Personalized Prediction of Patient Radiation Exposure for Therapy of Urolithiasis: An Application and Comparison of Six Machine Learning Algorithms.

The prediction of radiation exposure is an important tool for the choice of therapy modality and becomes, as a component of patient-informed consent, increasingly important for both surgeon and patient. The final goal is the implementation of a trained and tested machine learning model in a real-time computer system allowing the surgeon and patient to better assess patient's personal radiation risk. In summary, 995 patients with ureterorenoscopy over a period from May 2016 to December 2019 were included. According to the suggestions based on actual literature evidence, dose area product (DAP) was categorized into 'low doses' ≤ 2.8 Gy·cm2 and 'high doses' > 2.8 Gy·cm2 for ureterorenoscopy (URS). To forecast the level of radiation exposure during treatment, six different machine learning models were trained, and 10-fold crossvalidated and their model performances evaluated in training and independent test samples. The negative predictive value for low DAP during ureterorenoscopy was 94% (95% CI: 92-96%). Factors influencing the radiation exposure were: age (p = 0.0002), gender (p = 0.011), weight (p < 0.0001), stone size (p < 0.000001), surgeon experience (p = 0.039), number of stones (p = 0.0007), stone density (p = 0.023), use of flexible endoscope (p < 0.0001) and preoperative stone position (p < 0.00001). The machine learning algorithm identified a subgroup of patients of 81% of the total sample, for which highly accurate predictions (94%) were possible allowing the surgeon to assess patient's personal radiation risk. Patients without prediction (19%), the medical expert can make decisions as usual. Next step will be the implementation of the trained model in real-time computer systems for clinical decision processes in daily practice.

Optimized Camera-Based Patient Positioning in CT: Impact on Radiation Exposure.

OBJECTIVE: The aim of this study was to evaluate whether a 3-dimensional (3D) camera can outperform highly trained technicians in precision of patient positioning and whether this transforms into a reduction in patient exposure. MATERIALS AND METHODS: In a single-center study, 3118 patients underwent computer tomography (CT) scans of the chest and/or abdomen on a latest generation single-source CT scanner supported with an automated patient positioning system by 3D camera. One thousand five hundred fifty-seven patients were positioned laser-guided by a highly trained radiographer (camera off) and 1561 patients with 3D camera (camera on) guidance. Radiation parameters such as effective dose, organ doses, CT dose index, and dose length product were analyzed and compared. Isocenter accuracy and table height were evaluated between the 2 groups. RESULTS: Isocenter positioning was significantly improved with the 3D camera ( P < 0.001) as compared with visual laser-guided positioning. Absolute table height differed significantly ( P < 0.001), being higher with camera positioning (165.6 ± 16.2 mm) as compared with laser-guided positioning (170.0 ± 20.4 mm). Radiation exposure decreased using the 3D camera as indicated by dose length product (321.1 ± 266.6 mGy·cm; camera off: 342.0 ± 280.7 mGy·cm; P = 0.033), effective dose (3.3 ± 2.7 mSv; camera off: 3.5 ± 2.9; P = 0.053), and CT dose index (6.4 ± 4.3 mGy; camera off: 6.8 ± 4.6 mGy; P = 0.011). Exposure of radiation-sensitive organs such as colon ( P = 0.015) and red bone marrow ( P = 0.049) were also lower using the camera. CONCLUSIONS: The introduction of a 3D camera improves patient positioning in the isocenter of the scanner, which results in a lower and also better balanced dose reduction for the patients.

Handling of Incidents in the Clinical Application of Ionizing Radiation in Diagnostic and Interventional Radiology - a Multi-center Study. / Umgang mit unbeabsichtigten Expositionen bei medizinischen Anwendungen ionisierender Strahlung in der Röntgendiagnostik und interventionellen Radiologie ­ eine Multi-Center-Studie.

PURPOSE: According to the German legislation and regulation of radiation protection, i. e. Strahlenschutzgesetz und Strahlenschutzverordnung (StrlSchG and StrlSchV), which came into force on 31st December 2018, significant unintended or accidential exposures have to be reported to the competent authority. Furthermore, facilities have to implement measures to prevent and to recognize unintended or accidental exposures as well as to reduce their consequences. We developed a process to register incidents and tested its application in the framework of a multi-center-study. MATERIALS AND METHODS: Over a period of 12 months, 16 institutions for x-ray diagnostics and interventions, documented their incidents. Documentation of the incidents was conducted using the software CIRSrad, which was developed, released for testing purposes and implemented in the frame of the study. Reporting criteria of the project were selected to be more sensitive compared to the legal criteria specifying "significant incidents". Reported incidents were evaluated after four, eight, and twelve months. Finally, all participating institutions were interviewed on their experience with the software and the correlated effort. RESULTS: The rate of reported incidents varied between institutions as well as between modalities. The majority of incidents were reported in conventional x-ray imaging, followed by computed tomography and therapeutic interventions. Incidents were attributed to several different causes, amongst others to the technical setup and patient positioning (19 %) and patient movement or insufficient cooperativeness of the patient (18 %). Most incidents were below corresponding thresholds stated in StrlSchV. The workload for documenting the incidents was rated as appropriate. CONCLUSION: It is possible to monitor and handle incidents complient with legal requirements with an acceptable effort. The number of reported incidents can be increased by frequent trainings on the detection and the processing workflow, on the software and legal regulation as well as by a transparent error handling within the institution. KEY POINTS: · The software CIRSrad was developed to enable the present study and as prototype platform for a future radiological incident management system.. · 586 exceedances of thresholds were recorded by 16 facilities in a period of one year.. · Frequent trainings of all users increase the number of reported cases.. CITATION FORMAT: · Müller BS, Singer J, Stamm G et al. Handling of Incidents in the Clinical Application of Ionizing Radiation in Diagnostic and Interventional Radiology - a Multi-center Study. Fortschr Röntgenstr 2022; 194: 400 - 408.