Convolutional neural network-based kidney volume estimation from low-dose unenhanced computed tomography scans.

PURPOSE: Kidney volume is important in the management of renal diseases. Unfortunately, the currently available, semi-automated kidney volume determination is time-consuming and prone to errors. Recent advances in its automation are promising but mostly require contrast-enhanced computed tomography (CT) scans. This study aimed at establishing an automated estimation of kidney volume in non-contrast, low-dose CT scans of patients with suspected urolithiasis. METHODS: The kidney segmentation process was automated with 2D Convolutional Neural Network (CNN) models trained on manually segmented 2D transverse images extracted from low-dose, unenhanced CT scans of 210 patients. The models' segmentation accuracy was assessed using Dice Similarity Coefficient (DSC), for the overlap with manually-generated masks on a set of images not used in the training. Next, the models were applied to 22 previously unseen cases to segment kidney regions. The volume of each kidney was calculated from the product of voxel number and their volume in each segmented mask. Kidney volume results were then validated against results semi-automatically obtained by radiologists. RESULTS: The CNN-enabled kidney volume estimation took a mean of 32 s for both kidneys in a CT scan with an average of 1026 slices. The DSC was 0.91 and 0.86 and for left and right kidneys, respectively. Inter-rater variability had consistencies of ICC = 0.89 (right), 0.92 (left), and absolute agreements of ICC = 0.89 (right), 0.93 (left) between the CNN-enabled and semi-automated volume estimations. CONCLUSION: In our work, we demonstrated that CNN-enabled kidney volume estimation is feasible and highly reproducible in low-dose, non-enhanced CT scans. Automatic segmentation can thereby quantitatively enhance radiological reports.

Commercial AI solutions in detecting COVID-19 pneumonia in chest CT: not yet ready for clinical implementation?

OBJECTIVES: In response to the COVID-19 pandemic, many researchers have developed artificial intelligence (AI) tools to differentiate COVID-19 pneumonia from other conditions in chest CT. However, in many cases, performance has not been clinically validated. The aim of this study was to evaluate the performance of commercial AI solutions in differentiating COVID-19 pneumonia from other lung conditions. METHODS: Four commercial AI solutions were evaluated on a dual-center clinical dataset consisting of 500 CT studies; COVID-19 pneumonia was microbiologically proven in 50 of these. Sensitivity, specificity, positive and negative predictive values, and AUC were calculated. In a subgroup analysis, the performance of the AI solutions in differentiating COVID-19 pneumonia from other conditions was evaluated in CT studies with ground-glass opacities (GGOs). RESULTS: Sensitivity and specificity ranges were 62-96% and 31-80%, respectively. Negative and positive predictive values ranged between 82-99% and 19-25%, respectively. AUC was in the range 0.54-0.79. In CT studies with GGO, sensitivity remained unchanged. However, specificity was lower, and ranged between 15 and 53%. AUC for studies with GGO was in the range 0.54-0.69. CONCLUSIONS: This study highlights the variable specificity and low positive predictive value of AI solutions in diagnosing COVID-19 pneumonia in chest CT. However, one solution yielded acceptable values for sensitivity. Thus, with further improvement, commercial AI solutions currently under development have the potential to be integrated as alert tools in clinical routine workflow. Randomized trials are needed to assess the true benefits and also potential harms of the use of AI in image analysis. KEY POINTS: • Commercial AI solutions achieved a sensitivity and specificity ranging from 62 to 96% and from 31 to 80%, respectively, in identifying patients suspicious for COVID-19 in a clinical dataset. • Sensitivity remained within the same range, while specificity was even lower in subgroup analysis of CT studies with ground-glass opacities, and interrater agreement between the commercial AI solutions was minimal to nonexistent. • Thus, commercial AI solutions have the potential to be integrated as alert tools for the detection of patients with lung changes suspicious for COVID-19 pneumonia in a clinical routine workflow, if further improvement is made.

Radiation dose management systems-requirements and recommendations for users from the ESR EuroSafe Imaging initiative.

The European Directive 2013/59/Euratom requires member states of the European Union to ensure justification and optimisation of radiological procedures and store information on patient exposure for analysis and quality assurance. The EuroSafe Imaging campaign of the European Society of Radiology created a working group (WG) on "Dose Management" with the aim to provide European recommendations on the implementation of dose management systems (DMS) in clinical practice. The WG follows Action 4: "Promote dose management systems to establish local, national, and European diagnostic reference levels (DRL)" of the EuroSafe Imaging Call for Action 2018. DMS are designed for medical practitioners, radiographers, medical physics experts (MPE) and other health professionals involved in imaging to support their tasks and duties of radiation protection in accordance with local and national requirements. The WG analysed requirements and critical points when installing a DMS and classified the individual functions at different performance levels. KEY POINTS: • DMS are very helpful software tools for monitoring patient exposure, optimisation, compliance with DRLs and quality assurance. • DMS can help to fulfil dosimetric aspects of the European Directive 2013/59/Euratom. • The EuroSafe WG analyses DMS requirements and gives recommendations for users.

An International Survey of Quality and Safety Programs in Radiology.

PURPOSE: The aim of this study was to determine the status of radiology quality improvement programs in a variety of selected nations worldwide. METHODS: A survey was developed by select members of the International Economics Committee of the American College of Radiology on quality programs and was distributed to committee members. Members responded on behalf of their country. The 51-question survey asked about 12 different quality initiatives which were grouped into 4 themes: departments, users, equipment, and outcomes. Respondents reported whether a designated type of quality initiative was used in their country and answered subsequent questions further characterizing it. RESULTS: The response rate was 100% and represented Australia, Canada, China, England, France, Germany, India, Israel, Japan, the Netherlands, Russia, and the United States. The most frequently reported quality initiatives were imaging appropriateness (91.7%) and disease registries (91.7%), followed by key performance indicators (83.3%) and morbidity and mortality rounds (83.3%). Peer review, equipment accreditation, radiation dose monitoring, and structured reporting were reported by 75.0% of respondents, followed by 58.3% of respondents for quality audits and critical incident reporting. The least frequently reported initiatives included Lean/Kaizen exercises and physician performance assessments, implemented by 25.0% of respondents. CONCLUSION: There is considerable diversity in the quality programs used throughout the world, despite some influence by national and international organizations, from whom further guidance could increase uniformity and optimize patient care in radiology.

Validation of Semantic Analyses of Unstructured Medical Data for Research Purposes.

HINTERGRUND: In Sekundärdaten existieren oftmals unstrukturierte Freitexte. In dieser Arbeit wird ein Text-Mining-System validiert, um unstrukturierte medizinische Daten für Forschungszwecke zu extrahieren. METHODEN: Aus einer radiologischen Klinik wurden aus 7102 CT-Befunden 1000 zufällig ausgewählt. Diese wurden von 2 Medizinern manuell in definierte Befundgruppen eingeteilt. Zur automatisierten Verschlagwortung und Klassifizierung wurde die Textanalyse-Software Averbis Extraction Platform (AEP) eingesetzt. Besonderheiten des Systems sind u. a. eine morphologische Analyse zur Zerlegung zusammengesetzter Wörter sowie die Erkennung von Nominalphrasen, Abkürzungen und negierten Aussagen. Anhand der extrahierten standardisierten Schlüsselwörter werden Befundberichte mithilfe maschineller Lernverfahren den vorgegebenen Befundgruppen zugeordnet. Zur Bewertung von Reliabilität und Validität des automatisierten Verfahrens werden die automatisierten und 2 unabhängige manuelle Klassifizierungen in mehreren Durchläufen auf Übereinstimmungen hin verglichen. ERGEBNISSE: Die manuelle Klassifizierung war zu zeitaufwendig. Bei der automatisierten Verschlagwortung stellte sich in unseren Daten die Klassifizierung nach ICD-10 als ungeeignet heraus. Ebenfalls zeigte sich, dass die Stichwortsuche keine verlässlichen Ergebnisse liefert. Computerunterstütztes Textmining in Kombination mit maschinellem Lernen führte zu verlässlichen Klassifizierungen. Die Inter-Rater-Reliabilität der beiden manuellen Klassifizierungen, sowie der maschinellen und der manuellen Klassifizierung war sehr hoch. Beide manuelle Klassifizierungen stimmten in 93% aller Befunde überein. Der Kappa-Koeffizient beträgt 0,89 [95% Konfidenzintervall (KI) 0,87-0,92]. Die automatische Klassifizierung stimmte in 86% aller Befunde mit der unabhängigen, zweiten manuellen Klassifizierung überein (Kappa-Koeffizient 0,79 [95% KI 0,75-0,81]). DISKUSSION: Die Klassifizierung der Software AEP war sehr gut. In unserer Studie folgte sie allerdings einem systematischen Muster. Die meisten falschen Zuordnungen finden sich in Befunden, die auf ein erhöhtes Krebsrisiko hinweisen. Die Freitextstruktur der Befunde lässt Bedenken hinsichtlich der Machbarkeit einer rein automatisierten Analyse aufkommen. Die Kombination aus menschlichem Intellekt und einer intelligenten, lernfähigen Software erscheint als zukunftsweisend, um unstrukturierte aber wichtige Textinformationen der Forschung zugänglich machen zu können.

Comparison of medical-grade and calibrated consumer-grade displays for diagnosis of subtle bone fissures.

OBJECTIVE: To compare the diagnostic accuracy of medical-grade and calibrated consumer-grade digital displays for the detection of subtle bone fissures. METHODS: Three experienced radiologists assessed 96 digital radiographs, 40 without and 56 with subtle bone fissures, for the presence or absence of fissures in various bones using one consumer-grade and two medical-grade displays calibrated according to the DICOM-Grayscale Standard Display Function. The reference standard was consensus reading. Subjective image quality was also assessed by the three readers. Statistical analysis was performed using receiver operating characteristic analysis and by calculating the sensitivity, specificity, and Youden's J for each combination of reader and display. Cohen's unweighted kappa was calculated to assess inter-rater agreement. Subjective image quality was compared using the Wilcoxon signed-rank test. RESULTS: No significant differences were found for the assessment of subjective image quality. Diagnostic performance was similar across all readers and displays, with Youden's J ranging from 0.443 to 0.661. The differences were influenced more by the reader than by the display used for the assessment. CONCLUSION: No significant differences were found between medical-grade and calibrated consumer-grade displays with regard to their diagnostic performance in assessing subtle bone fissures. Calibrated consumer-grade displays may be sufficient for most radiological examinations. KEY POINTS: • Diagnostic performance of calibrated consumer-grade displays is comparable to medical-grade displays. • There is no significant difference with regard to subjective image quality. • Use of calibrated consumer-grade displays could cut display costs by 60-80%.

Visceral artery aneurysms: Incidence, management, and outcome analysis in a tertiary care center over one decade.

OBJECTIVES: To evaluate the incidence, management, and outcome of visceral artery aneurysms (VAA) over one decade. METHODS: 233 patients with 253 VAA were analyzed according to location, diameter, aneurysm type, aetiology, rupture, management, and outcome. RESULTS: VAA were localized at the splenic artery, coeliac trunk, renal artery, hepatic artery, superior mesenteric artery, and other locations. The aetiology was degenerative, iatrogenic after medical procedures, connective tissue disease, and others. The rate of rupture was much higher in pseudoaneurysms than true aneurysms (76.3% vs.3.1%). Fifty-nine VAA were treated by intervention (n = 45) or surgery (n = 14). Interventions included embolization with coils or glue, covered stents, or combinations of these. Thirty-five cases with ruptured VAA were treated on an emergency basis. There was no difference in size between ruptured and non-ruptured VAA. After interventional treatment, the 30-day mortality was 6.7% in ruptured VAA compared to no mortality in non-ruptured cases. Follow-up included CT and/or MRI after a mean period of 18.0 ± 26.8 months. The current status of the patient was obtained by a structured telephone survey. CONCLUSIONS: Pseudoaneurysms of visceral arteries have a high risk for rupture. Aneurysm size seems to be no reliable predictor for rupture. Interventional treatment is safe and effective for management of VAA. KEY POINTS: • Diagnosis of visceral artery aneurysms is increasing due to CT and MRI. • Diameter of visceral arterial aneurysms is no reliable predictor for rupture. • False aneurysms/pseudoaneurysms and symptomatic cases need emergency treatment. • Interventional treatment is safe and effective.

Structured reporting for efficient epidemiological and in-hospital prevalence analysis of pulmonary embolisms. / Strukturierte Befundung zur effizienten Erhebung von epidemiologischen Daten und Analyse der innerklinischen Prävalenz bei Lungenarterienembolien.

Structured reporting (SR) not only offers advantages regarding report quality but, as an IT-based method, also the opportunity to aggregate and analyze large, highly structured datasets (data mining). In this study, a data mining algorithm was used to calculate epidemiological data and in-hospital prevalence statistics of pulmonary embolism (PE) by analyzing structured CT reports.All structured reports for PE CT scans from the last 5 years (n = 2790) were extracted from the SR database and analyzed. The prevalence of PE was calculated for the entire cohort and stratified by referral type and clinical referrer. Distributions of the manifestation of PEs (central, lobar, segmental, subsegmental, as well as left-sided, right-sided, bilateral) were calculated, and the occurrence of right heart strain was correlated with the manifestation.The prevalence of PE in the entire cohort was 24% (n = 678). The median age of PE patients was 71 years (IQR 58-80), and the sex distribution was 1.2/1 (M/F). Outpatients showed a lower prevalence of 23% compared to patients from regular wards (27%) and intensive care units (30%). Surgically referred patients had a higher prevalence than patients from internal medicine (34% vs. 22%). Patients with central and bilateral PEs had a significantly higher occurrence of right heart strain compared to patients with peripheral and unilateral embolisms.Data mining of structured reports is a simple method for obtaining prevalence statistics, epidemiological data, and the distribution of disease characteristics, as demonstrated by the PE use case. The generated data can be helpful for multiple purposes, such as for internal clinical quality assurance and scientific analyses. To benefit from this, consistent use of SR is required and is therefore recommended. · SR-based data mining allows simple epidemiologic analyses for PE.. · The prevalence of PE differs between outpatients and inpatients.. · Central and bilateral PEs have an increased risk of right heart strain.. · Jorg T, Halfmann MC, Graafen D et al. Structured reporting for efficient epidemiological and in-hospital prevalence analysis of pulmonary embolisms. Fortschr Röntgenstr 2024; DOI 10.1055/a-2301-3349.

A novel reporting workflow for automated integration of artificial intelligence results into structured radiology reports.

OBJECTIVES: Artificial intelligence (AI) has tremendous potential to help radiologists in daily clinical routine. However, a seamless, standardized, and time-efficient way of integrating AI into the radiology workflow is often lacking. This constrains the full potential of this technology. To address this, we developed a new reporting pipeline that enables automated pre-population of structured reports with results provided by AI tools. METHODS: Findings from a commercially available AI tool for chest X-ray pathology detection were sent to an IHE-MRRT-compliant structured reporting (SR) platform as DICOM SR elements and used to automatically pre-populate a chest X-ray SR template. Pre-populated AI results could be validated, altered, or deleted by radiologists accessing the SR template. We assessed the performance of this newly developed AI to SR pipeline by comparing reporting times and subjective report quality to reports created as free-text and conventional structured reports. RESULTS: Chest X-ray reports with the new pipeline could be created in significantly less time than free-text reports and conventional structured reports (mean reporting times: 66.8 s vs. 85.6 s and 85.8 s, respectively; both p < 0.001). Reports created with the pipeline were rated significantly higher quality on a 5-point Likert scale than free-text reports (p < 0.001). CONCLUSION: The AI to SR pipeline offers a standardized, time-efficient way to integrate AI-generated findings into the reporting workflow as parts of structured reports and has the potential to improve clinical AI integration and further increase synergy between AI and SR in the future. CRITICAL RELEVANCE STATEMENT: With the AI-to-structured reporting pipeline, chest X-ray reports can be created in a standardized, time-efficient, and high-quality manner. The pipeline has the potential to improve AI integration into daily clinical routine, which may facilitate utilization of the benefits of AI to the fullest. KEY POINTS: • A pipeline was developed for automated transfer of AI results into structured reports. • Pipeline chest X-ray reporting is faster than free-text or conventional structured reports. • Report quality was also rated higher for reports created with the pipeline. • The pipeline offers efficient, standardized AI integration into the clinical workflow.

[Interdisciplinary case discussions]. / Interdisziplinäre Besprechungen.

BACKGROUND: Interdisciplinary case discussions, especially tumor conferences, represent a large part of the clinical radiologist's daily work. Radiology plays a key role in tumor conferences, since imaging findings have a direct influence on therapy decisions. METHODS AND OBJECTIVES: This article discusses the requirements for the radiologist in preparing and conducting tumor conferences. Furthermore, the general conditions and forms of implementation of tumor conferences will be highlighted. Information technology (IT) tools for process automation and systems for assessing the course of tumor diseases will be presented. RESULTS: Detailed preparation of tumor conferences and clear communication of findings is essential. The radiological expertise in tumor conferences often leads to changes or adjustments of initially planned therapies. In addition to traditional face-to-face meetings, hybrid solutions have become established for tumor conferences in which the core team is on site and other participants (external referring physicians, internal participants outside the core team) are connected via video conference. Various systems have been established for assessing the course of tumor diseases. Due to its broad applicability, RECIST 1.1. is the most widely used. IT tools enable previously marked lesions to be displayed over time in a matrix view (lesion tracking). Artificial intelligence (AI) can also be used to automatically detect lesions and assess their volumes. CONCLUSION: Preparing and conducting tumor conferences is time-consuming for radiologists. IT tools can automate and thus facilitate the processes. Hybrid solutions combining face-to-face meetings and video conferences make it easier for external referring physicians to present their patients in tumor conferences.

[Image interpretation and the radiological report]. / Befundkommunikation.

BACKGROUND: The radiological report is the cornerstone of communication between radiologists and referring physicians and patients, respectively. The report is comprised of image interpretation on the one hand and communication of this interpretation on the other hand. OBJECTIVES AND METHODS: To outline different types of radiological reports (regarding content as well as structure) and their communication. To this end, current guidelines are summarized and clinical examples are presented. RESULTS: The radiological report is typically a written piece of free text prose and highly individualized regarding its quality, precision, and structure. In order to improve the understanding of the written report, additional material (e.g., annotations, images, tables) can be supplemented (multimedia-enhanced reporting). In terms of standardization, national and international radiological associations promote structured reporting in radiology. However, this is not without issues. CONCLUSION: Effective communication should improve patient care and it should be clear and provided in a timely manner. As communication in clinical reality is often hampered by various factors, internal standard operating procedures (SOPs) should be developed to improve communication workflows. to improve communication procedures.

[Validation and implementation of artificial intelligence in radiology : Quo vadis in 2022?] / Validierung und Implementierung von künstlicher Intelligenz in der radiologischen Versorgung : Quo vadis im Jahr 2022?

BACKGROUND: The hype around artificial intelligence (AI) in radiology continues and the number of approved AI tools is growing steadily. Despite the great potential, integration into clinical routine in radiology remains limited. In addition, the large number of individual applications poses a challenge for clinical routine, as individual applications have to be selected for different questions and organ systems, which increases the complexity and time required. OBJECTIVES: This review will discuss the current status of validation and implementation of AI tools in clinical routine, and identify possible approaches for an improved assessment of the generalizability of results of AI tools. MATERIALS AND METHODS: A literature search in various literature and product databases as well as publications, position papers, and reports from various stakeholders was conducted for this review. RESULTS: Scientific evidence and independent validation studies are available for only a few commercial AI tools and the generalizability of the results often remains questionable. CONCLUSIONS: One challenge is the multitude of offerings for individual, specific application areas by a large number of manufacturers, making integration into the existing site-specific IT infrastructure more difficult. Furthermore, remuneration for the use of AI tools in clinical routine by health insurance companies in Germany is lacking. But in order for reimbursement to be granted, the clinical utility of new applications must first be proven. Such proof, however, is lacking for most applications.

Efficient structured reporting in radiology using an intelligent dialogue system based on speech recognition and natural language processing.

BACKGROUND: Structured reporting (SR) is recommended in radiology, due to its advantages over free-text reporting (FTR). However, SR use is hindered by insufficient integration of speech recognition, which is well accepted among radiologists and commonly used for unstructured FTR. SR templates must be laboriously completed using a mouse and keyboard, which may explain why SR use remains limited in clinical routine, despite its advantages. Artificial intelligence and related fields, like natural language processing (NLP), offer enormous possibilities to facilitate the imaging workflow. Here, we aimed to use the potential of NLP to combine the advantages of SR and speech recognition. RESULTS: We developed a reporting tool that uses NLP to automatically convert dictated free text into a structured report. The tool comprises a task-oriented dialogue system, which assists the radiologist by sending visual feedback if relevant findings are missed. The system was developed on top of several NLP components and speech recognition. It extracts structured content from dictated free text and uses it to complete an SR template in RadLex terms, which is displayed in its user interface. The tool was evaluated for reporting of urolithiasis CTs, as a use case. It was tested using fictitious text samples about urolithiasis, and 50 original reports of CTs from patients with urolithiasis. The NLP recognition worked well for both, with an F1 score of 0.98 (precision: 0.99; recall: 0.96) for the test with fictitious samples and an F1 score of 0.90 (precision: 0.96; recall: 0.83) for the test with original reports. CONCLUSION: Due to its unique ability to integrate speech into SR, this novel tool could represent a major contribution to the future of reporting.

Structured reporting in radiology enables epidemiological analysis through data mining: urolithiasis as a use case.

PURPOSE: To investigate the epidemiology and distribution of disease characteristics of urolithiasis by data mining structured radiology reports. METHODS: The content of structured radiology reports of 2028 urolithiasis CTs was extracted from the department's structured reporting (SR) platform. The investigated cohort represented the full spectrum of a tertiary care center, including mostly symptomatic outpatients as well as inpatients. The prevalences of urolithiasis in general and of nephro- and ureterolithasis were calculated. The distributions of age, sex, calculus size, density and location, and the number of ureteral and renal calculi were calculated. For ureterolithiasis, the impact of calculus characteristics on the degree of possible obstructive uropathy was calculated. RESULTS: The prevalence of urolithiasis in the investigated cohort was 72%. Of those patients, 25% had nephrolithiasis, 40% ureterolithiasis, and 35% combined nephro- and ureterolithiasis. The sex distribution was 2.3:1 (M:F). The median patient age was 50 years (IQR 36-62). The median number of calculi per patient was 1. The median size of calculi was 4 mm, and the median density was 734 HU. Of the patients who suffered from ureterolithiasis, 81% showed obstructive uropathy, with 2nd-degree uropathy being the most common. Calculus characteristics showed no impact on the degree of obstructive uropathy. CONCLUSION: SR-based data mining is a simple method by which to obtain epidemiologic data and distributions of disease characteristics, for the investigated cohort of urolithiasis patients. The added information can be useful for multiple purposes, such as clinical quality assurance, radiation protection, and scientific or economic investigations. To benefit from these, the consistent use of SR is mandatory. However, in clinical routine SR usage can be elaborate and requires radiologists to adapt.

Implementation of structured reporting in clinical routine: a review of 7 years of institutional experience.

BACKGROUND: To evaluate the implementation process of structured reporting (SR) in a tertiary care institution over a period of 7 years. METHODS: We analysed the content of our image database from January 2016 to December 2022 and compared the numbers of structured reports and free-text reports. For the ten most common SR templates, usage proportions were calculated on a quarterly basis. Annual modality-specific SR usage was calculated for ultrasound, CT, and MRI. During the implementation process, we surveyed radiologists and clinical referring physicians concerning their views on reporting in radiology. RESULTS: As of December 2022, our reporting platform contained more than 22,000 structured reports. Use of the ten most common SR templates increased markedly since their implementation, leading to a mean SR usage of 77% in Q4 2022. The highest percentages of SR usage were shown for trauma CT, focussed assessment with ultrasound for trauma (FAST), and prostate MRI: 97%, 95%, and 92%, respectively, in 2022. Overall modality-specific SR usage was 17% for ultrasound, 13% for CT, and 6% for MRI in 2022. Both radiologists and referring physicians were more satisfied with structured reports and rated SR better than free-text reporting (FTR) on various attributes. CONCLUSIONS: The increasing SR usage during the period under review and the positive attitude towards SR among both radiologists and clinical referrers show that SR can be successfully implemented. We therefore encourage others to take this step in order to benefit from the advantages of SR. KEY POINTS: 1. Structured reporting usage increased markedly since its implementation at our institution in 2016. 2. Mean usage for the ten most popular structured reporting templates was 77% in 2022. 3. Both radiologists and referring physicians preferred structured reports over free-text reports. 4. Our data shows that structured reporting can be successfully implemented. 5. We strongly encourage others to implement structured reporting at their institutions.

Natural language processing for automatic evaluation of free-text answers - a feasibility study based on the European Diploma in Radiology examination.

BACKGROUND: Written medical examinations consist of multiple-choice questions and/or free-text answers. The latter require manual evaluation and rating, which is time-consuming and potentially error-prone. We tested whether natural language processing (NLP) can be used to automatically analyze free-text answers to support the review process. METHODS: The European Board of Radiology of the European Society of Radiology provided representative datasets comprising sample questions, answer keys, participant answers, and reviewer markings from European Diploma in Radiology examinations. Three free-text questions with the highest number of corresponding answers were selected: Questions 1 and 2 were "unstructured" and required a typical free-text answer whereas question 3 was "structured" and offered a selection of predefined wordings/phrases for participants to use in their free-text answer. The NLP engine was designed using word lists, rule-based synonyms, and decision tree learning based on the answer keys and its performance tested against the gold standard of reviewer markings. RESULTS: After implementing the NLP approach in Python, F1 scores were calculated as a measure of NLP performance: 0.26 (unstructured question 1, n = 96), 0.33 (unstructured question 2, n = 327), and 0.5 (more structured question, n = 111). The respective precision/recall values were 0.26/0.27, 0.4/0.32, and 0.62/0.55. CONCLUSION: This study showed the successful design of an NLP-based approach for automatic evaluation of free-text answers in the EDiR examination. Thus, as a future field of application, NLP could work as a decision-support system for reviewers and support the design of examinations being adjusted to the requirements of an automated, NLP-based review process. CLINICAL RELEVANCE STATEMENT: Natural language processing can be successfully used to automatically evaluate free-text answers, performing better with more structured question-answer formats. Furthermore, this study provides a baseline for further work applying, e.g., more elaborated NLP approaches/large language models. KEY POINTS: • Free-text answers require manual evaluation, which is time-consuming and potentially error-prone. • We developed a simple NLP-based approach - requiring only minimal effort/modeling - to automatically analyze and mark free-text answers. • Our NLP engine has the potential to support the manual evaluation process. • NLP performance is better on a more structured question-answer format.

Pediatric computed tomography doses in Germany from 2016 to 2018 based on large-scale data collection.

PURPOSE: Accumulating evidence from epidemiological studies that pediatric computed tomography (CT) examinations can be associated with a small but non-zero excess risk for developing leukemia or brain tumor highlights the need to optimize doses of pediatric CT procedures. Mandatory dose reference levels (DRL) can support reduction of collective dose from CT imaging. Regular surveys of applied dose-related parameters are instrumental to decide when technological advances and optimized protocol design allow lower doses without sacrificing image quality. Our aim was to collect dosimetric data to support adapting current DRL to changing clinical practice. METHOD: Dosimetric data and technical scan parameters from common pediatric CT examinations were retrospectively collected directly from Picture Archiving and Communication Systems (PACS), Dose Management Systems (DMS), and Radiological Information Systems (RIS). RESULTS: We collected data from 17 institutions on 7746 CT series from the years 2016 to 2018 from examinations of the head, thorax, abdomen, cervical spine, temporal bone, paranasal sinuses and knee in patients below 18 years of age. Most of the age-stratified parameter distributions were lower than distributions from previously-analyzed data from before 2010. Most of the third quartiles were lower than German DRL at the time of the survey. CONCLUSIONS: Directly interfacing PACS, DMS, and RIS installations allows large-scale data collection but relies on high data-quality at the documentation stage. Data should be validated by expert knowledge or guided questionnaires. Observed clinical practice in pediatric CT imaging suggests lowering some DRL in Germany is reasonable.

Computed tomographies and cancer risk in children: a literature overview of CT practices, risk estimations and an epidemiologic cohort study proposal.

Radiation protection is a topic of great public concern and of many scientific investigations, because ionizing radiation is an established risk factor for leukaemia and many solid tumours. Exposure of the public to ionizing radiation includes exposure to background radiation, as well as medical and occupational exposures. A large fraction of the exposure from diagnostic procedures comes from medical imaging. Computed tomography (CT) is the major single contributor of diagnostic radiation exposure. An increase in the use of CTs has been reported over the last decades in many countries. Children have smaller bodies and lower shielding capacities, factors that affect the individual organ doses due to medical imaging. Several risk models have been applied to estimate the cancer burden caused by ionizing radiation from CT. All models predict higher risks for cancer among children exposed to CT as compared to adults. However, the cancer risk associated with CT has not been assessed directly in epidemiological studies. Here, plans are described to conduct an historical cohort study to investigate the cancer incidence in paediatric patients exposed to CT before the age of 15 in Germany. Patients will be recruited from radiology departments of several hospitals. Their individual exposure will be recorded, and time-dependent cumulative organ doses will be calculated. Follow-up for cancer incidence via the German Childhood Cancer Registry will allow computation of standardized incidence ratios using population-based incidence rates for childhood cancer. Dose-response modelling and analyses for subgroups of children based on the indication for and the result of the CT will be performed.