RPS 613 - Modern strategies, risk, and safety management in medical imaging

Author Block: J. I. Peltonen¹, A-M. Hoyos-Garcia², D. Y. Nersissian³, T. Li Kuo⁴, I. Fitton⁵, V. Tsapaki⁶, O. CIRAJ BJELAC⁶, M. Kortesniemi¹; ¹Helsinki/FI, ²na/MX, ³São Paulo/BR, ⁴Kuala Lumpur/MY, ⁵Paris/FR, ⁶Vienna/AT
Purpose: To test previously created automated quality assurance (QA) framework for computed tomography (CT) as a multi-center international pilot use with various different CT models as part of the International Atomic Energy Agency (IAEA) Coordinated Research Project (CRP) entitled “Advanced Tools for Quality and Dosimetry of Digital Imaging in Radiology” (E24025).
Methods or Background: The QA framework was built on free open-source code components and libraries to facilitate future development and for freely sharing knowledge within global medical physics and radiation protection community.

The framework core components included results database, in-house built database communication interface, interactive user interface, data processing backend and a DICOM server.

The QA framework requires only the standard CT calibration (vendor specific) phantom with homogenous section included in any commercial CT system installation. The QA analysis provides all essential technical image quality features (e.g. CT numbers and their variation, contrast, noise, resolution) including also model-observer based values (detectability index).
Results or Findings: The pilot installation was deployed in 6 countries with 12 CT systems in total representing various scanner generations and platforms, including conventional multi-slice scanners and new photon counting detector-based CT. The results demonstrated excellent applicability with different systems and environments.
Conclusion: Large scale CT QA framework built on free open-source code has been successfully tested on multi-center international environment within IAEA pilot project. The framework can be prepared for large scale global use.
Limitations: This research outlines early findings from the system's broad implementation.
Funding for this study: No dedicated funding was provided for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: A. Romanyukha¹, L. Garajová², N. Fitousi¹, A. M. Sprinkart²; ¹Leuven/BE, ²Bonn/DE
Purpose: New scanner technology comes with promise of lower exposure and improved workflow. The purpose of this study was to QA a new spectral CT device by evaluating the difference in standard scan parameters, exposure and technique between old and new scanners of the same vendor.
Methods or Background: 6 months of data was exported from the dose management system including study compositions, scan parameters (kVp, pitch, collimation, exposure time per rotation, scanlength), and indicators of exposure (CTDIvol, SSDEWED) and technique (positioning offsets, lengths of scans outside localizer borders) for the same exams on iCT 256 (old) and spectral CT 7500 (new) scanners (Philips Healthcare): chest-abdomen with contrast (n=295 and n=694, respectively), chest with contrast (n=106 and n=220, respectively) and chest without contrast (n=252 and n=393, respectively). All data was collected from radiation dose structured reports.
Results or Findings: Study compositions, kVp and collimation were identical, while pitch was 1.8 times higher and exposure time per rotation was 0.06 s lower on the spectral CT 7500 for all protocols. Median scanlength was 1-6% higher on the new scanner for standard exams, likely attributed to patient differences. CTDIvol and SSDEWED were 50% lower for both chest protocols on the new scanner, and 14% and 22% lower, respectively, for the chest-abdomen protocol on the new scanner. Higher ranges of CTDIvol per patient WED were observed on the new scanner. Vertical and horizontal positioning offsets improved by 47-86% and 57-89%, respectively, on the new scanner. Scanning outside localizer borders was uncommon on both scanners and reduced further on the new scanner.
Conclusion: Differences in scan parameters were found, and spectral CT 7500 demonstrated reduced dose and improved technique. Investigation into scan settings and their impact on dose will be performed.
Limitations: None
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: S. Sawall, L. T. Rotkopf, T. D. Do, H-U. Kauczor, C. H. Ziener, H-P. Schlemmer, M. Kachelrieß; Heidelberg/DE
Purpose: To design and evaluate a cost-efficient, semi-anthropomorphic phantom for quality assurance of CT lung cancer screening protocols, with emphasis on accessibility and applicability in resource-limited settings.
Methods or Background: The phantom was built from readily available materials for task-based image quality assessment at clinically relevant intersection lengths and dose levels. Candidate materials including ABS, PMMA, polycarbonate, polyethylene, and polyurethane (PU) foams were tested for machinability, attenuation, and water equivalence. A semi-anthropomorphic phantom with thoracic dimensions of 20 × 30 cm approximating adults was constructed. It includes features for the assessment of task transfer function, slice sensitivity profile, noise power spectrum, image noise, and lesion detectability. Cylindrical inserts and nodules of varying sizes and contrasts were embedded to mimic solid and ground-glass nodules. Imaging experiments were performed on an energy-integrating CT (Somatom Force) and a photon-counting CT (Naeotom Alpha.Peak) at multiple tube voltages and prefilters. Furthermore, a calibration approach is introduced in which CT values are modeled as a function of PU density or alternative materials, ensuring consistent evaluation despite local material variability.
Results or Findings: ABS and PU foams proved most suitable for simulating tissue and low-contrast lesions. ABS served as the phantom body and solid nodules, while PU foams of different densities represented lung parenchyma and ground-glass nodules. These materials provided water-equivalent attenuation, stable CT values across scanners and voltages, and realistic intersection lengths. The phantom enabled robust image quality assessment.
Conclusion: Our low-cost, reproducible phantom for CT lung cancer screening QA can be manufactured on site with simple tools, supporting broad use in clinical practice. To support adoption, design files and specifications will be made available.
Limitations: Manual fabrication may reduce structural accuracy, although CNC machining can improve precision where available.
Funding for this study: N/A
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: C. Lee; Rockville/US
Purpose: Traditional dosimetry methods in diagnostic imaging often rely on reference patient models, overlooking the diversity of age, sex, body size, and pregnancy status in real-world practice. This “one-size-fits-all” approach limits accuracy and underestimates variability in radiation dose. Recent advances with the National Cancer Institute dosimetry systems for Computed Tomography (NCICT) and Radiography and Fluoroscopy (NCIRF) move beyond population averages to provide personalized dose estimates.
Methods or Background: NCICT and NCIRF integrate Monte Carlo–derived dose libraries with advanced computational human phantoms that span pediatric through adult anatomies, both sexes, varying body habitus, and more recently pregnant patients. Organ dose estimates are individualized by merging scanner- or procedure-specific parameters with anatomically matched phantoms. NCIRF further models spatial dose distributions to estimate local peak skin dose—critical for patients undergoing complex, high-dose interventions.
Results or Findings: By accounting for patient-specific characteristics, individualized organ dose estimates can differ several-fold from reference values, particularly in pediatric, obese, or pregnant cohorts. NCIRF’s peak skin dose calculations enable proactive management of deterministic effects. Both systems are designed for seamless integration with dose reports, enabling high-throughput monitoring in clinical practice and retrospective dose reconstruction in research cohorts. Large-scale applications have demonstrated their utility for personalized radiation protection, patient counseling, and epidemiological studies.
Conclusion: NCICT and NCIRF represent a paradigm shift in diagnostic imaging dosimetry—moving beyond “one size fits all” to precision dose assessment. By leveraging the most advanced computational human phantoms and real-world imaging data, these systems provide powerful tools for patient-specific monitoring and long-term risk evaluation in both CT and fluoroscopy.
Limitations: Although NCICT and NCIRF employ individualized computational phantoms to improve accuracy, residual anatomical discrepancies from actual patients remain and may introduce uncertainty in organ and skin dose estimates.
Funding for this study: This research was supported by the Intramural Research Program of the National Institutes of Health (NIH). The contributions of the NIH author(s) were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered Works of the United States Government. However, the findings and conclusions presented in this paper are those of the author(s) and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: P. Toroi¹, I. Stojanović², M. Živanović², A. Ciccotelli³, N. Kiiskinen⁴, M. Borowski⁵; ¹Vantaa/FI, ²Vinča, Belgrade/RS, ³Rome/IT, ⁴Helsinki/FI, ⁵Braunschweig/DE
Purpose: Measurement results should always be reported together with the estimated uncertainties. Measurement uncertainties must be sufficiently low to achieve the accuracy requested for clinical measurements. For air kerma measurements the target uncertainty is well established e.g., in IAEA TRS-457. However, for the other quality control (QC) parameters such as tube voltage, half-value layer, and irradiation time, the target uncertainty has not yet been clearly defined [Komatina I, et al. 2025 Physica Medica 136 105055].
Methods or Background: The dosimeter performance and associated uncertainties were analyzed within a European project using data from surveys, publications and experimental measurements. The achievable uncertainties were compared with established targets, and their clinical relevance was evaluated.
Results or Findings: For air kerma measurements, a target uncertainty of 5% (k = 2) can be reliably achieved with proper calibration. In contrast, for the other QC parameters the situation is less clear. IEC standards provide some acceptance criteria. For example, tube voltage should be within 8% in general radiology and 5% in mammography. To verify compliance with such limits, the measurement uncertainty must be considerably lower than the acceptance threshold. For the other QC parameters, an uncertainty of 2% is often set as a goal, although this is challenging to achieve. While reproducibility can be excellent (uncertainty <1%), absolute accuracy strongly depends on proper calibration. If a device is used under conditions different from those of its calibration, systematic errors exceeding 5% may occur.
Conclusion: Dosimeters used in X-ray imaging enable measurement of a wide range of QC parameters. However, for quantities other than air kerma, the associated uncertainties are still poorly known. This study clarifies achievable uncertainty levels and supports a more informed evaluation of whether observed measurement deviations are significant.
Limitations: Not applicable.
Funding for this study: This work was funded by the project 22NRM01 TraMeXI which has received funding from the European Partnership on Metrology, cofinanced from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States. Funded by the European Union.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: V. Mari, N. Ntoufas, S. Koukouraki, K. Perisinakis; Iraklion/GR
Purpose: To estimate the contribution of CT exposure and FDG administration to the life attributable radiogenic risk (LAR) of cancer associated with PET/CT examinations and compare the total LAR to the nominal life intrinsic risk (LIR) of cancer.
Methods or Background: The radiation dose to radiosensitive organs was calculated in a series of 24 patients subjected to PET/CT. Organ radiation doses from the administered FDG activity were determined using organ dose per activity factors. To estimate CT exposure-related organ doses, a Monte Carlo simulation software package was used that simulates the actual patient CT exposure on a patient-specific mathematical phantom generating dose images in 1-1 correspondence to original CT images. The FDG-related, CT-related, cumulative organ doses and published organ-specific radiogenic cancer risk factors were employed to estimate corresponding LAR estimates. The mean total LAR was compared to recently published LIR of cancer.
Results or Findings: The FDG-related organ doses ranged from 1.3 mGy (skin) to 27.9 mGy (bladder). The CT exposure-related organ doses ranged from 7.2 mGy (uterus) to 18.1 mGy (thyroid). The CT-related and FDG-related LAR was estimated to be 3.9 x10-4 and 2.0 x10-4 for males and 10.4 x10-4 and 3.4 x10-4 for females, respectively. The total LAR from a PET/CT examination was found to increase the risk for carcinogenesis to a cancer-free individual subjected to PET/CT by only 1.0015 and 1.0042 for male and female individuals, respectively.
Conclusion: The main contributor to the radiogenic risk of cancer from PET/CT examinations is CT, contributing by 78% and 65% to the total LAR of female and male patients, respectively. The increase of the nominal risk of cancer induction following a PET/CT examination with a modern system is < 0.42% and may be considered marginal.
Limitations: Cohort size.
Funding for this study: None.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: B. Gricienė, I. Andriulevičiūtė, K. Skovorodko, L. Krynke, M. Jeršova, V. Statkus, A. Samuilis; Vilnius/LT
Purpose: The aim of this study is to provide an overview of the Lithuanian legal framework for the registration of accidental and unintended radiation exposures and to present national statistics on patient and occupational exposure cases exceeding investigation levels.
Methods or Background: Across Europe, the requirements for recording and reporting unplanned or emergency exposure situations vary considerably, reflecting differences in national radiation protection strategies and regulatory approaches. In Lithuania, these requirements established through national hygiene norms, which provide the legal basis for mandatory reporting. In addition, hospitals are required to implement local orders that regulate the reporting and analysis of unintended events. Patient cases in which radiation exposure during interventional radiology procedures leads to observable skin effects, except when associated with complex procedures, must be reported to the competent authority. For occupational exposures, hospitals are obliged to establish local investigation levels. These serve as threshold values that trigger further analysis, corrective actions, and preventive measures. In practice, this requires the systematic registration of unintended events, the application of optimization strategies, and the use of structured procedures for case analysis.
Results or Findings: Statistics on unintended events in diagnostic radiology reported by Lithuanian hospitals from 2020 to 2025 were analyzed. The majority of events were related to human error or equipment malfunction. Occupational exposure cases exceeding investigation levels were analysed, with most associated with demanding workloads and complex fluoroscopy guided interventional procedures.
Conclusion: Our findings highlight the importance and challenges of implementing a clear legal framework for the systematic registration of accidental and unintended radiation exposure cases. National statistics confirm that the application of error system registration and investigation levels provides a practical and effective tool for harmonizing case analysis, supporting optimization, ensuring consistent radiation protection standards.
Limitations: Not applicable.
Funding for this study: No funding was provided for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: A. Papachristodoulou¹, C. Paraskevopoulou², C. Domingos Filho³, D. Baltazar⁴, A. Roncacci⁵; ¹Thessaloniki/GR, ²Athens/GR, ³Barcelona/ES, ⁴Amadora/PT, ⁵The Hague/NL
Purpose: To evaluate trends in clinical incidents related to patient identification and procedure matching across a European diagnostic and radiotherapy network, and to assess the impact of targeted safety interventions.
Methods or Background: Clinical incident (CI) reports were collected from 286 diagnostic imaging and radiotherapy centres across 15 European countries. Incidents included repeated examinations, unintended radiation exposures, wrong procedure, wrong patient, wrong side and wrong site imaged or treated. Data was analyzed from January 2023 to August 2025. An action plan was introduced in response to peaks observed in March 2023 and April 2024. Safety interventions included extensive staff training, safety campaigns, systematic review of incidents, and deployment of digital identification tools.
Results or Findings: In 2023, 36 incidents were reported in January, with reductions to 10–18 per month from May onward, after the first intervention. The overall burden was highest in 2023 compared with subsequent years. In 2024, incident counts peaked at 21 (March) but stabilized at lower averages (8–13). Data from 2025 (Q1–Q3) showed further improvement, with monthly counts between 2–13, consistently below 2023–2024 levels. High-risk categories such as wrong patient and wrong site incidents declined significantly, while repeated examinations remained the most frequent category. The downward trend correlates with the roll-out of the action plan and harmonized safety protocols across countries.
Conclusion: A structured patient safety program, triggered by peak periods and supported by training, campaigns, and clinical review, produced sustained reductions in errors. Embedding these actions into clinical governance frameworks advances value-based healthcare by reducing harm and optimizing processes and resources. Continuous analysis of incidents is essential to identify trends and enable proactive mitigating actions that further strengthen safety culture across imaging networks.
Limitations: N/A
Funding for this study: N/A
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: J. H. J. Ketola, S. Inkinen; Helsinki/FI
Purpose: Diagnostic ultrasound (US) transducers are prone to mechanical damage due to wearing or impact. Many defects can be identified by evaluating in-air reverberation patterns. Defective transducers can be either repaired or replaced. This study aimed to evaluate the outcomes associated with our technical quality control (QC) program for US transducers in our unit covering over 80 US devices with ca. 300 transducers.
Methods or Background: In January 2024, our hospital district implemented a digital quality control platform. Each diagnostic transducer is tested monthly by taking in-air reverberation images, which are available on the platform. If a transducer is found defective, medical engineering staff are notified to arrange repair or replacement.
Results or Findings: By October 2025, a total of 187 transducer defects had been identified by medical physicists or local radiographers/sonographers. Crystal damage was confirmed in 53 transducers through phantom measurements. Of these, 20 transducers were repaired and 33 were replaced — one with an existing unused probe, seven under warranty claims, and 25 through new purchases. Electrical safety assessments detected excess leakage current in three transducers, while one was found to have a short circuit. Faulty connectors and beamformers accounted for eight defects. Following repairs, five transducers exhibited recurring issues, prompting compensation claims. Additionally, one repaired transducer demonstrated reduced penetration depth in phantom testing and was subsequently replaced.
Conclusion: Our technical quality control program effectively detects transducer defects. Significant cost savings can be achieved by utilizing warranty claims for replacement and opting to transducer repairs rather than acquiring new transducers. However, our recent findings emphasize importance of QC for repaired transducers as they may present recurrent or novel issues, potentially invalidating financial benefits.
Limitations: No limitations were identified.
Funding for this study: No external funding.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: