Eliminating inconsistencies in simulation and treatment planning orders in radiation therapy.

PURPOSE: To identify deficiencies with simulation and treatment planning orders and to develop corrective measures to improve safety and quality. METHODS AND MATERIALS: At Washington University, the DMAIIC formalism is used for process management, whereby the process is understood as comprising Define, Measure, Analyze, Improve, Implement, and Control activities. Two complementary tools were used to provide quantitative assessments: failure modes and effects analysis and reported event data. The events were classified by the user according to severity. The event rates (ie, number of events divided by the number of opportunities to generate an event) related to simulation and treatment plan orders were determined. RESULTS: We analyzed event data from the period 2008-2009 to design an intelligent SIMulation and treatment PLanning Electronic (SIMPLE) order system. Before implementation of SIMPLE, event rates of 0.16 (420 of 2558) for a group of physicians that were subsequently used as a pilot group and 0.13 (787 of 6023) for all physicians were obtained. An interdisciplinary group evaluated and decided to replace the Microsoft Word-based form with a Web-based order system. This order system has mandatory fields and context-sensitive logic, an ability to create templates, and enables an automated process for communication of orders through an enterprise management system. After the implementation of the SIMPLE order, the event rate decreased to 0.09 (96 of 1001) for the pilot group and to 0.06 (145 of 2140) for all physicians (P<.0001). The average time to complete the SIMPLE form was 3 minutes, as compared with 7 minutes for the Word-based form. The number of severe events decreased from 10.7% (45 of 420) and 12.1% (96 of 787) to 6.2% (6 of 96) and 10.3% (15 of 145) for the pilot group and all physicians, respectively. CONCLUSIONS: There was a dramatic reduction in the total and the number of potentially severe events through use of the SIMPLE system. In addition, the order process has become more efficient and reliable.

Technical note: electronic chart checks in a paperless radiation therapy clinic.

PURPOSE: EcCk, which stands for Electronic Chart ChecK, is a computer software and database system. It was developed to improve quality and efficiency of patient chart checking in radiation oncology departments. The core concept is to automatically collect and analyze patient treatment data, and to report discrepancies and potential concerns. METHODS: EcCk consists of several different computer technologies, including relational database, DICOM, dynamic HTML, and image processing. Implemented in MATLAB and C#, EcCk processes patient data in DICOM, PDF, Microsoft Word, database, and Pinnacle native formats. Generated reports are stored on the storage server and indexed in the database. A standalone report-browser program is implemented to allow users to view reports on any computer in the department. Checks are performed according to predefined logical rules, and results are presented through color-coded reports in which discrepancies are summarized and highlighted. Users examine the reports and take appropriate actions. The core design is intended to automate human task and to improve the reliability of the performed tasks. The software is not intended to replace human audits but rather to aid as a decision support tool. RESULTS: The software was successfully implemented in the clinical environment and has demonstrated the feasibility of automation of this common task with modern clinical tools. The software integrates multiple disconnected systems and successfully supports analysis of data in diverse formats. CONCLUSIONS: While the human is the ultimate expert, EcCk has a significant potential to improve quality and efficiency of patient treatment record audits, and to allow verification of tasks that are not easily performed by humans. EcCk can potentially relieve human experts from simple and repetitive tasks, and allow them to work on other important tasks, and in the end to improve the quality and safety of radiation therapy treatments.

Regarding the focal treatment of prostate cancer: inference of the Gleason grade from magnetic resonance spectroscopic imaging.

PURPOSE: To quantify, as a function of average magnetic resonance spectroscopy (MRS) score and tumor volume, the probability that a cancer-suspected lesion has an elevated Gleason grade. METHODS AND MATERIALS: The data consist of MRS imaging ratios R stratified by patient, lesion (contiguous abnormal voxels), voxels, biopsy and pathologic Gleason grade, and lesion volume. The data were analyzed using a logistic model. RESULTS: For both low and high Gleason score biopsy lesions, the probability of pathologic Gleason score >/=4+3 increases with lesion volume. At low values of R a lesion volume of at least 15-20 voxels is needed to reach a probability of success of 80%; the biopsy result helps reduce the prediction uncertainty. At larger MRS ratios (R > 6) the biopsy result becomes essentially uninformative once the lesion volume is >12 voxels. With the exception of low values of R, for lesions with low Gleason score at biopsy, the MRS ratios serve primarily as a selection tool for assessing lesion volumes. CONCLUSIONS: In patients with biopsy Gleason score >/=4+3, high MRS imaging tumor volume and (creatine + choline)/citrate ratio may justify the initiation of voxel-specific dose escalation. This is an example of biologically motivated focal treatment for which intensity-modulated radiotherapy and especially brachytherapy are ideally suited.

Calibration procedures for seeds preloaded in cartridges.

Radioactive seeds preloaded in sterilized cartridges or needles are commonly obtainable from manufacturers. Under the US regulations for control of radioactive materials, seed users are required to account for all seeds and independently verify their air kerma strength (SK). As a result, the viability of inspection schemes that rely on measurement of aggregate seeds is of interest. In this paper we consider the conditions (if any) under which cartridge inspection can satisfy regulatory requirements and still provide practical benefit (i.e., time savings) against the regular single-seed assay. The standards for comparison are the recommendations of AAPM TG40, AAPM TG56, and ACR's "Standard for the Performance of Manually Loaded Brachytherapy Sources." The practical benefit is judged in comparison to the effort required to apply the 10% assay recommendation of TG40 to seeds in cartridges. Two specific cartridge inspection schemes are considered: (a) measuring the SK of each cartridge in a batch; (b) measuring a single cartridge sampled at random from the batch. Unlike the 10% assay, which is defined (imperfectly, in our view) without reference to the prevalence of in-calibration seeds, the estimation of the relative merits of cartridge inspection methods must necessarily include such information and, as such, is manufacturer specific. In this paper results are provided for Oncura model 6711 125I seeds in shielded and unshielded Mick cartridges. We show that the only practically useful cartridge inspection scheme is the batch scheme applied to unshielded cartridges. The false positive rates associated with the other schemes are such that we expect to open a cartridge (and perform the 10% assay) at least 80% of the time. While anything less than 100% of the time is theoretically an improvement, this neglects the additional effort required to assay the cartridges.