Program Information
Automated Daily Patient Treatment Chart Checks Using Data Mining System
A Dhabaan*, X Jiang , E Schreibmann , E Elder , Winship Cancer Institute of Emory University, Atlanta, GA
Presentations
SU-E-T-140 (Sunday, July 12, 2015) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose:The aim of this study is the use of data mining techniques to establish an automated daily chart check. This automated technique increases the likelihood of error capture before subsequent treatment delivery, allowing the delivery of the prescribed treatment course with high confidence level. Furthermore, the scope of inspected parameters expands compared to that able to be performed manually.
Methods:Considering the significant number of parameters that are checked, performing chart check manually has a number of shortfalls such as human error, comparison of incorrect field and inability to verify MLC shape and leaves positions. Developing a computer-driven automatic chart checkers eliminates human error, improves the accuracy and the frequency of the process, shortens the time of error discovery and correction, increases the number of parameters that are checked, allows inspection of parameters that are impossible to check manually and reduces the time the physicist need to spend to check a chart. This work will utilizes in house data mining framework (RT Metrix) to establish this QA function. The system utilize ARIA data base to verify the treatment delivery by comparing the plan record in ARIA to record verification which is the delivery parameters. Acceptable tolerances were defined in the software. If the delivery variation from the tolerance is minor then the system will show a warning flag but if the variation is major the system will flag the delivery as failed.
Results:This system has been developed, tested, compared to ARIA Chart QA and is currently able to check parameters that routinely checked by the clinical physicist.
Conclusion:The automated chart check system is feasible and offers great potential in the improvement of patient delivery quality assurance. The system can be developed further to include more QA functions.
Contact Email: