Potential of GPT-4 for Detecting Errors in Radiology Reports: Implications for Reporting Accuracy.

Background Errors in radiology reports may occur because of resident-to-attending discrepancies, speech recognition inaccuracies, and large workload. Large language models, such as GPT-4 (ChatGPT; OpenAI), may assist in generating reports. Purpose To assess effectiveness of GPT-4 in identifying common errors in radiology reports, focusing on performance, time, and cost-efficiency. Materials and Methods In this retrospective study, 200 radiology reports (radiography and cross-sectional imaging [CT and MRI]) were compiled between June 2023 and December 2023 at one institution. There were 150 errors from five common error categories (omission, insertion, spelling, side confusion, and other) intentionally inserted into 100 of the reports and used as the reference standard. Six radiologists (two senior radiologists, two attending physicians, and two residents) and GPT-4 were tasked with detecting these errors. Overall error detection performance, error detection in the five error categories, and reading time were assessed using Wald χ2 tests and paired-sample t tests. Results GPT-4 (detection rate, 82.7%;124 of 150; 95% CI: 75.8, 87.9) matched the average detection performance of radiologists independent of their experience (senior radiologists, 89.3% [134 of 150; 95% CI: 83.4, 93.3]; attending physicians, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; residents, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; P value range, .522-.99). One senior radiologist outperformed GPT-4 (detection rate, 94.7%; 142 of 150; 95% CI: 89.8, 97.3; P = .006). GPT-4 required less processing time per radiology report than the fastest human reader in the study (mean reading time, 3.5 seconds ± 0.5 [SD] vs 25.1 seconds ± 20.1, respectively; P < .001; Cohen d = -1.08). The use of GPT-4 resulted in lower mean correction cost per report than the most cost-efficient radiologist ($0.03 ± 0.01 vs $0.42 ± 0.41; P < .001; Cohen d = -1.12). Conclusion The radiology report error detection rate of GPT-4 was comparable with that of radiologists, potentially reducing work hours and cost. © RSNA, 2024 See also the editorial by Forman in this issue.

GPT-4 Analysis of MRI Reports in Suspected Myocarditis: A Multicenter Study.

PURPOSE: Diagnosing myocarditis relies on multimodal data including magnetic resonance imaging (MRI), clinical symptoms, and blood values. The correct interpretation and integration of MRI findings requires radiological expertise and knowledge. We aimed to investigate the performance of Generative Pre-trained Transformer 4 (GPT-4), a large language model, for report-based medical decision-making in the context of cardiac MRI for suspected myocarditis. METHODS: This retrospective study includes MRI reports from 396 patients with suspected myocarditis and eight centers, respectively. MRI reports and patient data including blood values, age, and further clinical information were provided to GPT-4 and to radiologists with 1 (Resident 1), 2 (Resident 2), and 4 years (Resident 3) of experience in cardiovascular MRI and knowledge of the 2018 Lake Louise Criteria. The final impression of the report regarding the radiological assessment of whether myocarditis is present or not was not provided. The performance of GPT-4 and of the human readers were compared to a consensus reading (two board-certified radiologists with 8 and 10 years of experience in cardiovascular MRI). Sensitivity, specificity, and accuracy were calculated. RESULTS: GPT-4 yielded an accuracy of 83%, sensitivity of 90%, and specificity of 78%, which was comparable to the physician with 1 year of experience (R1: 86%, 90%, 84%, p=.14) and lower than that of more experienced physicians (R2: 89%, 86%, 91%, p=.007 and R3: 91%, 85%, 96%, p<.001). GPT-4 and human readers showed a higher diagnostic performance when results from T1- and T2-mapping sequences were part of the reports, for Residents 1 and Resident 3 with statistical significance (p=.004 and p=.02, respectively). CONCLUSION: GPT-4 yielded good accuracy for diagnosing myocarditis based on MRI reports in a large dataset from multiple centers and therefore holds the potential to serve as a diagnostic decision supporting tool in this capacity, particularly for less experienced physicians. Further studies are required to explore the full potential and elucidate educational aspects of the integration of large language models in medical decision-making.

GPT-4 for Automated Determination of Radiological Study and Protocol based on Radiology Request Forms: A Feasibility Study.

Published under a CC BY 4.0 license.

AI support for accurate and fast radiological diagnosis of COVID-19: an international multicenter, multivendor CT study.

OBJECTIVES: Differentiation between COVID-19 and community-acquired pneumonia (CAP) in computed tomography (CT) is a task that can be performed by human radiologists and artificial intelligence (AI). The present study aims to (1) develop an AI algorithm for differentiating COVID-19 from CAP and (2) evaluate its performance. (3) Evaluate the benefit of using the AI result as assistance for radiological diagnosis and the impact on relevant parameters such as accuracy of the diagnosis, diagnostic time, and confidence. METHODS: We included n = 1591 multicenter, multivendor chest CT scans and divided them into AI training and validation datasets to develop an AI algorithm (n = 991 CT scans; n = 462 COVID-19, and n = 529 CAP) from three centers in China. An independent Chinese and German test dataset of n = 600 CT scans from six centers (COVID-19 / CAP; n = 300 each) was used to test the performance of eight blinded radiologists and the AI algorithm. A subtest dataset (180 CT scans; n = 90 each) was used to evaluate the radiologists' performance without and with AI assistance to quantify changes in diagnostic accuracy, reporting time, and diagnostic confidence. RESULTS: The diagnostic accuracy of the AI algorithm in the Chinese-German test dataset was 76.5%. Without AI assistance, the eight radiologists' diagnostic accuracy was 79.1% and increased with AI assistance to 81.5%, going along with significantly shorter decision times and higher confidence scores. CONCLUSION: This large multicenter study demonstrates that AI assistance in CT-based differentiation of COVID-19 and CAP increases radiological performance with higher accuracy and specificity, faster diagnostic time, and improved diagnostic confidence. KEY POINTS: • AI can help radiologists to get higher diagnostic accuracy, make faster decisions, and improve diagnostic confidence. • The China-German multicenter study demonstrates the advantages of a human-machine interaction using AI in clinical radiology for diagnostic differentiation between COVID-19 and CAP in CT scans.

Feasibility of artificial intelligence-supported assessment of bone marrow infiltration using dual-energy computed tomography in patients with evidence of monoclonal protein - a retrospective observational study.

OBJECTIVES: To demonstrate the feasibility of an automated, non-invasive approach to estimate bone marrow (BM) infiltration of multiple myeloma (MM) by dual-energy computed tomography (DECT) after virtual non-calcium (VNCa) post-processing. METHODS: Individuals with MM and monoclonal gammopathy of unknown significance (MGUS) with concurrent DECT and BM biopsy between May 2018 and July 2020 were included in this retrospective observational study. Two pathologists and three radiologists reported BM infiltration and presence of osteolytic bone lesions, respectively. Bone mineral density (BMD) was quantified CT-based by a CE-certified software. Automated spine segmentation was implemented by a pre-trained convolutional neural network. The non-fatty portion of BM was defined as voxels > 0 HU in VNCa. For statistical assessment, multivariate regression and receiver operating characteristic (ROC) were conducted. RESULTS: Thirty-five patients (mean age 65 ± 12 years; 18 female) were evaluated. The non-fatty portion of BM significantly predicted BM infiltration after adjusting for the covariable BMD (p = 0.007, r = 0.46). A non-fatty portion of BM > 0.93% could anticipate osteolytic lesions and the clinical diagnosis of MM with an area under the ROC curve of 0.70 [0.49-0.90] and 0.71 [0.54-0.89], respectively. Our approach identified MM-patients without osteolytic lesions on conventional CT with a sensitivity and specificity of 0.63 and 0.71, respectively. CONCLUSIONS: Automated, AI-supported attenuation assessment of the spine in DECT VNCa is feasible to predict BM infiltration in MM. Further, the proposed method might allow for pre-selecting patients with higher pre-test probability of osteolytic bone lesions and support the clinical diagnosis of MM without pathognomonic lesions on conventional CT. KEY POINTS: • The retrospective study provides an automated approach for quantification of the non-fatty portion of bone marrow, based on AI-supported spine segmentation and virtual non-calcium dual-energy CT data. • An increasing non-fatty portion of bone marrow is associated with a higher infiltration determined by invasive biopsy after adjusting for bone mineral density as a control variable (p = 0.007, r = 0.46). • The non-fatty portion of bone marrow might support the clinical diagnosis of multiple myeloma when conventional CT images are negative (sensitivity 0.63, specificity 0.71).

Two-dimensional CT measurements enable assessment of body composition on head and neck CT.

OBJECTIVES: The aim of this study was to evaluate whether simple 2D measurements in axial slices of head and neck CT examinations correlate with generally established measurements of body composition in abdominal CT at the height of the third lumbar vertebra and thus allow for an estimation of muscle and fat masses. METHODS: One hundred twenty-two patients who underwent concurrent CT of the head and neck and the abdomen between July 2016 and July 2020 were retrospectively included. For a subset of 30 patients, additional bioelectrical impedance analysis (BIA) was available. Areas of paraspinal muscles at the height of the third (C3) and fifth cervical vertebrae (C5) as well as the total cross-sectional area at the height of C3 and at the submandibular level were correlated with the results of abdominal measurements and BIA. Furthermore, intra- and interreader variabilities of all measurements were assessed. RESULTS: Regarding adipose tissue, good correlations were found between the total cross-sectional area of the patient's body at the submandibular level and at the height of C3 between both abdominal measurements and BIA results (r = 0.8-0.92; all p < 0.001). Regarding muscle, the total paraspinal muscle area at the height of C3 and C5 showed strong correlations with abdominal measurements and moderate to strong correlations with BIA results (r = 0.44-0.80; all p < 0.001), with the muscle area on C5 yielding slightly higher correlations. CONCLUSIONS: Body composition information can be obtained with comparable reliability from head and neck CT using simple biplanar measurements as from abdominal CT. KEY POINTS: • The total paraspinal muscle area at the height of C3 and C5 correlates strongly with abdominal muscle mass. • The total cross-sectional area at the submandibular level and at the height of C3 shows good correlations with abdominal fat mass. • The described measurements facilitate a rapid, opportunistic assessment of relevant body composition parameters.

Thrombus Enhancement Is a Predictor of Clinical Outcome in Acute Ischemic Stroke after Mechanical Thrombectomy.

BACKGROUND: Ex vivo computed tomography (CT) studies of artificial blood thrombi showed that contrast enhancement (CE) is determined by fibrin-content, while unenhanced density is associated with red blood cells. Thus, the present study investigates patient outcome in association with combined thrombus density measures in native and contrast-enhanced CT (CECT) of acute ischemic stroke patients. METHODS: This retrospective study includes 137 patients with M1 occlusions treated by mechanical thrombectomy (MT) between 2010 and 2016. Clinical outcome was determined with modified Rankin Scale (mRS) at 90 days. Differentiation of complete and incomplete large vessel occlusion (CLVO/ILVO) was based on CT and angiography. Two blinded readers classified blood thrombi based on native non-enhanced CT (NECT) as (a) hypo-, (b) iso-, and (c) hyperdense and in CECT angio measurements as (d) not-enhancing, (e) intermediate and (f) enhancing. To make sure that the mean is not represented in any of the maximum/minimum groups, thresholds in both cases were selected in a way that all values within one SD around the mean value form the isodense/intermediate group. In addition, the CE per se was correlated with the outcome. Correlations between imaging and clinical scales were performed with Spearman's Rho. For the group testing Pearson chi-square test, Mann-Whitney U, as well parametric and nonparametric one-factor ANOVA "Kruskal-Wallis" test including Bonferroni correction for multiple tests ware used. RESULTS: Twenty-three patients with ILVO (16.8%) differed significantly from patients with CLVO in mRS at admission (median 4 vs. 5) and after 90 days (median 1 vs. 4; p < 0.05) and thus were excluded. In the ILVO cohort, the classification according to NECT did not show statistical difference between hypo-, iso- and hyperdense CLVOs in regard to outcome. Classification of CLVOs according to CECT allowed an outcome prediction between the intermediate (median 3) and enhancing group (median 5) and between the enhancing and non-enhancing group (median 3; both p < 0.05) with a correlation of 291 between CE and higher mRS after 90 days (p < 0.005). CONCLUSIONS: CE of thrombi - especially in a range from over 18.4 to 40.35 Hounsfield Units - is an independent predictor of poor clinical outcome in patients undergoing MT due to acute middle cerebral artery occlusion.

Treatment of intracranial aneurysms with large-diameter (≥5.5 mm) Derivo Embolization Devices, with particular focus on 7 and 8 mm diameter devices.

OBJECTIVE: There are few studies on flow diverters with diameters >5 mm. We present our preliminary experience with the 5.5-8 mm Derivo Embolization Device (DED) implants for the treatment of cerebral aneurysms. METHODS: A consecutive series of 26 patients (median age: 60 years) treated for 32 aneurysms in 26 procedures was retrospectively analyzed for procedural characteristics, complications, and mid-term angiographic results. RESULTS: The median aneurysm size was 10.5 mm, 2 of 30 (6%) aneurysms were ruptured and 9 (28%) had a fusiform or dissecting morphology. DED implantation was performed in the internal carotid artery in 18 of 26 (69%) procedures and in the vertebrobasilar artery in 8 (31%). Device deployment failed in 1 (4%) procedure. The 7 and 8 mm implants were successfully deployed in 5 cases. Additional balloon angioplasty or stent implantation was performed in 3 (12%) cases to improve wall apposition. Complications included 1 (4%) major stroke and 2 (8%) minor strokes. Angiographic follow up at a mean of 6 months showed complete occlusion in 8 of30 (27%) aneurysms and favorable occlusion in 14 (47%). CONCLUSIONS: The use of large diameter DEDs was safe and feasible. The mid-term occlusion rates are acceptable considering the complex subset of aneurysms studied. Further studies are warranted to define the indications for large-diameter DEDs and to evaluate their long-term efficacy.

Long-term Safety and Efficacy of the Derivo Embolization Device in a Single-center Series.

PURPOSE: This study analyzes the long-term clinical and angiographic outcomes of the Derivo Embolization Device (DED), an advanced flow diverter device with an electropolished surface, for the treatment of intracranial aneurysms. METHODS: A consecutive series of 101 patients (mean age: 58 years, 72% female) treated with the DED for 122 aneurysms at a single center between 2017 and 2023 was retrospectively analyzed for major (change in National Institutes of Health Stroke Scale [NIHSS] score ≥ 4 points) and minor (change in NIHSS score < 4 points) neurological events, procedural morbidity (increase of at least one point on the modified Rankin Scale), and angiographic results. RESULTS: There were 14 (11%) recurrent aneurysms, 15 (12%) ruptured aneurysms, 26 (21%) posterior circulation aneurysms and 16 (13%) fusiform or dissecting aneurysms. Device deployment failed in 1 case (1%). Procedure-related symptomatic procedural complications consisted of 2 (2%) major events (1 major stroke and 1 vessel perforation with intracranial hemorrhage and infarction) and 6 minor events (6 minor strokes). Procedural morbidity was 5%. There were no late ischemic or hemorrhagic events during follow-up. Complete and favorable aneurysm occlusion was achieved in 54% (40/74) and 62% (46/74) at a mean of 5 months, 71% (27/38) and 87% (33/38) at a mean of 12 months, and 76% (25/33) and 97% (32/33) at a mean of 35 months, respectively. CONCLUSION: The results demonstrate progressive aneurysm occlusion beyond 12 months after DED implantation with an almost 100% favorable occlusion rate. Procedural morbidity was low and there were no late complications.

Pipeline Vantage Embolization Device for the treatment of intracranial aneurysms: A systematic review and meta-analysis.

OBJECTIVE: The Pipeline Vantage Embolization Device (PVED) is a novel coated flow diverter with reduced wire diameters to improve neoendothelialization and stent porosity. This systematic review evaluates the safety and efficacy of the PVED based on the current literature. METHODS: Following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, a thorough literature search was conducted using PubMed, EMBASE, and Cochrane. The random effects model was used to calculate estimates with major neurological complications within 30 days of treatment as the primary safety endpoint and ≤1-year complete occlusion rate as the primary efficacy endpoint. RESULTS: Six single-arm studies (5 retrospective, 1 prospective) with 392 patients and 439 aneurysms (6.8% ruptured) were included. Antiplatelet regimens varied, but dual antiplatelet therapy was administered in the majority. The pooled technical success rate was 99.0% (95%CI, 98.0%-100%) with an average of 1.2 devices implanted per procedure. Balloon angioplasty was performed in 17.0% (95%CI, 6.4-27.6%) and adjunctive coiling in 28.0% (95%CI, 17.8-38.2%), with significant heterogeneity for both variables. Pooled estimates for major neurological complications were 3.5% (95%CI, 1.7%-5.2%) with total ischemic events in 4.1% (95% CI, 1.6%-6.6%) and hemorrhagic events in 1.0% (95% CI, 0.0%-1.9%). The rate of complete angiographic occlusion was 75.7% (95%CI, 70.7%-80.6%) at a mean follow-up of 7 months, with in-stent stenoses in 8.1% (95%CI, 4.5%-11.8%). CONCLUSIONS: The safety and efficacy profile of the PVED appears comparable to competing devices, with potentially fewer complications than first-generation flow diverters. Long-term and comparative studies are needed to further confirm these results.

Feasibility, Safety, and Efficacy of Endovascular vs. Surgical Treatment of Unruptured Multi-Sac Intracranial Aneurysms in a Single-Center Retrospective Series.

PURPOSE: Multi-sac aneurysms (MSAs) are not uncommon, but studies on their management are scarce. This study aims to evaluate and compare the feasibility, safety, and efficacy of MSAs treated with either clipping or coiling after interdisciplinary case discussion at our center. MATERIALS AND METHODS: We retrospectively analyzed MSAs treated by microsurgical clipping, coiling, or stent-assisted coiling (SAC). Treatment modalities, complications, angiographic results, and clinical outcomes were evaluated. Major neurological events were defined as a safety endpoint and complete occlusion as an efficacy endpoint. RESULTS: Ninety patients (mean age, 53.2±11.0 years; 73 [81.1%] females) with MSAs met our inclusion criteria (clipping, 50; coiling, 19; SAC, 21). Most aneurysms were located in the middle cerebral artery (48.9%). All clipping procedures were technically successful, but endovascular treatment failed in 1 coiling case, and a switch from coiling to SAC was required in 2 cases. The major event rates were 4.0% after clipping (1 major stroke and 1 intracranial hemorrhage) and 0% after endovascular therapy (P=0.667). At mid-term angiographic follow-up (mean 12.0±8.9 months), all 37 followed clipped aneurysms were completely occluded, compared to 8/17 (41.7%) after coiling and 11/15 (73.3%) after SAC (P<0.001). Coiling was significantly associated with incomplete occlusion in the adjusted analysis (odds ratio, 11.7; 95% confidence interval, 2.7-52.6; P=0.001). CONCLUSION: Both endovascular and surgical treatment were feasible and safe for MSAs. As coiling was associated with comparatively high recanalization rates, endovascular treatment may be preferred with stent support.

Aneurysm Treatment With the Pipeline Vantage Embolization Device in Retrospective Evaluation: Periprocedural Results from the Pipe-VADER Study.

OBJECTIVE: The Pipeline Vantage Embolization Device is a fourth-generation flow diverter with an antithrombotic coating and a reduced profile compared to previous Pipeline versions. The objective of this study was to evaluate the procedural feasibility, safety, and efficacy of this device. METHODS: The Pipe-VADER study was designed as a retrospective, observational study of consecutive patients treated with the Vantage at 3 neurovascular centers. Patient and aneurysm characteristics, procedural parameters, early complications, and extent of postinterventional contrast retention were analyzed on an intention-to-treat basis. RESULTS: Twenty-eight patients with 31 aneurysms (median size: 5.0 mm, posterior circulation: 4 [12.9%], ruptured: 5 [16.1%]) were included. The technical success rate was 100%, with multiple stents used in 4/30 (13.3%) procedures. Of the 30 procedures, adjunctive coiling was performed in 3 (10.0%) and balloon angioplasty in 2 (6.7%). Median procedure time was 62 minutes. Procedural ischemic stroke occured in 4 (13.3%) cases, whereof 2 were major strokes (6.6%). There were no hemorrhagic complications. Initial contrast retention was observed in 29/31 (93.5%) aneurysms. All 27 overstented side vessels were patent at the end of the procedure. Short-term follow-up (median: 5 months) showed complete and favorable occlusion rates of 70% (14/20) and 80% (16/20), respectively. CONCLUSIONS: The new Pipeline Vantage appears to be safe and feasible for the treatment of intracranial aneurysms and warrants further evaluation.

ChatGPT yields low accuracy in determining LI-RADS scores based on free-text and structured radiology reports in German language.

Background: To investigate the feasibility of the large language model (LLM) ChatGPT for classifying liver lesions according to the Liver Imaging Reporting and Data System (LI-RADS) based on MRI reports, and to compare classification performance on structured vs. unstructured reports. Methods: LI-RADS classifiable liver lesions were included from German written structured and unstructured MRI reports with report of size, location, and arterial phase contrast enhancement as minimum inclusion requirements. The findings sections of the reports were propagated to ChatGPT (GPT-3.5), which was instructed to determine LI-RADS scores for each classifiable liver lesion. Ground truth was established by two radiologists in consensus. Agreement between ground truth and ChatGPT was assessed with Cohen's kappa. Test-retest reliability was assessed by passing a subset of n = 50 lesions five times to ChatGPT, using the intraclass correlation coefficient (ICC). Results: 205 MRIs from 150 patients were included. The accuracy of ChatGPT at determining LI-RADS categories was poor (53% and 44% on unstructured and structured reports). The agreement to the ground truth was higher (k = 0.51 and k = 0.44), the mean absolute error in LI-RADS scores was lower (0.5 ± 0.5 vs. 0.6 ± 0.7, p < 0.05), and the test-retest reliability was higher (ICC = 0.81 vs. 0.50), in free-text compared to structured reports, respectively, although structured reports comprised the minimum required imaging features significantly more frequently (Chi-square test, p < 0.05). Conclusions: ChatGPT attained only low accuracy when asked to determine LI-RADS scores from liver imaging reports. The superior accuracy and consistency throughout free-text reports might relate to ChatGPT's training process. Clinical relevance statement: Our study indicates both the necessity of optimization of LLMs for structured clinical data input and the potential of LLMs for creating machine-readable labels based on large free-text radiological databases.

Immediate angiographic control after intra-arterial nimodipine administration underestimates the vasodilatory effect.

Intra-arterial nimodipine administration is a widely used rescue therapy for cerebral vasospasm. Although it is known that its effect sets in with delay, there is little evidence in current literature. Our aim was to prove that the maximal vasodilatory effect is underestimated in direct angiographic controls. We reviewed all cases of intra-arterial nimodipine treatment for subarachnoid hemorrhage-related cerebral vasospasm between January 2021 and December 2022. Inclusion criteria were availability of digital subtraction angiography runs before and after nimodipine administration and a delayed run for the most affected vessel at the end of the procedure to decide on further escalation of therapy. We evaluated nimodipine dose, timing of administration and vessel diameters. Delayed runs were performed in 32 cases (19 patients) with a mean delay of 37.6 (± 16.6) min after nimodipine administration and a mean total nimodipine dose of 4.7 (± 1.2) mg. Vessel dilation was more pronounced in delayed vs. immediate controls, with greater changes in spastic vessel segments (n = 31: 113.5 (± 78.5%) vs. 32.2% (± 27.9%), p < 0.0001) vs. non-spastic vessel segments (n = 32: 23.1% (± 13.5%) vs. 13.3% (± 10.7%), p < 0.0001). In conclusion intra-arterially administered nimodipine seems to exert a delayed vasodilatory effect, which should be considered before escalation of therapy.

The worldwide COVID-19 pandemic caused a decline in sonographic examinations - is this a continuing trend?

PURPOSE: Due to the increasing number of COVID-19 infections since spring 2020 the patient care workflow underwent changes in Germany. To minimize face-to-face exposure and reduce infection risk, non-time-critical elective medical procedures were postponed. Since ultrasound examinations include non-time-critical elective examinations and often can be substituted by other imaging modalities not requiring direct patient contact, the number of examinations has declined significantly. The aim of this study is to quantify the baseline number of ultrasound examinations in the years before, during, and in the early post-pandemic period of the COVID-19 pandemic (since January 2015 to September 2023), and to measure the number of examinations at different German university hospitals. MATERIALS AND METHODS: The number of examinations was assessed based on a web-based database at all participating clinics at the indicated time points. RESULTS: N = 288 562 sonographic examinations from four sites were included in the present investigation. From January 2020 to June 2020, a significantly lower number of examinations of n = 591.21 vs. 698.43 (p = 0.01) per month and included center was performed. Also, excluding the initial pandemic period until June 2020, significantly fewer ultrasound examinations were performed compared to pre-pandemic years 648.1 vs. 698.4 (p < 0.05), per month and included center, while here differences between the individual centers were observed. In the late phase of the pandemic (n = 681.96) and in the post-pandemic phase (as defined by the WHO criteria from May 2023; n = 739.95), the number of sonographic examinations returned to pre-pandemic levels. CONCLUSION: The decline in the number of sonographic examinations caused by the COVID-19 pandemic was initially largely intentional and can be illustrated quantitatively. After an initial abrupt decline in sonographic examinations, the pre-pandemic levels could not be reached for a long time, which could be due to restructuring of patient care and follow-up treatment. In the post-pandemic phase, the pre-pandemic level has been achieved again. The reasons for a prolonged reduction in ultrasound examinations are discussed in this article. KEY POINTS: · During the pandemic, significantly fewer ultrasound examinations were performed in the included centers.. · The number of examinations could not be reach the pre-pandemic level for a long time, which could be due to restructuring of patient care and follow-up treatment.. · Identifying causes for sonographic exam reduction is crucial in pandemic preparedness to uphold healthcare quality and continuity for all patients.. · The prolonged decline in sonographic examinations during the pandemic does not represent a lasting trend, as evidenced by the return to pre-pandemic levels..

Multicenter Experience with the Pipeline Flex and Vantage with Shield Technology for Intracranial Aneurysm Treatment.

BACKGROUND AND PURPOSE: The Pipeline Embolization Device is a safe and effective treatment option for intracranial aneurysms. The newer Pipeline generations have received structural refinements and a surface modification to improve deliverability, procedural safety, and angiographic outcomes. This multicenter study evaluated the clinical safety and efficacy of the 2 surface-modified Pipeline iterations, Pipeline Vantage and Pipeline Flex with Shield Technology (PFS). MATERIALS AND METHODS: Consecutive patients treated between 2017 and 2023 were retrospectively reviewed for aneurysm characteristics, procedural details, complications, and angiographic outcomes. The safety end point was the rate of procedural and postprocedural major neurologic events occurring during the hospital stay. The efficacy end point was the rate of complete occlusion at last follow-up. RESULTS: One hundred forty-one patients underwent 112 Pipeline Vantage procedures and 32 PFS procedures for 147 aneurysms with a mean size of 8.0 (SD, 5.9) mm (11% ruptured, 16% posterior circulation, 18% nonsaccular morphology). All procedures were technically successful with a mean of 1.2 devices implanted. Balloon angioplasty was required in 20/144 (13.9%) procedures. Major neurologic adverse events occurred in 6/144 (4.2%) procedures (all ischemic stroke), resulting in death in 2 (1.4%) patients. There were no hemorrhagic complications. At a mean of 11 months, complete occlusion was achieved in 85/112 (75.9%) aneurysms, 15/112 (13.4%) had an entry remnant, and 12/112 (10.7%) had an aneurysm remnant. CONCLUSIONS: The results demonstrate high feasibility, procedural safety, and efficacy of the surface-modified Pipeline flow diverters.

Safety and efficacy of coated flow diverters in the treatment of ruptured intracranial aneurysms: a retrospective multicenter study.

BACKGROUND: This multicenter study evaluated the safety and efficacy of coated flow diverters (cFDs) for the treatment of ruptured intracranial aneurysms. METHODS: Consecutive patients treated with different cFDs for ruptured aneurysms under tirofiban at eight neurovascular centers between 2016 and 2023 were retrospectively analyzed. The majority of patients were loaded with dual antiplatelet therapy after the treatment. Aneurysm occlusion was determined using the O'Kelly-Marotta (OKM) grading scale. Primary outcome measures were major procedural complications and aneurysmal rebleeding during hospitalization. RESULTS: The study included 60 aneurysms (posterior circulation: 28 (47%)) with a mean size of 5.8±4.7 mm. Aneurysm morphology was saccular in 28 (47%), blister-like in 12 (20%), dissecting in 13 (22%), and fusiform in 7 (12%). Technical success was 100% with a mean of 1.1 cFDs implanted per aneurysm. Adjunctive coiling was performed in 11 (18%) aneurysms. Immediate contrast retention was observed in 45 (75%) aneurysms. There was 1 (2%) major procedural complication (a major stroke, eventually leading to death) and no aneurysmal rebleeding. A good outcome (modified Rankin Scale 0-2) was achieved in 40 (67%) patients. At a mean follow-up of 6 months, 27/34 (79%) aneurysms were completely occluded (OKM D), 3/34 (9%) had an entry remnant (OKM C), and 4/34 (12%) had residual filling (OKM A or B). There was 1 (3%) severe in-stent stenosis during follow-up that was treated with balloon angioplasty. CONCLUSIONS: Treatment of ruptured aneurysms with cFDs was reasonably safe and efficient and thus represents a valid treatment option, especially for complex cases.

Cooperative AI training for cardiothoracic segmentation in computed tomography: An iterative multi-center annotation approach.

PURPOSE: Radiological reporting is transitioning to quantitative analysis, requiring large-scale multi-center validation of biomarkers. A major prerequisite and bottleneck for this task is the voxelwise annotation of image data, which is time-consuming for large cohorts. In this study, we propose an iterative training workflow to support and facilitate such segmentation tasks, specifically for high-resolution thoracic CT data. METHODS: Our study included 132 thoracic CT scans from clinical practice, annotated by 13 radiologists. In three iterative training experiments, we aimed to improve and accelerate segmentation of the heart and mediastinum. Each experiment started with manual segmentation of 5-25 CT scans, which served as training data for a nnU-Net. Further iterations incorporated AI pre-segmentation and human correction to improve accuracy, accelerate the annotation process, and reduce human involvement over time. RESULTS: Results showed consistent improvement in AI model quality with each iteration. Resampled datasets improved the Dice similarity coefficients for both the heart (DCS 0.91 [0.88; 0.92]) and the mediastinum (DCS 0.95 [0.94; 0.95]). Our AI models reduced human interaction time by 50 % for heart and 70 % for mediastinum segmentation in the most potent iteration. A model trained on only five datasets achieved satisfactory results (DCS > 0.90). CONCLUSIONS: The iterative training workflow provides an efficient method for training AI-based segmentation models in multi-center studies, improving accuracy over time and simultaneously reducing human intervention. Future work will explore the use of fewer initial datasets and additional pre-processing methods to enhance model quality.