Pub Date : 2024-09-18DOI: 10.1016/j.acra.2024.09.006
Bridget Kowalczyk, Phil Ramis, Andrew Hillman, Regan City, Elizabeth Stukins, Krishna Nallamshetty, Eric M Rohren
Rationale and objectives: To investigate and discern if preferences and expectations regarding the stylistics of the radiology report varied across roles, specialties, and practice location amongst referring providers.
Materials and methods: A total of 579 referring clinicians were invited to complete our survey electronically and were asked to identify themselves as either physicians or advanced practice providers (APPs), specify their specialty, and primary practice environment. They were asked to rank the three reports on appearance, formatting, level of detail, and overall preference, with additional queries about their preferences regarding literature citation inclusions and placement of dose reduction statements.
Results: 477 surveys were completed and returned for analysis, resulting in an 82.2% response rate. The most preferred reporting style was the blended report (62.5%), followed by the narrative report (18.9%) and the highly templated report (18.7%), respectively. There were no statistically significant differences in the most preferred reporting style between provider types (F(1, 475) = [0.69], p = 0.4067), between different practice settings (F(2, 474) = [2.32], p = 0.0995), and between different medical specialties (F(5, 471) = [2.23], p = 0.051). Among the three report styles, blended reporting received the highest satisfaction scores overall. The highly templated report was rated lowest for appearance and detail, while narrative reports received moderate satisfaction scores for appearance and detail. A majority favored inclusion of literature citations and similarly, the placement of dose-optimization statements at the end of the report. Preferences were consistent across specialties and practice settings.
Conclusion: This survey highlights that a majority of clinicians across a variety of specialties prefer a mix of structured reporting with narrative elements. The standardization of required metrics included in the radiology report may have far-reaching consequences for future reimbursement.
理论依据和目标:调查并确定不同角色、专业和执业地点的转诊医生对放射报告文体的偏好和期望是否存在差异:我们共邀请了 579 位转诊临床医生通过电子方式完成调查,并要求他们表明自己是医生或高级医疗服务提供者 (APP),说明自己的专业和主要执业环境。他们被要求对三份报告的外观、格式、详细程度和总体偏好进行排序,并询问他们对文献引用和减少剂量声明位置的偏好:共有 477 份调查问卷完成并返回进行分析,回复率为 82.2%。最受欢迎的报告风格是混合报告(62.5%),其次分别是叙述报告(18.9%)和高度模板化报告(18.7%)。不同医疗服务提供者类型之间(F(1, 475) = [0.69], p = 0.4067)、不同医疗机构之间(F(2, 474) = [2.32], p = 0.0995)以及不同医学专业之间(F(5, 471) = [2.23], p = 0.051)在最喜欢的报告风格上没有明显的统计学差异。在三种报告风格中,混合报告的总体满意度得分最高。高度模板化的报告在外观和细节方面得分最低,而叙述式报告在外观和细节方面的满意度得分中等。大多数人赞成在报告中引用文献,同样,也赞成在报告末尾加入剂量优化说明。不同专业和执业环境的偏好是一致的:这项调查表明,不同专科的大多数临床医生都倾向于将结构化报告与叙述性内容相结合。放射学报告中所要求的指标标准化可能会对未来的报销产生深远影响。
{"title":"Radiology Reporting Preferences: What Do Referring Clinicians Want?","authors":"Bridget Kowalczyk, Phil Ramis, Andrew Hillman, Regan City, Elizabeth Stukins, Krishna Nallamshetty, Eric M Rohren","doi":"10.1016/j.acra.2024.09.006","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.006","url":null,"abstract":"<p><strong>Rationale and objectives: </strong>To investigate and discern if preferences and expectations regarding the stylistics of the radiology report varied across roles, specialties, and practice location amongst referring providers.</p><p><strong>Materials and methods: </strong>A total of 579 referring clinicians were invited to complete our survey electronically and were asked to identify themselves as either physicians or advanced practice providers (APPs), specify their specialty, and primary practice environment. They were asked to rank the three reports on appearance, formatting, level of detail, and overall preference, with additional queries about their preferences regarding literature citation inclusions and placement of dose reduction statements.</p><p><strong>Results: </strong>477 surveys were completed and returned for analysis, resulting in an 82.2% response rate. The most preferred reporting style was the blended report (62.5%), followed by the narrative report (18.9%) and the highly templated report (18.7%), respectively. There were no statistically significant differences in the most preferred reporting style between provider types (F(1, 475) = [0.69], p = 0.4067), between different practice settings (F(2, 474) = [2.32], p = 0.0995), and between different medical specialties (F(5, 471) = [2.23], p = 0.051). Among the three report styles, blended reporting received the highest satisfaction scores overall. The highly templated report was rated lowest for appearance and detail, while narrative reports received moderate satisfaction scores for appearance and detail. A majority favored inclusion of literature citations and similarly, the placement of dose-optimization statements at the end of the report. Preferences were consistent across specialties and practice settings.</p><p><strong>Conclusion: </strong>This survey highlights that a majority of clinicians across a variety of specialties prefer a mix of structured reporting with narrative elements. The standardization of required metrics included in the radiology report may have far-reaching consequences for future reimbursement.</p>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":3.8,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142300176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-17DOI: 10.1016/j.acra.2024.08.054
Gyan Moorthy,Leah Bush,Anne Zimmerman,Saurabh Jha
Radiology began as a translation of quantum physics to clinical medicine. Advances in computing and engineering enabled the differentiation of the field into diagnostic radiology, interventional radiology, and radiation oncology as practical responses to rapidly proliferating medical knowledge. Radiology has itself transformed modern medicine, helping clinicians identify, track, and intervene on multiple once deadly diseases. It is practiced in academic departments and hospital based, outpatient center based, or fully remote private groups of varying sizes, often with direct physicist support to optimize the use of complicated equipment. Importantly, radiology was guided to its current form not just by scientific advances, but by the interplay of cultural and governmental forces, as well as hard lessons, the results of constantly shifting balances of competing interests as follows: insurance, pharmaceutical, medical device, hospital, physician, physician extender, and patient. The purpose of this review is to describe the historical legal landscape of diagnostic radiology in the context of ethics, public health initiatives, and patient protections. For clarity, the review is divided into two parts.
{"title":"Laws That Have Shaped Radiology: Part I.","authors":"Gyan Moorthy,Leah Bush,Anne Zimmerman,Saurabh Jha","doi":"10.1016/j.acra.2024.08.054","DOIUrl":"https://doi.org/10.1016/j.acra.2024.08.054","url":null,"abstract":"Radiology began as a translation of quantum physics to clinical medicine. Advances in computing and engineering enabled the differentiation of the field into diagnostic radiology, interventional radiology, and radiation oncology as practical responses to rapidly proliferating medical knowledge. Radiology has itself transformed modern medicine, helping clinicians identify, track, and intervene on multiple once deadly diseases. It is practiced in academic departments and hospital based, outpatient center based, or fully remote private groups of varying sizes, often with direct physicist support to optimize the use of complicated equipment. Importantly, radiology was guided to its current form not just by scientific advances, but by the interplay of cultural and governmental forces, as well as hard lessons, the results of constantly shifting balances of competing interests as follows: insurance, pharmaceutical, medical device, hospital, physician, physician extender, and patient. The purpose of this review is to describe the historical legal landscape of diagnostic radiology in the context of ethics, public health initiatives, and patient protections. For clarity, the review is divided into two parts.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-17DOI: 10.1016/j.acra.2024.09.002
Pingping Wang,Danlei Song,JiaHao Han,Jing Zhang,Huihui Chen,Ruixia Gao,Huiming Shen,Jia Li
RATIONALE AND OBJECTIVESTo compare the diagnostic accuracy and grading ability of ultrasound-derived fat fraction (UDFF), controlled attenuation parameters (CAP), and hepatic/renal ratio (HRR) for hepatic steatosis in metabolic dysfunction-associated steatotic liver disease (MASLD) using magnetic resonance imaging proton density fat fraction (PDFF) as the gold standard.METHODSPatients suspected of having MASLD in our hospital between October 2023 and May 2024 were divided into the MASLD group and the control group. All patients underwent UDFF, CAP, and PDFF examinations. HRR was measured during routine ultrasound examination. In statistical analysis, we initially assessed the correlation between UDFF, CAP, HRR, and general characteristics of subjects with PDFF. Subsequently, receiver operating characteristic curve were employed to evaluate and compare the diagnostic performance of UDFF, CAP, and HRR for different grades of hepatic steatosis in MASLD. Their area under the curve, optimal cut-off value, sensitivity, and specificity were also determined. Finally, predictive factors determined hepatic steatosis in MASLD (PDFF≥6%) were identified through binary logistic regression analysis.RESULTS115 individuals were ultimately included in the MASLD group, while 102 were included in the control group. UDFF, CAP, and HRR were all positively correlated with PDFF. Among them, UDFF exhibited the strongest correlation with PDFF (ρ = 0.91). Furthermore, in the comparison of diagnostic efficacy among different grades of hepatic steatosis, UDFF outperformed CAP and HRR (p < 0.05). However, there were no statistically significant differences in AUCs between CAP and HRR across all three grades. The AUCs for UDFF in ≥S1, ≥S2, and ≥S3 were 0.99 (95% CI 0.97 to 1.00), 0.96 (95% CI 0.93 to 0.98), and 0.97 (95% CI 0.94 to 0.99), respectively. The optimal thresholds for UDFF are determined as follows: ≥ 6% for grade S1; ≥ 15% for grade S2; and ≥ 23% for grade S3. Multivariate analysis revealed that only age, UDFF, and CAP were important influencing factors for hepatic steatosis in MASLD.CONCLUSIONThe diagnostic accuracy of UDFF surpassed that of CAP and HRR in the detection and grading of hepatic steatosis in MASLD.
{"title":"Comparing Three Ultrasound-Based Techniques for Diagnosing and Grading Hepatic Steatosis in Metabolic Dysfunction-Associated Steatotic Liver Disease.","authors":"Pingping Wang,Danlei Song,JiaHao Han,Jing Zhang,Huihui Chen,Ruixia Gao,Huiming Shen,Jia Li","doi":"10.1016/j.acra.2024.09.002","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.002","url":null,"abstract":"RATIONALE AND OBJECTIVESTo compare the diagnostic accuracy and grading ability of ultrasound-derived fat fraction (UDFF), controlled attenuation parameters (CAP), and hepatic/renal ratio (HRR) for hepatic steatosis in metabolic dysfunction-associated steatotic liver disease (MASLD) using magnetic resonance imaging proton density fat fraction (PDFF) as the gold standard.METHODSPatients suspected of having MASLD in our hospital between October 2023 and May 2024 were divided into the MASLD group and the control group. All patients underwent UDFF, CAP, and PDFF examinations. HRR was measured during routine ultrasound examination. In statistical analysis, we initially assessed the correlation between UDFF, CAP, HRR, and general characteristics of subjects with PDFF. Subsequently, receiver operating characteristic curve were employed to evaluate and compare the diagnostic performance of UDFF, CAP, and HRR for different grades of hepatic steatosis in MASLD. Their area under the curve, optimal cut-off value, sensitivity, and specificity were also determined. Finally, predictive factors determined hepatic steatosis in MASLD (PDFF≥6%) were identified through binary logistic regression analysis.RESULTS115 individuals were ultimately included in the MASLD group, while 102 were included in the control group. UDFF, CAP, and HRR were all positively correlated with PDFF. Among them, UDFF exhibited the strongest correlation with PDFF (ρ = 0.91). Furthermore, in the comparison of diagnostic efficacy among different grades of hepatic steatosis, UDFF outperformed CAP and HRR (p < 0.05). However, there were no statistically significant differences in AUCs between CAP and HRR across all three grades. The AUCs for UDFF in ≥S1, ≥S2, and ≥S3 were 0.99 (95% CI 0.97 to 1.00), 0.96 (95% CI 0.93 to 0.98), and 0.97 (95% CI 0.94 to 0.99), respectively. The optimal thresholds for UDFF are determined as follows: ≥ 6% for grade S1; ≥ 15% for grade S2; and ≥ 23% for grade S3. Multivariate analysis revealed that only age, UDFF, and CAP were important influencing factors for hepatic steatosis in MASLD.CONCLUSIONThe diagnostic accuracy of UDFF surpassed that of CAP and HRR in the detection and grading of hepatic steatosis in MASLD.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-17DOI: 10.1016/j.acra.2024.09.005
Muhammed Said Beşler,Laura Oleaga,Vanesa Junquero,Cristina Merino
RATIONALE AND OBJECTIVESThis study aims to evaluate the performance of generative pre-trained transformer (GPT)-4o in the complete official European Board of Radiology (EBR) exam, designed to assess radiology knowledge, skills, and competence.MATERIALS AND METHODSQuestions based on text, image, or video and in the format of multiple choice, free-text reporting, or image annotation were uploaded into GPT-4o using standardized prompting. The results were compared to the average scores of radiologists taking the exam in real time.RESULTSIn Part 1 (multiple response questions and short cases), GPT-4o outperformed both the radiologists' average scores and the maximum pass score (70.2% vs. 58.4% and 60%, respectively). In Part 2 (clinically oriented reasoning evaluation), the performance of GPT-4o was below both the radiologists' average scores and the minimum pass score (52.9% vs. 66.1% and 55%, respectively). The accuracy on questions involving ultrasound images was higher compared to other imaging modalities (accuracy rate, 87.5-100%). For video-based questions, the performance was 50.6%. The model achieved the highest accuracy on most likely diagnosis questions but showed lower accuracy in free-text reporting and direct anatomical assessment in images (100% vs. 31% and 28.6%, respectively).CONCLUSIONThe abilities of GPT-4o in the official EBR exam are particularly noteworthy. This study demonstrates the potential of large language models to assist radiologists in assessing and managing cases from diagnosis to treatment or follow-up recommendations, even with zero-shot prompting.
{"title":"Evaluating GPT-4o's Performance in the Official European Board of Radiology Exam: A Comprehensive Assessment.","authors":"Muhammed Said Beşler,Laura Oleaga,Vanesa Junquero,Cristina Merino","doi":"10.1016/j.acra.2024.09.005","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.005","url":null,"abstract":"RATIONALE AND OBJECTIVESThis study aims to evaluate the performance of generative pre-trained transformer (GPT)-4o in the complete official European Board of Radiology (EBR) exam, designed to assess radiology knowledge, skills, and competence.MATERIALS AND METHODSQuestions based on text, image, or video and in the format of multiple choice, free-text reporting, or image annotation were uploaded into GPT-4o using standardized prompting. The results were compared to the average scores of radiologists taking the exam in real time.RESULTSIn Part 1 (multiple response questions and short cases), GPT-4o outperformed both the radiologists' average scores and the maximum pass score (70.2% vs. 58.4% and 60%, respectively). In Part 2 (clinically oriented reasoning evaluation), the performance of GPT-4o was below both the radiologists' average scores and the minimum pass score (52.9% vs. 66.1% and 55%, respectively). The accuracy on questions involving ultrasound images was higher compared to other imaging modalities (accuracy rate, 87.5-100%). For video-based questions, the performance was 50.6%. The model achieved the highest accuracy on most likely diagnosis questions but showed lower accuracy in free-text reporting and direct anatomical assessment in images (100% vs. 31% and 28.6%, respectively).CONCLUSIONThe abilities of GPT-4o in the official EBR exam are particularly noteworthy. This study demonstrates the potential of large language models to assist radiologists in assessing and managing cases from diagnosis to treatment or follow-up recommendations, even with zero-shot prompting.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
RATIONALE AND OBJECTIVESTo develop and validate a deep learning model for automated pathological grading and prognostic assessment of lung cancer using CT imaging, thereby providing surgeons with a non-invasive tool to guide surgical planning.MATERIAL AND METHODSThis study utilized 572 cases from the National Lung Screening Trial cohort, dividing them randomly into training (461 cases) and internal validation (111 cases) sets in an 8:2 ratio. Additionally, 224 cases from four cohorts obtained from the Cancer Imaging Archive, all diagnosed with non-small cell lung cancer, were included for external validation. The deep learning model, built on the MobileNetV3 architecture, was assessed in both internal and external validation sets using metrics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The model's prognostic value was further analyzed using Cox proportional hazards models.RESULTSThe model achieved high accuracy, sensitivity, specificity, and AUC in the internal validation set (accuracy: 0.888, macro AUC: 0.968, macro sensitivity: 0.798, macro specificity: 0.956). External validation demonstrated comparable performance (accuracy: 0.807, macro AUC: 0.920, macro sensitivity: 0.799, macro specificity: 0.896). The model's predicted signatures correlated significantly with patient mortality and provided valuable insights for prognostic assessment (adjusted HR 2.016 [95% CI: 1.010, 4.022]).CONCLUSIONSThis study successfully developed and validated a deep learning model for the preoperative grading of lung cancer pathology. The model's accurate predictions could serve as a useful adjunct in treatment planning for lung cancer patients, enabling more effective and customized interventions to improve patient outcomes.
{"title":"Deep Learning Model for Pathological Grading and Prognostic Assessment of Lung Cancer Using CT Imaging: A Study on NLST and External Validation Cohorts.","authors":"Runhuang Yang,Weiming Li,Siqi Yu,Zhiyuan Wu,Haiping Zhang,Xiangtong Liu,Lixin Tao,Xia Li,Jian Huang,Xiuhua Guo","doi":"10.1016/j.acra.2024.08.028","DOIUrl":"https://doi.org/10.1016/j.acra.2024.08.028","url":null,"abstract":"RATIONALE AND OBJECTIVESTo develop and validate a deep learning model for automated pathological grading and prognostic assessment of lung cancer using CT imaging, thereby providing surgeons with a non-invasive tool to guide surgical planning.MATERIAL AND METHODSThis study utilized 572 cases from the National Lung Screening Trial cohort, dividing them randomly into training (461 cases) and internal validation (111 cases) sets in an 8:2 ratio. Additionally, 224 cases from four cohorts obtained from the Cancer Imaging Archive, all diagnosed with non-small cell lung cancer, were included for external validation. The deep learning model, built on the MobileNetV3 architecture, was assessed in both internal and external validation sets using metrics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The model's prognostic value was further analyzed using Cox proportional hazards models.RESULTSThe model achieved high accuracy, sensitivity, specificity, and AUC in the internal validation set (accuracy: 0.888, macro AUC: 0.968, macro sensitivity: 0.798, macro specificity: 0.956). External validation demonstrated comparable performance (accuracy: 0.807, macro AUC: 0.920, macro sensitivity: 0.799, macro specificity: 0.896). The model's predicted signatures correlated significantly with patient mortality and provided valuable insights for prognostic assessment (adjusted HR 2.016 [95% CI: 1.010, 4.022]).CONCLUSIONSThis study successfully developed and validated a deep learning model for the preoperative grading of lung cancer pathology. The model's accurate predictions could serve as a useful adjunct in treatment planning for lung cancer patients, enabling more effective and customized interventions to improve patient outcomes.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
RATIONALE AND OBJECTIVESTraumatic neuroradiological emergencies necessitate rapid and accurate diagnosis, often relying on computed tomography (CT). However, the associated ionizing radiation poses long-term risks. Modern artificial intelligence reconstruction algorithms have shown promise in reducing radiation dose while maintaining image quality. Therefore, we aimed to evaluate the dose reduction capabilities of a deep learning-based denoising (DLD) algorithm in traumatic neuroradiological emergency CT scans.MATERIALS AND METHODSThis retrospective single-center study included 100 patients with neuroradiological trauma CT scans. Full-dose (100%) and low-dose (25%) simulated scans were processed using iterative reconstruction (IR2) and DLD. Subjective and objective image quality assessments were performed by four neuroradiologists alongside clinical endpoint analysis. Bayesian sensitivity and specificity were computed with 95% credible intervals.RESULTSSubjective analysis showed superior scores for 100% DLD compared to 100% IR2 and 25% IR2 (p < 0.001). No significant differences were observed between 25% DLD and 100% IR2. Objective analysis revealed no significant CT value differences but higher noise at 25% dose for DLD and IR2 compared to 100% (p < 0.001). DLD exhibited lower noise than IR2 at both dose levels (p < 0.001). Clinical endpoint analysis indicated equivalence to 100% IR2 in fracture detection for all datasets, with sensitivity losses in hemorrhage detection at 25% IR2. DLD (25% and 100%) maintained comparable sensitivity to 100% IR2. All comparisons demonstrated robust specificity.CONCLUSIONSThe evaluated algorithm enables high-quality, fully diagnostic CT scans at 25% of the initial radiation dose and improves patient care by reducing unnecessary radiation exposure.
{"title":"Deep Learning-Based Denoising Enables High-Quality, Fully Diagnostic Neuroradiological Trauma CT at 25% Radiation Dose.","authors":"Georg Gohla,Arne Estler,Leonie Zerweck,Jessica Knoppik,Christer Ruff,Sebastian Werner,Konstantin Nikolaou,Ulrike Ernemann,Saif Afat,Andreas Brendlin","doi":"10.1016/j.acra.2024.08.018","DOIUrl":"https://doi.org/10.1016/j.acra.2024.08.018","url":null,"abstract":"RATIONALE AND OBJECTIVESTraumatic neuroradiological emergencies necessitate rapid and accurate diagnosis, often relying on computed tomography (CT). However, the associated ionizing radiation poses long-term risks. Modern artificial intelligence reconstruction algorithms have shown promise in reducing radiation dose while maintaining image quality. Therefore, we aimed to evaluate the dose reduction capabilities of a deep learning-based denoising (DLD) algorithm in traumatic neuroradiological emergency CT scans.MATERIALS AND METHODSThis retrospective single-center study included 100 patients with neuroradiological trauma CT scans. Full-dose (100%) and low-dose (25%) simulated scans were processed using iterative reconstruction (IR2) and DLD. Subjective and objective image quality assessments were performed by four neuroradiologists alongside clinical endpoint analysis. Bayesian sensitivity and specificity were computed with 95% credible intervals.RESULTSSubjective analysis showed superior scores for 100% DLD compared to 100% IR2 and 25% IR2 (p < 0.001). No significant differences were observed between 25% DLD and 100% IR2. Objective analysis revealed no significant CT value differences but higher noise at 25% dose for DLD and IR2 compared to 100% (p < 0.001). DLD exhibited lower noise than IR2 at both dose levels (p < 0.001). Clinical endpoint analysis indicated equivalence to 100% IR2 in fracture detection for all datasets, with sensitivity losses in hemorrhage detection at 25% IR2. DLD (25% and 100%) maintained comparable sensitivity to 100% IR2. All comparisons demonstrated robust specificity.CONCLUSIONSThe evaluated algorithm enables high-quality, fully diagnostic CT scans at 25% of the initial radiation dose and improves patient care by reducing unnecessary radiation exposure.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-17DOI: 10.1016/j.acra.2024.08.034
Sung-Hye You,Byungjun Kim,InSeong Kim,Kyung-Sook Yang,Kyung Min Kim,Bo Kyu Kim,Jae Ho Shin
RATIONALE AND OBJECTIVESThe role of MR imaging in patients with cognitive impairment is to evaluate each component of Alzheimer's disease (AD), small vessel disease (SVD), and glymphatic function. We want to validate the diagnostic performance of the comprehensive interpretation of these parameters to predict the cognitive impairment stage. MATERIALS AND METHODS: This retrospective single-center study included 359 patients with cognitive impairment who had undergone MRI (FLAIR, T2WI, 3D-T1WI, susceptibility-weighted imaging, and diffusion tensor imaging [DTI]) and a neuropsychological screening battery between January 2020 and July 2022. Each AD and SVD-related MR parameter was visually evaluated, and DTI analysis along the perivascular space (ALPS) index was calculated. Volumetry analysis was performed using Neurophet AQUA AI-based software. Using logistic regression analysis, four types of models were developed and compared by adding the components in the following order: (1) clinical factors and AD, (2) SVD, (3) glymphatic function-related MR parameters, and (4) volumetric data. Chi-square automatic interaction detection algorithm was used to develop diagnostic tree analysis (DTA) model to predict late-stage cognitive impairment.RESULTSAPOE4 status, years of education, medial temporal lobe atrophy score, Fazekas scale score, DTI-ALPS index, and white matter hyperintensity were significant predictors of late-stage cognitive impairment. The performance of the prediction model increased from Model 1 to Model 4 (AUC: 0.880, 0.899, 0.914, and 0.945, respectively). The overall accuracy of the DTA model was 87.47%.CONCLUSIONIntegrative brain MRI assessments in patients with cognitive impairment, AD, SVD, and glymphatic function-related MR parameters, improve the prediction of late-stage cognitive impairment.
{"title":"Integrative MR Imaging Interpretation in Cognitive Impairment with Alzheimer's Disease, Small Vessel Disease, and Glymphatic Function-Related MR Parameters.","authors":"Sung-Hye You,Byungjun Kim,InSeong Kim,Kyung-Sook Yang,Kyung Min Kim,Bo Kyu Kim,Jae Ho Shin","doi":"10.1016/j.acra.2024.08.034","DOIUrl":"https://doi.org/10.1016/j.acra.2024.08.034","url":null,"abstract":"RATIONALE AND OBJECTIVESThe role of MR imaging in patients with cognitive impairment is to evaluate each component of Alzheimer's disease (AD), small vessel disease (SVD), and glymphatic function. We want to validate the diagnostic performance of the comprehensive interpretation of these parameters to predict the cognitive impairment stage. MATERIALS AND METHODS: This retrospective single-center study included 359 patients with cognitive impairment who had undergone MRI (FLAIR, T2WI, 3D-T1WI, susceptibility-weighted imaging, and diffusion tensor imaging [DTI]) and a neuropsychological screening battery between January 2020 and July 2022. Each AD and SVD-related MR parameter was visually evaluated, and DTI analysis along the perivascular space (ALPS) index was calculated. Volumetry analysis was performed using Neurophet AQUA AI-based software. Using logistic regression analysis, four types of models were developed and compared by adding the components in the following order: (1) clinical factors and AD, (2) SVD, (3) glymphatic function-related MR parameters, and (4) volumetric data. Chi-square automatic interaction detection algorithm was used to develop diagnostic tree analysis (DTA) model to predict late-stage cognitive impairment.RESULTSAPOE4 status, years of education, medial temporal lobe atrophy score, Fazekas scale score, DTI-ALPS index, and white matter hyperintensity were significant predictors of late-stage cognitive impairment. The performance of the prediction model increased from Model 1 to Model 4 (AUC: 0.880, 0.899, 0.914, and 0.945, respectively). The overall accuracy of the DTA model was 87.47%.CONCLUSIONIntegrative brain MRI assessments in patients with cognitive impairment, AD, SVD, and glymphatic function-related MR parameters, improve the prediction of late-stage cognitive impairment.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1016/j.acra.2024.09.014
Matthew Vickery,Erica Lanser,Kevin M Koch,Douglas Pierce,Joseph Budovec
RATIONALE AND OBJECTIVESTraditional medical student radiology experiences often lack interactivity and fail to replicate the clinical experience of being a radiologist. This study introduces SCRAPS, a novel simulation-based paradigm designed to improve the medical student experience and provide an active learning opportunity as part of their radiology rotation.MATERIALS AND METHODSSCRAPS utilizes a consumer-grade laptop, common word processing software, a free to use PACS and resident instructors to place students in a simulated reading-room environment. Students interpret pre-selected cases, dictate reports, and discuss findings with resident debriefing. Sessions lasted 60 to 90 min. Feedback was collected from 120 participating students (23 third year and 97 fourth year) via an anonymous survey.RESULTSStudents rated SCRAPS highly for its unique nature, enjoyability, and for providing insight into the process of performing clinical radiology tasks and endorsed it as valuable to their education.CONCLUSIONSCRAPS demonstrates promise for medical student education. It aligns with constructivist educational principles and is relatively easy to implement and adapt to new educational challenges.
{"title":"SCRAPS: Introducing a Student-Centered Resident-Administered PACS Simulator for Medical Student Radiology Education.","authors":"Matthew Vickery,Erica Lanser,Kevin M Koch,Douglas Pierce,Joseph Budovec","doi":"10.1016/j.acra.2024.09.014","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.014","url":null,"abstract":"RATIONALE AND OBJECTIVESTraditional medical student radiology experiences often lack interactivity and fail to replicate the clinical experience of being a radiologist. This study introduces SCRAPS, a novel simulation-based paradigm designed to improve the medical student experience and provide an active learning opportunity as part of their radiology rotation.MATERIALS AND METHODSSCRAPS utilizes a consumer-grade laptop, common word processing software, a free to use PACS and resident instructors to place students in a simulated reading-room environment. Students interpret pre-selected cases, dictate reports, and discuss findings with resident debriefing. Sessions lasted 60 to 90 min. Feedback was collected from 120 participating students (23 third year and 97 fourth year) via an anonymous survey.RESULTSStudents rated SCRAPS highly for its unique nature, enjoyability, and for providing insight into the process of performing clinical radiology tasks and endorsed it as valuable to their education.CONCLUSIONSCRAPS demonstrates promise for medical student education. It aligns with constructivist educational principles and is relatively easy to implement and adapt to new educational challenges.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1016/j.acra.2024.09.001
Xiaoyu Huang,Yong Huang,Kexin Liu,Fenglin Zhang,Zhou Zhu,Kai Xu,Ping Li
RATIONALE AND OBJECTIVESThis study aimed to develop a deep learning (DL) prognostic model to evaluate the significance of intra- and peritumoral radiomics in predicting outcomes for high-grade serous ovarian cancer (HGSOC) patients receiving platinum-based chemotherapy.MATERIALS AND METHODSA DL model was trained and validated on retrospectively collected unenhanced computed tomography (CT) scans from 474 patients at two institutions, which were divided into a training set (N = 362), an internal test set (N = 86), and an external test set (N = 26). The model incorporated tumor segmentation and peritumoral region analysis, using various input configurations: original tumor regions of interest (ROIs), ROI subregions, and ROIs expanded by 1 and 3 pixels. Model performance was assessed via hazard ratios (HRs) and receiver operating characteristic (ROC) curves. Patients were stratified into high- and low-risk groups on the basis of the training set's optimal cutoff value.RESULTSAmong the input configurations, the model using an ROI with a 1-pixel peritumoral expansion achieved the highest predictive accuracy. The DL model exhibited robust performance for predicting progression-free survival, with HRs of 3.41 (95% CI: 2.85, 4.08; P < 0.001) in training set, 1.14 (95% CI: 1.03, 1.26; P = 0.012) in internal test set, and 1.32 (95% CI: 1.07, 1.63; P = 0.011) in external test set. KM survival analysis revealed significant differences between the high-risk and low-risk groups (P < 0.05).CONCLUSIONThe DL model effectively predicts survival outcomes in HGSOC patients receiving platinum-based chemotherapy, offering valuable insights for prognostic assessment and personalized treatment planning.
{"title":"Intratumoral and Peritumoral Radiomics for Predicting the Prognosis of High-grade Serous Ovarian Cancer Patients Receiving Platinum-Based Chemotherapy.","authors":"Xiaoyu Huang,Yong Huang,Kexin Liu,Fenglin Zhang,Zhou Zhu,Kai Xu,Ping Li","doi":"10.1016/j.acra.2024.09.001","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.001","url":null,"abstract":"RATIONALE AND OBJECTIVESThis study aimed to develop a deep learning (DL) prognostic model to evaluate the significance of intra- and peritumoral radiomics in predicting outcomes for high-grade serous ovarian cancer (HGSOC) patients receiving platinum-based chemotherapy.MATERIALS AND METHODSA DL model was trained and validated on retrospectively collected unenhanced computed tomography (CT) scans from 474 patients at two institutions, which were divided into a training set (N = 362), an internal test set (N = 86), and an external test set (N = 26). The model incorporated tumor segmentation and peritumoral region analysis, using various input configurations: original tumor regions of interest (ROIs), ROI subregions, and ROIs expanded by 1 and 3 pixels. Model performance was assessed via hazard ratios (HRs) and receiver operating characteristic (ROC) curves. Patients were stratified into high- and low-risk groups on the basis of the training set's optimal cutoff value.RESULTSAmong the input configurations, the model using an ROI with a 1-pixel peritumoral expansion achieved the highest predictive accuracy. The DL model exhibited robust performance for predicting progression-free survival, with HRs of 3.41 (95% CI: 2.85, 4.08; P < 0.001) in training set, 1.14 (95% CI: 1.03, 1.26; P = 0.012) in internal test set, and 1.32 (95% CI: 1.07, 1.63; P = 0.011) in external test set. KM survival analysis revealed significant differences between the high-risk and low-risk groups (P < 0.05).CONCLUSIONThe DL model effectively predicts survival outcomes in HGSOC patients receiving platinum-based chemotherapy, offering valuable insights for prognostic assessment and personalized treatment planning.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142262201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1016/j.acra.2024.09.008
Xiuzhen Yao,Shuitang Deng,Xiaoyu Han,Danjiang Huang,Zhengyu Cao,Xiaoxiang Ning,Weiqun Ao
RATIONALE AND OBJECTIVESTo develop and validate multimodal deep-learning models based on clinical variables, multiparametric MRI (mp-MRI) and hematoxylin and eosin (HE) stained pathology slides for predicting microsatellite instability (MSI) status in rectal cancer patients.MATERIALS AND METHODSA total of 467 surgically confirmed rectal cancer patients from three centers were included in this study. Patients from center 1 were randomly divided into a training set (242 patients) and an internal validation (invad) set (105 patients) in a 7:3 ratio. Patients from centers 2 and 3 (120 patients) were included in an external validation (exvad) set. HE and immunohistochemistry (IHC) staining were analyzed, and MSI status was confirmed by IHC staining. Independent predictive factors were identified through univariate and multivariate analyses based on clinical evaluations and were used to construct a clinical model. Deep learning with ResNet-101 was applied to preoperative MRI (T2WI, DWI, and contrast-enhanced T1WI sequences) and postoperative HE-stained images to calculate deep-learning radiomics score (DLRS) and deep-learning pathomics score (DLPS), respectively, and to DLRS and DLPS models. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was used to evaluate and compare the predictive performance of each model.RESULTSAmong all rectal cancer patients, 82 (17.6%) had MSI. Long diameter (LD) and pathological T stage (pT) were identified as independent predictors and were used to construct the clinical model. After undergoing deep learning and feature selection, a final set of 30 radiomics features and 30 pathomics features were selected to construct the DLRS and DLPS models. A nomogram combining the clinical model, DLRS, and DLPS was created through weighted linear combination. The AUC values of the clinical model for predicting MSI were 0.714, 0.639, and 0.697 in the training, invad, and exvad sets, respectively. The AUCs of DLPS and DLRS ranged from 0.896 to 0.961 across the training, invad, and exvad sets. The nomogram achieved AUC values of 0.987, 0.987, and 0.974, with sensitivities of 1.0, 0.963, and 1.0 and specificities of 0.919, 0.949, and 0.867 in the training, invad, and exvad sets, respectively. The nomogram outperformed the other three models in all sets, with DeLong test results indicating superior predictive performance in the training set.CONCLUSIONThe nomogram, incorporating clinical data, mp-MRI, and HE staining, effectively reflects tumor heterogeneity by integrating multimodal data. This model demonstrates high predictive accuracy and generalizability in predicting MSI status in rectal cancer patients.
{"title":"Deep Learning Algorithm‑Based MRI Radiomics and Pathomics for Predicting Microsatellite Instability Status in Rectal Cancer: A Multicenter Study.","authors":"Xiuzhen Yao,Shuitang Deng,Xiaoyu Han,Danjiang Huang,Zhengyu Cao,Xiaoxiang Ning,Weiqun Ao","doi":"10.1016/j.acra.2024.09.008","DOIUrl":"https://doi.org/10.1016/j.acra.2024.09.008","url":null,"abstract":"RATIONALE AND OBJECTIVESTo develop and validate multimodal deep-learning models based on clinical variables, multiparametric MRI (mp-MRI) and hematoxylin and eosin (HE) stained pathology slides for predicting microsatellite instability (MSI) status in rectal cancer patients.MATERIALS AND METHODSA total of 467 surgically confirmed rectal cancer patients from three centers were included in this study. Patients from center 1 were randomly divided into a training set (242 patients) and an internal validation (invad) set (105 patients) in a 7:3 ratio. Patients from centers 2 and 3 (120 patients) were included in an external validation (exvad) set. HE and immunohistochemistry (IHC) staining were analyzed, and MSI status was confirmed by IHC staining. Independent predictive factors were identified through univariate and multivariate analyses based on clinical evaluations and were used to construct a clinical model. Deep learning with ResNet-101 was applied to preoperative MRI (T2WI, DWI, and contrast-enhanced T1WI sequences) and postoperative HE-stained images to calculate deep-learning radiomics score (DLRS) and deep-learning pathomics score (DLPS), respectively, and to DLRS and DLPS models. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was used to evaluate and compare the predictive performance of each model.RESULTSAmong all rectal cancer patients, 82 (17.6%) had MSI. Long diameter (LD) and pathological T stage (pT) were identified as independent predictors and were used to construct the clinical model. After undergoing deep learning and feature selection, a final set of 30 radiomics features and 30 pathomics features were selected to construct the DLRS and DLPS models. A nomogram combining the clinical model, DLRS, and DLPS was created through weighted linear combination. The AUC values of the clinical model for predicting MSI were 0.714, 0.639, and 0.697 in the training, invad, and exvad sets, respectively. The AUCs of DLPS and DLRS ranged from 0.896 to 0.961 across the training, invad, and exvad sets. The nomogram achieved AUC values of 0.987, 0.987, and 0.974, with sensitivities of 1.0, 0.963, and 1.0 and specificities of 0.919, 0.949, and 0.867 in the training, invad, and exvad sets, respectively. The nomogram outperformed the other three models in all sets, with DeLong test results indicating superior predictive performance in the training set.CONCLUSIONThe nomogram, incorporating clinical data, mp-MRI, and HE staining, effectively reflects tumor heterogeneity by integrating multimodal data. This model demonstrates high predictive accuracy and generalizability in predicting MSI status in rectal cancer patients.","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":null,"pages":null},"PeriodicalIF":4.8,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142261937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}