype) for quantitative PET/CT context and measurement-consistency research; and PET Response Tracker (Research prototype) for longitudinal PET review. AI Assistant remains an informational Demonstration. These tools do not establish clinical validation or replace professional judgment.
Labs explores further methods. Future analytics and applications need confirmed scope, governed inputs and evidence review. Use synthetic information in demonstrations and prototypes; do not submit patient or confidential organizational information.
Explore Research & TechnologyClear responsibility at every stage.
OraDigit supports improvement work. Clinical decisions remain with appropriately qualified healthcare professionals; operational authority remains with the healthcare organization. Medical physics responsibilities remain with qualified medical physicists where required.
Accreditation decisions remain with accrediting organizations. Compliance determinations remain with applicable authorities and the organization's qualified leadership. Readiness support does not confer accreditation or certify compliance. Technology supports this governance and does not replace professional judgment.
Start with a measurable imaging problem.
The strongest starting point is a specific concern, a defined scope and a useful baseline. Bring the question your imaging service needs to understand; together, we can define a focused next step.