TABLE 1.
List of failure modes and mitigations for the ATP workflow on the Unity
| Step | Process map step name | Potential failure mode(s) | Potential causes of failure | Mitigation(s) for failure mode |
|---|---|---|---|---|
| 1 | Preparation | |||
| Bring patient into room | Wrong patient | Name and birthday not verified | (1) Front desk checks patient name/birthday and gives wrist band; (2) therapists reverify name/birthday, and (3) confirm with wristbands given at front desk | |
| Wrong plan loaded on MOSAIQ | Photo check missed | Therapists have tablet that view the MOSAIQ computer which has patient photo | ||
| MRI questionnaire form filled and signed (patient, therapist, witness) | Questionnaire filled out incorrectly; patient contains ferromagnetic material | Patient did not read thoroughly | (1) Tablet to fill out MOSAIQ assessment with patient; (2) daily implant email for MRI patients | |
| Patient forgot about implant | (1) Daily automated implant email for MRI patients; (2) therapist to question scars | |||
| Daily/initial assessment in MOSAIQ | Filled out form incorrectly | Questionnaire incorrect so wrong information put in assessment as a result | (1) Tablet to fill out MOSAIQ assessment with patient; 2) daily implant email for MRI patients | |
| Therapist mistake when filling out assessment | Saying “yes” to implant or pregnant shows up red in MOSAIQ and MRI eligibility is not met so treatment will be prevented | |||
| Verify patient name and birthday with patient again | Second therapist did not verify twice | Thought first therapist verified | It is a good check to have but patient will have had three other verifications of id (front desk, first therapist, photo check) | |
| Wand patient in zone III | Wand malfunctions | Electronics error | (1) Have backup wand at MRI sim and (2) in‐room metal detectors (on wall) | |
| Wanding not complete | Did not wand all over | Must complete wand training module before wanding patient | ||
| Move wand too fast | Must complete wand training module before wanding patient | |||
| Wand not on | Thought on but it is not | Use wand only with sound on | ||
| Did not wand | Unaware of policy | (1) MRI training provides understanding dangers of ferromagnetic materials; (2) patients are in gowns so chance of carrying metal on body is low | ||
| Forgot | (1) In‐room metal detectors should go off; (2) patients are in gowns so chance of carrying metal on body is low | |||
| 2 | Positioning and imaging | |||
| Setup per simulation and drive to couch index per plan | Patient has moved in mold such that fields are deliverable (small field) | Patient discomfort | Verification scan will be thoroughly reviewed by planner and physicists to catch motion | |
| Couch goes to wrong location | Board level not entered correctly in sim | (1) Monaco does not allow automated shifts >5 cm; (2) department policy requires all shifts >0.5 cm to be rechecked | ||
| Take other sequences as needed for research | Motion monitoring stopped early to accommodate research scan | Unaware of policy | Training for designated MM watcher should include knowledge that stopping mm is only allowed for post MRI image | |
| Transfer from MARLIN to online using “Monaco” destination | Wrong scan sent | Did not check time or date | MRI tech to “call‐out” patient name, image date + time and MRI sequence; planner will verbally confirm | |
| 3 | Fusion/preoptimization | |||
| Decide which reference plan to used based on prior fraction notes as applicable | Wrong plan is chosen | Entered correctly in checklist but wrong plan chosen in the dropdown (ATP only treatment) | (1) Planner to “call‐out” MRI number and plan name/number and therapist and physicist to verify call‐out and on‐screen; (2) smart PIC to check if plan is based on most recent ATS | |
| Entered correctly in checklist but wrong plan chosen in the dropdown (ATP after ATS for prostate) | (1) therapists should update excel sheet with ATS plan number before proceeding to ATP; (2) planner to “call‐out” MRI number and plan name/number and therapist and physicist to verify call‐out and on‐screen; (3) Smart PIC to check if plan is based on most recent ATS | |||
| Reference plan name that should be used was entered incorrectly/vague on online checklist | Treating physicist should independently determine which reference plan to use beforehand | |||
| Select appropriate reference plan name in online Monaco under “reference plan” | After prostate ATS and ATP is now needed, the wrong scan is selected | Mistakenly selected wrong plan | (1) Planner to “call‐out” MRI number and plan name/number and therapist and physicist to verify call‐out and on‐screen; (2) smart PIC to check if plan is based on most recent ATS | |
| Fuse scan rigidly | Fused incorrectly | Fusion controls used improperly | Planner and physicist to independently review fusion | |
| Evaluate fusion | Primary and secondary scans mixed up during registration | Not clarified by planner | Always double check “show images” window to see which image is in active view | |
| Patient weight loss causes significant difference in fusion | Patient weight change | Call MD if overall change is 1 cm or more, as it may require ATS instead | ||
| Better to match on non‐bone structure but matched on bone | Matched on wrong anatomic area thinking it was correct | Out “fuse on” information in online checklist | ||
| 4 | Optimization | |||
| Optimize | Stray target contours leading to stray dose | Did not thoroughly evaluate contours in reference plan | Cannot make changes in ATP so need to address this during initial offline plan check | |
| Check ED under grid type > electron density | ED is not correct on reference scan | Planner did not check thoroughly, or wrong layer order or ED not assigned | (1) ED is double checked online: all planners (including therapist planners) are trained to expect max of 1.8 unless there is an implant; (2) review of ED assignment should include entire image in all three planes | |
| Take verification MRI | Did not take and patient moved | Forgot or unaware of policy | Online checklist states order of steps and includes taking the verification MRI | |
| 5 | Plan review and approval | |||
| Inspect DVH statistics (cursory) | Constraints incorrect for PTV | Incorrect because multiphase plan not scaled correctly | Cannot change constraints for ATP: checked thoroughly during initial offline plan check | |
| Constraints incorrect because template changed in prior fraction | (1) Online checklist states which reference plan to use; (2) physicist also independently determines this before treatment; (3) planner will use call‐out procedure to ensure selection is correct | |||
| Constraints incorrect for critical organ (limit) | Incorrect because multiphase plan not scaled correctly | Cannot change constraints for ATP: checked thoroughly during initial offline plan check | ||
| Inspect isodoses for bug in non‐contiguous structures | Did not check | (1) Planner and physicist should review dose together; (2) cannot change contours in ATP, so must be checked during initial offline plan check | ||
| Unaware of policy | (1) Instructions are in online checklist to review plan before approval; (2) cannot change contours in ATP, so must be checked during initial offline plan check | |||
| Inspect verification scan by toggling between MRI old and new MRI (in drop down bar) | Should have done ATS because of patient motion but did ATP | Did not review thoroughly verification images | Planner and physicist should independently review verification images | |
| Accidentally reviewed the wrong images (not new MRI) | Always double check “show images” window to see which image is in active view | |||
| Gas bubble pushing into PTV prompts second ATS which is missed | Did not review thoroughly verification images | Planner and physicist should independently review verification images | ||
| Approve plan and close Monaco | Transfer incorrect from Monaco to MOSAIQ | Control points or mu does not match; partial plan transfer; gantry does not match | Smart PIC will catch any of these by comparing Monaco to MOSAIQ | |
| 6 | Plan check | |||
| Run smart PIC and sanity check | Did not run PIC or sanity check | Forgot | Add check to IMU script to make sure PIC passes (acts as reminder to run PIC too) | |
| 7 | Treatment | |||
| Run IMU | IMU returns a false positive | Incorrect ED | ED is checked during plan optimization; planners/therapists are trained to know the expected range and to review ED in all three planes along the entire image | |
| Dose calculation issue | Do arc check after treatment | |||
| Watch MM | Did not watch MM | Other issues to attend to | (1) Will have designated MM watcher who sole duty is to watch the MM screen; (2) implement action levels of what to do if there is motion | |
| Unaware of policy | (1) Update online checklist to include a “watch MM” step; (2) assign roles prior to patient treatment of who does what | |||
| Contour on MM not representative of oar motion (i.e., wrong slice) | Center of contour not ideal | Cannot change contours in ATP: check center of MM structure during initial offline plan check or most recent ATS fraction | ||
| 8 | Post treatment | |||
| MD signs report in aria | Miss bad fusion | Did not review thoroughly | Planner and physicist will independently review fusion so this should not happen | |
| Miss bad DVH | Did not review thoroughly | Planner and physicist will review dose prior to treating so this should not happen |
Abbreviations: ATP, adapt‐to‐position; ATS, adapt‐to‐shape; DVH, dose volume histogram; ED, electron density; IMU, independent monitor unit verification; MM, motion monitoring, MRI, magnetic resonance imaging; PIC, plan integrity check; PTV, planning target volume.