Radiology最新文献

{"title":"Dice the Slice: MRI and CT Segmentation in Radiology.","authors":"Felipe C Kitamura","doi":"10.1148/radiol.250143","DOIUrl":"https://doi.org/10.1148/radiol.250143","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e250143"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TotalSegmentator MRI: Robust Sequence-independent Segmentation of Multiple Anatomic Structures in MRI.","authors":"Tugba Akinci D'Antonoli, Lucas K Berger, Ashraya K Indrakanti, Nathan Vishwanathan, Jakob Weiss, Matthias Jung, Zeynep Berkarda, Alexander Rau, Marco Reisert, Thomas Küstner, Alexandra Walter, Elmar M Merkle, Daniel T Boll, Hanns-Christian Breit, Andrew Phillip Nicoli, Martin Segeroth, Joshy Cyriac, Shan Yang, Jakob Wasserthal","doi":"10.1148/radiol.241613","DOIUrl":"https://doi.org/10.1148/radiol.241613","url":null,"abstract":"<p><p>Background Since the introduction of TotalSegmentator CT, there has been demand for a similar robust automated MRI segmentation tool that can be applied across all MRI sequences and anatomic structures. Purpose To develop and evaluate an automated MRI segmentation model for robust segmentation of major anatomic structures independent of MRI sequence. Materials and Methods In this retrospective study, an nnU-Net model (TotalSegmentator MRI) was trained on MRI and CT scans to segment 80 anatomic structures relevant for use cases such as organ volumetry, disease characterization, surgical planning, and opportunistic screening. Images were randomly sampled from routine clinical studies to represent real-world examples. Dice scores were calculated between the predicted segmentations and expert radiologist segmentations to evaluate model performance on an internal test set and two external test sets and against two publicly available models and TotalSegmentator CT. The Wilcoxon signed rank test was used to compare model performance. The proposed model was applied to a separate internal dataset containing abdominal MRI scans to investigate age-dependent volume changes. Results A total of 1143 scans (616 MRI, 527 CT; median patient age, 61 years [IQR, 50-72 years]) were split into a training set (<i>n</i> = 1088; CT and MRI) and an internal test set (<i>n</i> = 55; MRI only). The two external test sets (AMOS and CHAOS) contained 20 MRI scans each, and the aging-study dataset contained 8672 abdominal MRI scans (median patient age, 59 years [IQR, 45-70 years]). The proposed model had a Dice score of 0.839 for the 80 anatomic structures in the internal test set and outperformed two other models (Dice score of 0.862 vs 0.759 for 40 anatomic structures and 0.838 vs 0.560 for 13 anatomic structures; <i>P</i> < .001 for both). On the TotalSegmentator CT test set (89 CT scans), the performance of the proposed model almost matched that of TotalSegmentator CT (Dice score, 0.966 vs 0.970; <i>P</i> < .001). The aging study demonstrated a strong correlation between age and organ volume (eg, age and liver volume: ρ = -0.096; <i>P</i> < .0001). Conclusion The proposed open-source, easy-to-use model allows for automatic, robust segmentation of 80 structures, extending the capabilities of TotalSegmentator to MRI scans from any MRI sequence. The ready-to-use online tool is available at <i>https://totalsegmentator.com</i>; the model, at <i>https://github.com/wasserth/TotalSegmentator</i>; and the dataset, at <i>http://zenodo.org/records/14710732</i>. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Kitamura in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e241613"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging Large Language Models to Generate Clinical Histories for Oncologic Imaging Requisitions.","authors":"Rajesh Bhayana, Omar Alwahbi, Aly Muhammad Ladak, Yangqing Deng, Adriano Basso Dias, Khaled Elbanna, Jorge Abreu Gomez, Ankush Jajodia, Kartik Jhaveri, Sarah Johnson, Dilkash Kajal, David Wang, Christine Soong, Ania Kielar, Satheesh Krishna","doi":"10.1148/radiol.242134","DOIUrl":"https://doi.org/10.1148/radiol.242134","url":null,"abstract":"<p><p>Background Clinical information improves imaging interpretation, but physician-provided histories on requisitions for oncologic imaging often lack key details. Purpose To evaluate large language models (LLMs) for automatically generating clinical histories for oncologic imaging requisitions from clinical notes and compare them with original requisition histories. Materials and Methods In total, 207 patients with CT performed at a cancer center from January to November 2023 and with an electronic health record clinical note coinciding with ordering date were randomly selected. A multidisciplinary team informed selection of 10 parameters important for oncologic imaging history, including primary oncologic diagnosis, treatment history, and acute symptoms. Clinical notes were independently reviewed to establish the reference standard regarding presence of each parameter. After prompt engineering with seven patients, GPT-4 (version 0613; OpenAI) was prompted on April 9, 2024, to automatically generate structured clinical histories for the 200 remaining patients. Using the reference standard, LLM extraction performance was calculated (recall, precision, F1 score). LLM-generated and original requisition histories were compared for completeness (proportion including each parameter), and 10 radiologists performed pairwise comparison for quality, preference, and subjective likelihood of harm. Results For the 200 LLM-generated histories, GPT-4 performed well, extracting oncologic parameters from clinical notes (F1 = 0.983). Compared with original requisition histories, LLM-generated histories more frequently included parameters critical for radiologist interpretation, including primary oncologic diagnosis (99.5% vs 89% [199 and 178 of 200 histories, respectively]; <i>P</i> < .001), acute or worsening symptoms (15% vs 4% [29 and seven of 200]; <i>P</i> < .001), and relevant surgery (61% vs 12% [122 and 23 of 200]; <i>P</i> < .001). Radiologists preferred LLM-generated histories for imaging interpretation (89% vs 5%, 7% equal; <i>P</i> < .001), indicating they would enable more complete interpretation (86% vs 0%, 15% equal; <i>P</i> < .001) and have a lower likelihood of harm (3% vs 55%, 42% neither; <i>P</i> < .001). Conclusion An LLM enabled accurate automated clinical histories for oncologic imaging from clinical notes. Compared with original requisition histories, LLM-generated histories were more complete and were preferred by radiologists for imaging interpretation and perceived safety. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Tavakoli and Kim in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e242134"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143190253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI and Surgical Findings Refine Concepts of Type 2 Cerebrospinal Fluid Leaks in Spontaneous Intracranial Hypotension.","authors":"Niklas Lützen, Jürgen Beck, Lalani Carlton Jones, Christian Fung, Theo Demerath, Alexander Rau, Charlotte Zander, Katharina Wolf, Florian Volz, Amir El Rahal, Horst Urbach","doi":"10.1148/radiol.241653","DOIUrl":"https://doi.org/10.1148/radiol.241653","url":null,"abstract":"<p><p>Background Type 2 lateral spinal cerebrospinal fluid (CSF) leakage occurs in approximately 20% of cases of spontaneous intracranial hypotension (SIH); however, the underlying pathologic mechanism remains ambiguous. Purpose To characterize MRI features of type 2 leaks, correlate them with intraoperative observations, and evaluate their diagnostic value. Materials and Methods Patients with SIH and type 2 leaks diagnosed between January 2021 and February 2023 were retrospectively identified. Characteristic imaging features from heavily T2-weighted MR myelography (T2-MRM) images were reevaluated (independently and blinded) in the type 2 leak sample mixed with a sample of 40 patients with SIH and type 1 (ventral) leaks. Available intraoperative data were reviewed for lateral dural tears, arachnoid outpouching, and ruptured spinal meningeal diverticula. Results Twenty-eight patients with SIH (mean age, 37.3 years ± 8.2 [SD]; 22 [79%] female patients) had 29 type 2 leaks between the T7 and L2 levels without side predominance. Characteristic cystic lesions with a broad dural base on the exiting nerve root sleeve were identified at T2-MRM; this \"bud-on-branch\" sign reflects an arachnoid outpouching herniating through a lateral dural tear, distinct from a meningeal diverticulum, which yielded a sensitivity of 79% (22 of 28; 95% CI: 59, 92) and a specificity of 100% (40 of 40; 95% CI: 91, 100) for leak location. Arachnoid outpouching was confirmed intraoperatively in 23 of 25 patients (92%; 95% CI: 81, 100), originating from the nerve root sleeve axilla in most patients (19 of 25, 76%; 95% CI: 59, 93); two of 25 patients (8%; 95% CI: 0, 19) had a dural tear only, and none had an underlying ruptured meningeal diverticulum. Conclusion This study showed that type 2 leaks are actually due to a lateral dural nerve root sleeve tear through which the arachnoid herniates, which contrasted the common perception that these leaks result from ruptured meningeal diverticula. These leaks had a characteristic anatomic distribution and MRI appearance with substantially facilitated leak localization in patients with SIH. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Rovira and Torres-Ferrús in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e241653"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143391639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Another Step toward Thrombectomy for All Patients with Severe Ischemic Stroke.","authors":"Anass Benomar, Jean Raymond","doi":"10.1148/radiol.250211","DOIUrl":"https://doi.org/10.1148/radiol.250211","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e250211"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Case 338.","authors":"Amar Shah, Maria Zulfiqar","doi":"10.1148/radiol.242727","DOIUrl":"https://doi.org/10.1148/radiol.242727","url":null,"abstract":"<p><strong>History: </strong>A 25-year-old female patient presented to the emergency department with worsening abdominal discomfort over the past 2-3 months. The patient had not experienced fever, chills, or dysuria. Past medical history was notable for two completed pregnancies; otherwise, there was no pertinent medical history or family history. At physical examination, the patient was uncomfortable but not in acute distress. There was tenderness to palpation in the right upper quadrant and epigastric region, but no rebound tenderness or guarding. Vital signs were blood pressure of 141/85 mm Hg, pulse rate of 91/min, and temperature of 37.2 °C. The serum β-human chorionic gonadotropin test result was negative for pregnancy, and urinalysis showed no leukocyte esterase or nitrites. Routine blood investigations, including white blood cell count, were within normal limits. Initial evaluation with contrast-enhanced CT of the abdomen and pelvis was performed (Fig 1), followed by MRI of the abdomen without and with intravenous contrast material (Fig 2).</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e242727"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Phase-resolved Functional Lung MRI Reveals Distinct Lung Perfusion Phenotype in Children and Adolescents with Post-COVID-19 Condition.","authors":"Gesa H Pöhler, Andreas Voskrebenzev, Marc-Luca Heinze, Valentina Skeries, Filip Klimeš, Julian Glandorf, Jan Eckstein, Nigar Babazade, Marius Wernz, Alexander Pfeil, Gesine Hansen, Frank K Wacker, Jens Vogel-Claussen, Martin Wetzke, Diane Miriam Renz","doi":"10.1148/radiol.241596","DOIUrl":"10.1148/radiol.241596","url":null,"abstract":"<p><p>Background Although measurable organic dysfunctions are frequently absent in pediatric patients with post-COVID-19 condition (PCC), this condition adversely affects quality of life. Free-breathing phase-resolved functional lung (PREFUL) MRI may be useful for assessing lung function in pediatric patients with PCC. Purpose To detect lung changes in children and adolescents with PCC compared with healthy control participants using PREFUL MRI. Materials and Methods In this single-center, prospective, cross-sectional study conducted between April 2022 and April 2023, children and adolescents (age ≤17 years) with PCC and age- and sex-matched healthy participants underwent MRI. Subgroup analysis was performed in participants with PCC who had cardiopulmonary symptoms. Regional ventilation, flow-volume loop correlation metric (FVL-CM), quantified perfusion, ventilation and perfusion defect percentages, and ventilation-perfusion ratios were compared between participants with PCC and controls using the Wilcoxon signed rank test. Correlation of imaging parameters with spirometry, heart rate, respiratory rate, and Bell score (fatigue severity) in participants with PCC was assessed using the Spearman rank correlation coefficient. Results The final study sample included 54 participants (27 participants with PCC and 27 matched control participants; median age, 15 years [IQR, 11-17 years]; 14 male participants). Twenty-one participants had cardiopulmonary symptoms. Participants with PCC had lower regional ventilation (median, 0.2 mL/mL [IQR, 0.1-0.2 mL/mL] vs 0.2 mL/mL [IQR, 0.2-0.2 mL/mL]; <i>P</i> = .047) and quantified perfusion (49 mL/min per 100 mL [IQR, 33-60 mL/min per 100 mL] vs 78 mL/min per 100 mL [IQR, 59-89 mL/min per 100 mL]; <i>P</i> < .001). Participants with PCC and cardiopulmonary symptoms had lower FVL-CMs (median, 0.99 arbitrary units [au] [IQR, 0.98-0.99 au] vs 0.99 au [IQR, 0.99-0.99 au]; <i>P</i> = .01) and higher ventilation defect (median, 7.6% [IQR, 4.5%-15.1%] vs 5.4% [IQR, 2.7%-7.1%]; <i>P</i> = .047) and perfusion defect percentage (median, 3.2% [IQR, 2.4%-4.2%] vs 2.3% [IQR, 1.8%-3.5%]; <i>P</i> = .02) compared with matched control participants. In participants with PCC, greater lung perfusion correlated with increased chronic fatigue severity (ρ = 0.48; <i>P</i> = .009) and higher ventilation-perfusion mismatch correlated with increased heart rate (ρ = 0.44; <i>P</i> = .02). Conclusion Free-breathing phase-resolved functional lung MRI-derived parameters helped identify a distinct phenotype of lung perfusion in children and adolescents with PCC and were correlated with heart rate and chronic fatigue severity. Clinical trial registration no. DRKS00028963 © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Parraga and Svenningsen in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e241596"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11868851/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143493394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Heat Line for Burned Bone.","authors":"Xingshun Zhou, Qiaoling Zhang","doi":"10.1148/radiol.241883","DOIUrl":"https://doi.org/10.1148/radiol.241883","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e241883"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143190272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Arthroscopy-validated Diagnostic Performance of 7-Minute Five-Sequence Deep Learning Super-Resolution 3-T Shoulder MRI.","authors":"Jan Vosshenrich, Mary Bruno, Tatiane Cantarelli Rodrigues, Ricardo Donners, Meghan Jardon, Yannik Leonhardt, Shana G Neumann, Michael Recht, Aline Serfaty, Steven E Stern, Jan Fritz","doi":"10.1148/radiol.241351","DOIUrl":"https://doi.org/10.1148/radiol.241351","url":null,"abstract":"<p><p>Background Deep learning (DL) methods enable faster shoulder MRI than conventional methods, but arthroscopy-validated evidence of good diagnostic performance is scarce. Purpose To validate the clinical efficacy of 7-minute threefold parallel imaging (PIx3)-accelerated DL super-resolution shoulder MRI against arthroscopic findings. Materials and Methods Adults with painful shoulder conditions who underwent PIx3-accelerated DL super-resolution 3-T shoulder MRI and arthroscopy between March and November 2023 were included in this retrospective study. Seven radiologists independently evaluated the MRI scan quality parameters and the presence of artifacts (Likert scale rating ranging from 1 [very bad/severe] to 5 [very good/absent]) as well as the presence of rotator cuff tears, superior and anteroinferior labral tears, biceps tendon tears, cartilage defects, Hill-Sachs lesions, Bankart fractures, and subacromial-subdeltoid bursitis. Interreader agreement based on κ values was evaluated, and diagnostic performance testing was conducted. Results A total of 121 adults (mean age, 55 years ± 14 [SD]; 75 male) who underwent MRI and arthroscopy within a median of 39 days (range, 1-90 days) were evaluated. The overall image quality was good (median rating, 4 [IQR, 4-4]), with high reader agreement (κ ≥ 0.86). Motion artifacts and image noise were minimal (rating of 4 [IQR, 4-4] for each), and reconstruction artifacts were absent (rating of 5 [IQR, 5-5]). Arthroscopy-validated abnormalities were detected with good or better interreader agreement (κ ≥ 0.68). The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were 89%, 90%, 89%, and 0.89, respectively, for supraspinatus-infraspinatus tendon tears; 82%, 63%, 68%, and 0.68 for subscapularis tendon tears; 93%, 73%, 86%, and 0.83 for superior labral tears; 100%, 100%, 100%, and 1.00 for anteroinferior labral tears; 68%, 90%, 82%, and 0.80 for biceps tendon tears; 42%, 93%, 81%, and 0.64 for cartilage defects; 93%, 99%, 98%, and 0.94 for Hill-Sachs deformities; 100%, 99%, 99%, and 1.00 for osseous Bankart lesions; and 97%, 63%, 92%, and 0.80 for subacromial-subdeltoid bursitis. Conclusion Seven-minute PIx3-accelerated DL super-resolution 3-T shoulder MRI has good diagnostic performance for diagnosing tendinous, labral, and osteocartilaginous abnormalities. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Tuite in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e241351"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Top 10 Tips for Writing about AI in <i>Radiology</i>: A Brief Guide for Authors.","authors":"Sarah L Atzen","doi":"10.1148/radiol.243347","DOIUrl":"https://doi.org/10.1148/radiol.243347","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"314 2","pages":"e243347"},"PeriodicalIF":12.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143441878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}