Scientific Evidence

*This information may not be within the Indicated uses for Canon Medical System devices. The views and findings included here are those of the presenter. Clinical settings and other considerations may impact repeating these results. The results may not have been validated by Canon or submitted to FDA for market clearance.

CANON PCCT: Photon Counting CT

Evaluation of coronary stenosis, plaques, and stents in a phantom using the SHR Spectral mode on a CZT PCCT prototype scanner | T1- SSPH07-6


Ruoqiao Zhang– Canon Medical Research Institute USA

Conclusions: We compared image quality between NR and SHR Spectral modes on a full-size PCCT prototype system. Result demonstrated that SHR Spectral improved the spatial resolution significantly than NR Spectral while retaining spectral information for material discrimination.

Clinical Relevance/Application: The SHR Spectral mode in our PCCT system can potentially provide more accurate evaluation of different types and degrees of in-stent stenosis and plaques than conventional scanners in cardiac imaging.

Scientific Presentation Tuesday, Dec 3 | 8:00 AM - 9:00 AM CST | N229

Ultra High Resolution Spectral Imaging Using CZT-Based Photon-Counting CT: A Phantom Study for Evaluation of Spatial Resolution and Iodine Quantification Accuracy| T1-SSPH07-4


Shobhit Sharma, PhD – Canon Medical Research USA

Conclusions: Spectral-UHR imaging with CZT-based PCCT enables visualization of high-resolution features while preserving spectral information as indicated by good iodine quantification accuracy across all exposures.

Clinical Relevance/Application: Iodine quantification accuracy is important for all contrast-enhanced CT exams, especially in cardiovascular and oncological domains. The ability to visualize finer structures while preserving iodine accuracy would remarkably improve the diagnostic performance of CT images.

Scientific Presentation | Tuesday, Dec 3 8:00 AM - 9:00 AM CST N229

IMAGING-DETECTED EXTRANODAL EXTENSION IN HEAD AND NECK CANCER: CLINICAL IMPLICATIONS AND DIAGNOSTIC CRITERIA IN THE ERA OF HIGH-RESOLUTION IMAGING INCLUDING PHOTON-COUNTING DETECTOR CT | HNEE-49


Hirofumi Kuno, MD, PhD – National Cancer Center Hospital East

Conclusions: Teaching Points: The objectives of this presentation are: 1) to review the clinical implications of iENE in patients with head and neck cancer, and 2) to discuss the diagnostic criteria for iENE in the era of high-resolution imaging with radiol-pathologic / clinico-radiologic correlation.

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

SHARPNESS EVALUATION OF CHEST MULTI PLANAR RECONSTRUCTION IMAGES WITH NORMAL AND SUPER HIGH-RESOLUTION MODE OF CZT-BASED PHOTON-COUNTING DETECTOR CT | M5A-SPPH-7


Keiichi Nomura, PhD- National Cancer Center Hospital East

Conclusions: The multi planar SHR images with a CZT-based PCD-CT scanner were found to have approximately 1.5 times higher sharpness than the NR images.

Clinical Relevance/Application: This study highlights the advantages of using the SHR mode with a CZT-based PCD-CT scanner for evaluating the microstructure of lung nodules in thoracic imaging.

Poster Presentation Monday, Dec 2 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

EVALUATION OF LUNG NODULE VOLUMES USING CONVENTIONAL CT AND CDZNTE-PHOTON COUNTING CT WITH HYBRID AND DEEP LEARNING RECONSTRUCTION METHODS | T2-SPPH-1


Paulo R. Costa, PhD – Nijmegen Radboudumc

Conclusions: The best performance in terms of measurement precision and accuracy was found using PCCT/DLR, closely followed by the PCCT/HIR.

Clinical Relevance/Application: The accurate and precise estimation of lung nodule volumes is essential for the optimal follow-up of screening- or incidentally-detected nodules.

Poster Presentation Tuesday, Dec 3 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

INITIAL CLINICAL EXPERIENCES OF A NOVEL PHOTON COUNTING SYSTEM COMBINING CADMIUM ZINC TELLURIDE DETECTORS AND SUPER-HIGH RESOLUTION DEEP-LEARNING IMAGE RECONSTRUCTION | S3A-SPCH-9


Ewoud Joris Smit, MD, PhD – Nijmegen Radboudumc

Conclusions: Initial results show that PCCT combining CZT detectors and SHR-DLR reconstruction provides superior spatial resolution and excellent depiction of the bronchial system while keeping image noise low.

*Clinical Relevance/Application: Photon counting CT systems with SHR-DLR providing low-noise super-high-resolution images of the chest may push diagnostic abilities past their current limits.

Poster Presentation Sunday, Dec 1 | 11:45 AM - 12:15 PM CST | LEARNING CENTER

CHARACTERIZATION OF LUNG LESIONS WITH HYBRID LIFELIKE PHANTOMS IN LOW-DOSE PHOTON-COUNTING DETECTOR CT | M5B-SPPH-3


Kai Mei, PhD – University of Pennsylvania

Conclusions: A hybrid lifelike phantom with computational lesions was used to assess the performance of EIDCT and PCDCT for low-dose lung imaging. As indicated by the higher SSIM and lower RMSE compared to EIDCT at all doses, PCDCT images are superior for characterizing lung lesions, especially at lower radiation doses.

*Clinical Relevance/Application: Accurate visual characterization of morphological and texture features of lesions while reducing patient dose is clinically important for monitoring the severity and progression of lung cancer.

Poster Presentation Monday, Dec 2 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

REALISTIC 3D-PRINTED HIGH-RESOLUTION CT PHANTOMS FEATURING COMPUTATIONAL LUNG LESIONS FOR PHOTON-COUNTING COMPUTED TOMOGRAPHY | M5B-SPPH-4


Kai Mei, PhD – University of Pennsylvania

Conclusions: The study demonstrates that computational lung lesions can be incorporated into PixelPrint phantoms, facilitating the evaluation PCCT UHR imaging. Availability of such phantoms would enable accurate and customized evaluations of this technology by providing access to high-resolution ground-truth information.

*Clinical Relevance/Application: New medical imaging technologies require task-specific evaluations for effective translation to clinical use. Availability of a high-resolution lung phantom with access to ground-truth would allow for evaluation of PCCT UHR imaging for tasks involving detection and classification of lung nodules.

Poster Presentation Monday, Dec 2 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

QUANTITATIVE LUNG IMAGING WITH CZT-BASED PHOTON-COUNTING CT: BENEFITS OF ULTRA HIGH-RESOLUTION IMAGING AND DEEP-LEARNING RECONSTRUCTION | R5B-SPPH-1


Shobhit Sharma, PhD – Canon Medical Research, USA

Conclusions: DLR in UHR imaging with CZT-based PCCT improves noise and preserves spatial resolution, as indicated by CNRs much higher than FBP and comparable to dnFBP at all exposures, while maintaining superior accuracy for WA% measurements. These benefits have potential for improvements in quantitative lung imaging while allowing for valuable reductions in patient dose.

*Clinical Relevance/Application: The ability to accurately characterize changes in airway walls while reducing patient dose is crucial for monitoring progression of lung diseases such as COPD and pulmonary fibrosis.

Poster Presentation Thursday, Dec 5 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

PRELIMINARY CLINICAL EVALUATION OF LUNG ADENOCARCINOMAS USING A CADMIUM ZINC TELLURIDE-BASED PHOTON-COUNTING DETECTOR COMPUTED TOMOGRAPHY | M5A-SPCH-2


Tomoaki Sasaki, MD – National Cancer Center Hospital East

Conclusions: SHR modes showed improved diagnostic confidence for most imaging findings of lung adenocarcinomas. Furthermore, DLR added further improvement of diagnostic confidence and significant reduction of image noise.

Clinical Relevance/Application: SHR with DLR can depict the details of lung adenocarcinomas with higher diagnostic confidence with significant reduction of image noise.

Poster Presentation Monday, Dec 2 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

CANON PIQE: Precise IQ Engine – Super Resolution

EVALUATION OF STENT IMAGING USING SUPER-RESOLUTION DEEP LEARNING IMAGE RECONSTRUCTION IN CORONARY CT ANGIOGRAPHY : COMPARATIVE STUDY OF IMAGE RECONSTRUCTIONTECHNOLOGY BY 3 DIFFERENT CT MANUFACTURERS USING PHANTOMS AND INVITRO | T2-SPPH-6


Sung Min Ko, PhD - Seoul National University

Conclusions: SR-DLR resulted in significantly lower blooming artifact with clear stent lumen and strut, decreased the image noise, increased the SNR and CNR compared with FBP, HIR, and DLR.

Clinical Relevance/Application: Coronary CT angiography with DLR resulted in lower blooming artifacts, thereby improving visualization of increased in-stent stenosis and luminal changes.

Poster Presentation Tuesday, Dec 3 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

A PRELIMINARY STUDY OF SUPER-RESOLUTION DEEP LEARNING RECONSTRUCTION WITH CARDIAC OPTION FOR EVALUATION OF ENDOVASCULAR-TREATED INTRACRANIAL ANEURYSMS | W2-SPNR-13


Sung Min Ko, PhD – Seoul National University

Conclusions: SR-DLR with cardiac option is useful for follow-up imaging in stent-assisted coil embolization and flow-diverting stent placement in terms of lower image noise, higher SNR and CNR, superior subjective image analysis, and less blooming artifact than other image reconstructions.

Clinical Relevance/Application: SR-DLR with cardiac option allow better visualization of the peripheral and smaller cerebral arteries. SR-DLR with cardiac option can be beneficial for CT imaging of stent-assisted coil embolization and flow-diverting sten.

Poster Presentation Wednesday, Dec 4 |9:00 AM - 9:30 AM CST

INFLUENCE OF RADIATION DOSE REDUCTION AND RECONSTRUCTION ALGORITHMS TO QUANTITATIVE COPD EVALUATIONS ON HIGH-DEFINITION CT WITH SUPER-HIGH-RESOLUTION MODE IN SMOKERS | R2-SPPH-3


Dr. Ohno, MD – Fujita Health University

Conclusions: HDCT has little influence of radiation dose reduction and reconstruction algorithm to quantitative CT evaluation of COPD on HDCT with SHR mode in smokers.

Clinical Relevance/Application: Quantitatively assessed HDCT has little influence of radiation dose reduction and reconstruction algorithm to COPD evaluation in smokers.

Poster Presentation Thursday, Dec 5 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

SUPER-RESOLUTION DEEP LEARNING RECONSTRUCTION FOR ASSESSMENT OF PANCREATIC DUCTAL ADENOCARCINOMA ON THIN-SLICE ABDOMINAL CT | R5A-SPGI-1


Yasunori Nagayama, MD – Kumamoto University Hospital

Conclusions: Compared with HIR and NR-DLR, the novel SR-DLR algorithm enhanced spatial resolution, reduced image noise, and improved the delineation of pancreatic tumors and adjacent structures in thin-slice multi-phase pancreatic CT, thus increasing diagnostic confidence for the assessment of PDAC.

Clinical Relevance/Application: The recently introduced SR-DLR algorithm optimized for body CT may facilitate a more confident and detailed assessment of PDAC owing to its superior spatial resolution, noise performance, and visibility of tumor and adjacent structures compared to HIR and NR-DLR algorithms.

Poster Presentation Thursday, Dec 5 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

REDUCING RADIATION DOSE AND IMPROVING SPATIAL RESOLUTION USING SUPER-RESOLUTION DEEP-LEARNING-BASED RECONSTRUCTION FOR ABDOMINAL CT: PHANTOM STUDY | R2-SPPH-1


Yoshinori Funama, PhD – Kumamoto University Hospital

Conclusions: SR-DLR images for the dedicated abdomen demonstrate substantial noise reduction while improving spatial resolution, particularly beneficial for smaller FOV.

Clinical Relevance/Application: SR-DLR images for dedicated abdomen are expected to benefit not only adults but also pediatric patients in particular due to substantial radiation dose reduction and improvement of spatial resolution.

Poster Presentation Thursday, Dec 5 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

RADIATION DOSE REDUCTION IN CORONARY CT ANGIOGRAPHY WITH SUPER-RESOLUTION DEEP LEARNING RECONSTRUCTION | R5B-SPCA-3


Limiao Zou – Peking Union Medical College Hospital

Conclusions: Compared with HIR and NR-DLR, ULD scans with SR-DLR achieved 60% reduction of radiation dose from LD scans with HIR while maintaining or even improving image quality.

Clinical Relevance/Application: Implementation of SR-DLR enables a substantial dose reduction from CCTA with preserved or even improved image quality.

Poster Presentation Thursday, Dec 5 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

CANON DLR: Deep Learning Reconstruction

AI-BASED COMPUTER-AIDED VOLUMETRY FOR INVASIVENESS EVALUATION IN LUNG ADENOCARCINOMA: COMPARISON OF CAPABILITY FOR NODULE COMPONENT MEASUREMENT AND DIAGNOSTIC PERFORMANCE ON DIFFERENT DOSE LEVELS AND RECONSTRUCTION METHODS ON ULTRA-HIGH-RESOLUTION CT | M3-SSPH03-2


Yoshiyuki Ozawa, MD, PhD– Fujita Health University

Conclusions: DLR has better potential than hybrid-type IR for improving diagnostic performance of AI-based CADv for invasiveness evaluation in lung adenocarcinoma on UHR-CT with standard-reduced- and ultra-low-dose levels.

Clinical Relevance/Application: DLR has better potential than hybrid-type IR for improving diagnostic performance of AI-based CADv for invasiveness evaluation in lung adenocarcinoma on UHR-CT with standard-reduced- and ultra-low-dose levels.

Scientific Presentation Monday, Dec 2 | 9:30 AM - 10:30 AM CST | S401

MYOCARDIAL EXTRACELLULAR VOLUME QUANTIFICATION IN CARDIAC CT: COMPARISON OF THE EFFECTS OF DEEP LEARNING RECONSTRUCTION AND ITERATIVE RECONSTRUCTION ALGORITHMS WITH MRI AS A REFERENCE STANDARD | S3A-SPCA-7


Hao Woo – Affiliated Zhongshan Hospital of Dalian University

Conclusions: DLR-AiCE can improved accuracy of myocardial ECV quantification when compared with iterative reconstruction using MRI as a reference standard.

Clinical Relevance/Application: Because of the lengthy acquisition time, the numerous contraindications and complexity associated with the scanning process, the prevalence of CMR is limited. Cardiac CT, on the other hand, is widely used due to its ease of access and quick acquisition times. Therefore, CT-ECV quantification derived from DLR-AiCE may represent a useful non-invasive alternative for assessing ischemic and non-ischemic cardiomyopathies.

Poster Presentation Sunday, Dec 1 | 11:45 AM - 12:15 PM CST | LEARNING CENTER

DEEP LEARNING RECONSTRUCTION: CAPABILITIES FOR IMPROVING IMAGE QUALITY AND ANATOMICAL STRUCTURE VISUALIZATION ON HIGH-DEFINITION CT | S3B-SPHN-1


Masahiko Nomura, MD, PhD – Fujita Health University

Conclusions: Newly developed DLR showed superior potential to that of hybrid-type IR for better image quality and visualization of anatomical landmarks in middle and inner ears and temporal bones on HDCT used in clinical practice.

Clinical Relevance/Application: Newly developed DLR showed superior potential to that of hybrid-type IR for better image quality and visualization of anatomical landmarks in middle and inner ears and temporal bones on HDCT used in clinical practice.

Poster Presentation Sunday, Dec 1 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

DEEP LEARNING-BASED CT IMAGE RECONSTRUCTION: IMAGE QUALITY EVALUATION IN AMPULLARY AND PERIAMPULLARY LESIONS | S3B-SPGI-13


Jia Xu – Peking Union Medical College Hospital

Conclusions: DLR images yield significant lower the image noise, higher SNR and CNR compared to FBP and AIDR3D images. DLR images showed the best visual image quality of the ampullary and periampullary lesions.

Clinical Relevance/Application: DLR method can improve the image quality, resulted in a better observation of the ampullary and periampullary lesions.

Poster Presentation Sunday, Dec 1 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

ASSESSMENT OF LOW CONTRAST DETECTABILITY VIA AN IN-SILICO IMAGING TRIAL COMPARING DEEP LEARNING RECONSTRUCTION TO CONVENTIONAL RECONSTRUCTION ALGORITHMS | M5B-SPPH-2


Daniel W Shin, MS, BS – Canon Medical Systems Corp.

Conclusions: This in-silico imaging study illustrates significantly improved LCD with DLR compared to conventional reconstruction methods.

*Clinical Relevance/Application: The improved LCD with DLR demonstrated in Silico provides evidence DLR is likely to increase diagnostic confidence and facilitate dose optimization in acquired patient data.

Poster Presentation Monday, Dec 2 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

COMPREHENSIVE EVALUATION OF DEEP LEARNING RECONSTRUCTION IN ULTRA-LOW-DOSE CT FOR UROLITHIASIS: ASSESSING IMAGE QUALITY AND LESION DETECTION IN PHANTOM AND CLINICAL SCENARIOS | M5A-SPGU-9


Xiaoxiao Zhang – Peking Union Medical College Hospital

Conclusions: The AICEs reconstruction algorithm demonstrates promise in optimizing ULDCT for urolithiasis, offering superior lesion detection and image quality enhancement across phantom and clinical studies.

Clinical Relevance/Application: AICE emerges as a preferred reconstruction algorithm for ULDCT in urolithiasis, potentially enhancing diagnostic accuracy and patient care in urological imaging.

Poster Presentation Monday, Dec 2 |12:15 PM - 12:45 PM CST | LEARNING CENTER

SUBTRACTION IODINE MAPS COMBINED WITH DEEP LEARNING RECONSTRUCTION FOR EVALUATION OF TEMPORAL BONE TUMORS | M2-SPHN-3


Tong Su, MD – Peking Union Medical College Hospital

Conclusions: Compared with conventional enhanced CT, subtraction iodine maps combined with deep learning reconstruction can improve the image quality of CT images of temporal bone tumor. With equivalent capability to detect temporal bone lesions, SIM-DLR can notably enhance the visualization of lesion contour and bone involvement.

Clinical Relevance/Application: SIM-DLR can enhance the visualization of the extent of tumor invasion in the temporal bone, facilitating the formulation of surgical strategies.

Poster Presentation Monday, Dec 2 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

DEEP LEARNING RECONSTRUCTION IMPROVES PULMONARY NODULE DETECTION AND MEASUREMENT ACCURACY FOR ULTRA-LOW-DOSE CHEST CT COMPARED TO HYBRID ITERATIVE RECONSTRUCTION | T5A-SPCH-8


Jinhua Wang, MD – Peking Union Medical College Hospital

Conclusions: DLR was superior to HIR in terms of image noise and image quality at ultra-low-dose settings. DLR also increased the lung nodule detection rate and improved measurement accuracy of nodule volume in comparison with HIR.

Clinical Relevance/Application: The study prospectively evaluated the performance of DLR applied to chest ultra-low-dose CT for image quality improvement and lung nodule assessment using a deep learning-based nodule evaluation system.

Poster Presentation Tuesday, Dec 3 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

DETECTION OF INTRACRANIAL HEMORRHAGE USING ULTRALOW-DOSE BRAIN COMPUTED TOMOGRAPHY WITH DEEP LEARNING RECONSTRUCTION VERSUS CONVENTIONAL-DOSE COMPUTED TOMOGRAPHY | W2-SPNR-14


Chuluunbaatar Otgonbaatar, MD – Seoul National University

Conclusions: Ultralow-dose CT with DLR is an acceptable technique that provides higher image quality and diagnostic performance with a reduction in radiation dose of approximately 87.7% compared with conventional-dose CT.

Clinical Relevance/Application: The present study investigated the value of ultralow-dose CT with DLR for the assessment of image quality and the evaluation of intracranial hemorrhage compared with ultralow-dose CT with IR and conventional-dose CT. Image quality was significantly improved in terms of image noise, SNR, and CNR, and higher diagnostic performance was observed with ultralow-dose CT with DLR than with ultralow-dose CT with IR.

Poster Presentation, Dec 4 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

REDUCING BOTH RADIATION AND CONTRAST DOSES IN 80-KV HEAD AND NECK CT ANGIOGRAPHY WITH VARIOUS STRENGTH DEEP LEARNING IMAGE RECONSTRUCTION: COMPARISON WITH 100-KV CT ANGIOGRAPHY | R2-SPNR-2


Jiajing Tong – Peking Union Medical College Hospital

Conclusions: The application of DLR in head and neck CTA acquired with 80kV tube voltage and 37.5% lower contrast dose maintained diagnostic image quality while exhibited lower image noise and higher CNR compared to regular-dose with HIR

Clinical Relevance/Application: The DLR (especially DLR-H) has great potential to reduce radiation dose and improving image quality in head and neck CTA with 80-kV tube voltage.

Poster Presentation Thursday, Dec 5 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

CANON CVCT: Enhanced Cardio-Vascular CT

DYNAMIC CONTRAST-ENHANCED AREA-DETECTOR CT FOR DETECTION OF TISSUE BLOOD SUPPLY CHANGES IN STAGE I NSCLC PATIENTS WITH PF-ILD | R1-SSCH09-2


Yoshiharu Ohno, MD, PhD – Fujita Health University

Conclusions: Dynamic first-pass CE-perfusion ADCT is useful for detecting pathological structural changes in stage I NSCLC patients with PF-ILD.

Clinical Relevance/Application: Dynamic first-pass CE-perfusion ADCT is useful for detecting pathological structure changes in stage I NSCLC patients with PF-ILD.

Scientific Presentation Thursday, Dec 5 | 8:00 AM - 9:00 AM CST | E451A

THE IMPACT OF DEEP LEARNING RECONSTRUCTION COMBINED WITH CONTRAST-ENHANCED BOOST TECHNOLOGY ON CT PULMONARY ARTERY IMAGING WITH LOW CONTRAST VOLUME AND LOW FLOW RATE | R1-SSCH09-3


Dingxiang Xie – The First Affilated Hospital, Sun Yat-sen University

Conclusions: The application of DLR combined with CE-boost technology for CTPA can enhance pulmonary artery visualization and achieve better image quality, in condition of substantially reduced contrast flow rate and total volume.

Clinical Relevance/Application: This technological approach holds particular promise for patients susceptible to contrast-induced nephropathy or elderly individuals with compromised vascular conditions, potentially aiding in the detection of sub-segmental pulmonary artery embolism. Considering the current research findings, there remains potential for further reduction in radiation dose and contrast volume.

Scientific Presentation Thursday, Dec 5 | 8:00 AM - 9:00 AM CST | E451A

EVALUATION OF A WHOLE HEART MOTION COMPENSATION METHOD FOR CARDIAC CT | T2-SPPH-11


Qiulin Tang, PhD – Canon Medical Research USA

Conclusions: Evaluations based on clinical data have been performed for a cardiac motion-compensated reconstruction method. The result shows that the method significantly reduces cardiac motion artifacts across all region.

*Clinical Relevance/Application: Cardiac Computed Tomography Angiography (CTA) stands as one of the foremost applications in cardiac CT imaging, pivotal for precise clinical diagnosis of cardiac ailments. Mitigation of motion artifacts holds paramount importance in ensuring the fidelity and accuracy of diagnostic outcomes.

Poster Presentation Tuesday, Dec 3 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

NEW HORIZON OF CT IMAGE PROCESSING: INNOVATIVE DOSE REDUCING CTA SUBTRACTION TECHNIQUE | VAEE-4


Nobuaki Idobata - Uetuki Clinic

Teaching Points: Subtraction Computed Tomography Angiography (CTA) can accurately evaluate the vascular lumen in cases of high calcification, however, to perform the subtraction technique it is necessary to obtain non-contrast images as a mask in addition to the contrast enhanced images. Using recently developed image positioning technique in helical imaging, we propose a CTA subtraction that does not require the non-contrast mask images.

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

CT IMAGING USING VARIABLE HELICAL PITCH (VHP) SCANNING FOR LOWER EXTREMITY ARTERIAL DISEASE (LEAD): REDUCED CONTRAST MEDIUM DOSE, IMPROVED IMAGE QUALITY AND DIAGNOSTIC ACCURACY | M2-SPVA-2


Xiaoshi Li - Xi'an Daxing Hospital

Conclusions: In lower extremity arterial CTA scans, the use of VHP technology can significantly reduce the CM dose, improve subjective image quality, and enhance the objective image quality of distal small vessels.

Clinical Relevance/Application: It has the potential to replace conventional scans in clinical practice and become the standard technique for lower extremity arterial CT scans.

Educational Poster Monday, Dec 2 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

CLINICAL APPLICATION OF DARK-BLOOD IMAGING IN HEAD AND NECK CT ANGIOGRAPHY: EFFECT ON IMAGE QUALITY AND PLAQUE VISIBILITY | T5B-SPVA-3


Junchao Li - The First Affiliated Hospital of Xinjiang Medical University

Conclusions: Compared to conventional triphasic head and neck CT angiography, the CE-boost dark-blood imaging demonstrated a significant improvement in image quality and visibility for both calcified and non-calcified plaque assessment.

Clinical Relevance/Application: The CE-boost dark blood imaging technique enables rapid imaging of the carotid artery wall and improves the detection rate of plaques. Combining this technique with conventional CTA can lead to a more comprehensive identification of plaques.

Poster Presentation Tuesday, Dec 3 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

THE DIAGNOSTIC VALUE OF SUBTRACTION CT IN EVALUATING THE EFFICACY OF TRANSCATHETER ARTERIAL CHEMOEMBOLIZATION IN HEPATOCELLULAR CARCINOMA PATIENTS COMPARED TO CONVENTIONAL CT AND MRI | W5A-SPIR-6


Fengxian Hu, MMEd - Qingdao Third People's Hospital

Conclusions: The multiphase CT with subtraction images can significantly improve the sensitivity and accuracy of LR-TR algorithm , especially in lesions with a diameter ≤ 6mm. Subtraction images can improve the diagnostic confidence for detecting viable tumor in TACE-treated patients.

Clinical Relevance/Application: Comparing with multiphase CT and MR, the multiphase CT with subtraction images provide high accuracy for evaluation of viability of tumor in HCC patients after TACE.

Poster Presentation Wednesday, Dec 4 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

THE VALUE OF CONTRAST ENHANCED BOOST TECHNIQUE IN CT ANGIOGRAPHY OF AORTIC ANEURYSMS | R2-SPVA-1


Dingxiang Xie - The First Affiliated Hospital, Sun Yat-sen University

Conclusions: The application of the CE-Boost technique significantly enhances both the subjective and objective image quality of aortic aneurysm CTA, thereby facilitating a more precise evaluation of overall aortic conditions. Additionally, the utilization of ΔCT values following CE-Boost assists in a more accurate differentiation of uneven contrast mix and thrombus.

Clinical Relevance/Application: The CE-Boost technique effectively enhances the image quality and diagnostic accuracy of aortic aneurysms, thereby contributing to the reduction of scan repetitions (venous phase) and radiation exposure.

Poster Presentation Thursday, Dec 5 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

A RETROSPECTIVE STUDY OF PULMONARY EMBOLUS RADIOMICS MODEL IN PREDICTING THROMBOLYSIS OUTCOMES IN ACUTE PULMONARY EMBOLISM | R5B-SPCH-7


Lijuan Duan - Yunnan University Affiliated Hospital

Conclusions: The nomogram, constructed based on the prediction model established by pulmonary embolus radiomics and machine learning and clinical indicators, has been demonstrated to be an effective tool for predicting the thrombolysis of pulmonary embolism.

Clinical Relevance/Application: The radiomic model can be used to analyze the emboli of APE patients in time and in advance, and nomogram can further assist the clinical selection of appropriate treatment.

Poster Presentation Thursday, Dec 5 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

DIAGNOSTIC PERFORMANCE OF CT-DERIVED EXTRACELLULAR VOLUME FRACTION IN AORTIC STENOSIS, CARDIAC AMYLOIDOSIS AND DUAL PATHOLOGY | W2-SPCA-3


Shinpei Yamaguchi – Kumamoto University Hospital

Conclusions: CT-ECV of patients with AS-ATTR was lower than that of patients with lone ATTR, and the diagnostic performance and optimal cutoff values of CT-ECV for differentiating between patients with AS-ATTR and lone AS were lower than those between patients with lone ATTR and lone AS.

Clinical Relevance/Application: The findings from this study may contribute to the accurate detection of cardiac amyloidosis and lead to early drug treatment.

Poster Presentation Wednesday, Dec 4 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

WISDOM OF EVALUATING ENDOLEAK WITH 4D CT -HOW TO RELIABLY ACQUIRE IMAGES WITH LOW RADIATION DOSE AND SHORT SCANNING TIME- | VAEE-5


Takashi Nishiyama - Wakayama Medical University Hospital

Teaching Points: It is possible to differentiate the type of Endoleak using 4D CT However, there are issues with the extension of scanning time and the tradeoff relationship between radiation dose and number of frames. This time, we have developed imaging method that reduces radiation dose by 40% and increases the number of frames by 2.25 times compared to conventional methods.

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

PROPOSED CT PERFUSION PROTOCOL: SUB-MILLISIEVERT APPROACH WITH FOUR-DIMENSIONAL SIMILARITY FILTER | T5A-SPCA-2


Kazuki Yoshida, MD, PhD - Ehime University

Conclusions: The MS-CTP protocol improved the image quality and probability of capturing the optimal scan timing compared with SS-CTP, with significantly higher sensitivity for obstructive CAD compared with dynamic those of CTP and SS-CTP.

Clinical Relevance/Application: The MS-CTP protocol may reduce radiation exposure and increase CTP’s clinical value in managing ischemic heart disease.

Educational Poster Tuesday, Dec 3 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

CANON Advanced CT Imaging

CT TAKES THE LEAD WITH LOWER DOSES THAN CONVENTIONAL X-RAY: CHEST CT AT 0.098 MSV | M7-SSCH04-6


Patrik Rogalla, MD, MBA– UHN

Conclusion: By combining an additional silver beam filter with DLR, the effective radiation dose in non-contrast chest CT can be reduced to dose parity with a single chest X-ray while providing subjectively sufficient image quality for assessing the lung parenchyma post biopsy. Further research is needed to validate these findings across diverse clinical indications and patient populations.

Clinical Relevance: For a typical adult weighing up to 90kg body weight, the CT radiation dose can be reduced to less than 0.1 mSv while maintaining diagnostic image quality for excluding post chest biopsy complications, equivalent to a single chest X-ray.

Scientific Presentation Monday, Dec 2 | 3:00 PM - 4:00 PM CST | E451A

RADIATION-INDUCED IMAGING CHANGES OF THORACOLUMBAR SPINE: WHAT RADIOLOGISTS NEED TO KNOW? | ROEE-5


Jian Guan, MD, PhD – The First Affiliated Hospital, Sun Yat-sen University

Teaching Points:

  1. To review the pathological changes and imaging appearance of thoracolumbar spine after radiotherapy.
  2. To know special imaging appearance of thoracolumbar spine after radiotherapy.
  3. To realize the relationship between involved extent of spine after radiotherapy and radiation disease.
  4. To learn diagnosis and differential diagnosis for radiotherapy related complications by imaging appearance of thoracolumbar spine.

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

DEEP LEARNING-BASED MOTION CORRECTION (CLEAR MOTION): CAPABILITY OF QUANTITATIVE AND QUALITATIVE IMAGE QUALITY IMPROVEMENTS ON CHEST CT IN PATIENTS WITH THORACIC DISEASES | T5A-SPPH-8


Yoshiyuki Ozawa, MD, PhD – Fujita Health University

Conclusions: DL-based motion correction is considered as more useful for image quality improvement than conventional method on chest CT in patients with various thoracic diseases.

Clinical Relevance/Application: DL-based motion correction is considered as more useful for image quality improvement than conventional method on chest CT in patients with various thoracic diseases.

Poster Presentation Tuesday, Dec 3 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

ARTIFICIAL INTELLIGENCE IN IMAGING DIAGNOSIS OF LIVER TUMORS: CURRENT STATUS AND FUTURE PROSPECTS | GIEE-121


Masatoshi Hori, MD, PhD – Kobe University Hospital

Teaching Points:

  1. Explore artificial intelligence (AI) applications in liver tumor imaging, including lesion detection, differentiation, segmentation, image quality enhancement, and radiomics.
  2. Examine the integration of AI in liver tumor diagnostics within clinical practice.
  3. Discuss how radiologists utilize AI in diagnosing liver tumors.
  4. Review future advancements in AI technologies and their potential impact on patient management.

Education Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

ULTRA-HIGH-RESOLUTION CT FINDINGS OF PRE-INVASIVE ADENOCARCINOMA AND MINIMALLY-INVASIVE ADENOCARCINOMA OF THE LUNG | CHEE-14


Hiroshi Moriya, MD, PhD – Ohara General Hospital

Teaching Points: With the spread of CT screening, the chances of small lung nodule and early lung cancer are increasing. Therefore, the following points are presented by comparing the ultra-high resolution CT (UHRCT) and pathological findings of resected specimens.

  1. Classification of Lung Adenocarcinoma
  2. Pathological invasion criteria of lung adenocarcinoma
  3. Spatial resolution of UHRCT
  4. UHRCT findings of of pre-invasive adenocarcinoma and minimally-invasive adenocarcinoma
  5. Comparison of UHRCT findings and pathological findings

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

CLINICOPATHOLOGICAL CORRELATION OF MULTIPLE CYSTIC LESIONS IN THE LUNG | CHEE-5


Tae Iwasawa, MD, PhD – Kanagawa Cardiovascuiar and Respiratory Center

Teaching Points: A cyst is any circumscribed and well-defined air-containing structure in the lung parenchyma. Cystic lesions occur due to various mechanisms such as airway obstruction with distal airspace dilatation (a check-valve mechanism), necrosis of airway walls, or lung parenchymal destruction by proteases. Cysts typically contain air, although, in some cases, they may be filled with fluid.In this exhibit, we will present computed tomography images and the histology of patients with multiple cystic lesions to explore their pathogenesis. Some cystic lesions increase in size over time, and we will illustrate the progression of the cystic lesions.

Education Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST LEARNING CENTER

ENERGY SAVING STRATEGY USING OPTIMAL TUBE VOLTAGE AND TUBE CURRENT IN ABDOMINAL CT - PHANTOM STUDY | M5B-SPNPM-1


Jeong Woo Kim, MD, PhD – Korea University Guro Hospital

Conclusions: Compared to the protocol using 80 kVp and 500 mA, the protocol using 100 kVp and 200 mA had lower DLP and noise, and nearly 50% lower energy consumption without compromising CNR of FLLs.

Clinical Relevance/Application: Low tube voltage and high tube current have been reported to increase the CNR of hepatic hypervascular tumor while reducing patient radiation dose. Considering energy consumption, the combination of moderate tube voltage and moderate tube current may be a candidate for an energy saving strategy while reducing radiation dose and image noise and without compromising CNR.

Poster Presentation Monday, Dec 2 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

REDUCTION OF CT ARTIFACTS FROM THE LATEST-GENERATION LEADLESS PACEMAKER USING SINGLE-ENERGY METAL ARTIFACT REDUCTION ALGORITHM | W5B-SPCA-2


Hu Yahui – The First Affiliated Hospital of Xinjiang Medical University

Conclusions: SEMAR provides effective reduction of metal artifacts from leadless PMs. The artifact-impaired tissues are fewer on leadless PMs than that on transvenous PMs.

Clinical Relevance/Application: The results of this study demonstrate that SEMAR efficiently reduces metallic artifacts from leadless PMs, thereby improving the image quality of CCTA (Fig. 1-2). SEMAR can also enhance the diagnostic confidence of the cardiac and extra-cardiac complications associated with transvenous and leadless PMs.

Poster Presentation Wednesday, Dec 4 | 12:45 PM - 1:15 PM CST | LEARNING CENTER

VALUE OF INTRALESIONAL AND PERILESIONAL RADIOMICS FOR PREDICTING THE BIOACTIVITY OF HEPATIC ALVEOLAR ECHINOCOCCOSIS | S3A-SPGI-11


Simiao Zhang – The First Affilated Hospital of Xinjiang Medical University

Conclusions: The nomogram model based on clinical and GPLV radiomic signatures show the best performance in prediction of the bioactivity of HAE. Radiomics including perilesional tissue can significantly improve the prediction efficacy of HAE bioactivity.

Clinical Relevance/Application: To explore the value of intralesional and perilesional radiomics based on CT in predicting the bioactivity of HAE. It is expected to achieve preoperative prediction of the bioactivity of HAE, so as to guide clinical treatment better.

Poster Presentation Sunday, Dec 1 | 11:45 AM - 12:15 PM CST |LEARNING CENTER

PATIENT-SPECIFIC ANALYSIS OF IMAGE QUALITY OF VIRTUAL MONOCHROMATIC IMAGES AND ORGAN DOSES IN ABDOMINAL DUAL ENERGY CT EXAMINATIONS | S3B-SPPH-4


Keisuke Fujii, PhD – National Cancer Center Hospital East

Conclusions: 60 and 70 keV VM images from abdominal DECT examinations provide higher CNR than SECT images while liver doses in DECT examinations are similar to those in the SECT examinations.

*Clinical Relevance/Application: 60 and 70 keV VM images for patients in abdominal DECT examinations can be useful in the detection of low contrast lesions because CNR of the VM images are higher than SECT images.

Poster Presentation Sunday, Dec 1 | 12:15 PM - 12:45 PM CST | LEARNING CENTER

RADIATION PNEUMONITIS IN LOCALLY ADVANCED NON-SMALL CELL LUNG CANCER PATIENTS WITH ARTIFICIAL-INTELLIGENCE SCREENED INTERSTITIAL LUNG DISEASE | ROEE-7


Patrik Rogalla, MD, MBA – UHN

Teaching Points:

  • Interstitial lung disease (ILD) complicates cancer treatment
  • ILD increases lung cancer risk
  • AI can identify patients with ILD from planning CTs (in-house development)
  • Patients with AI-screened ILD have stronger correlations between dosimetric parameters and the development of Grade ≥2 and Grade ≥3 radiation pneumonitis
  • Screening for ILD helps consider the application of different dose constraints in lung cancer patients to mitigate the increased risk of severe toxicity

Educational Poster Sunday, Dec 1 | 7:30 AM - 6:00 PM CST | LEARNING CENTER

CYSTIC LUNG DISEASES: BLEND OF CLASSIC KNOWLEDGE AND CONTEMPORARY INSIGHTS | CHEE-118


Jonatas Favero Prietto Dos Santos, MD

Teaching Points:

  • A cyst appears as a circumscribed parenchymal low attenuating area, usually with thin and regular walls.
  • Solitary or few small thin-walled lung cysts incidentally detected on CT may be part of the normal aging process.
  • In scenarios involving young patients with multiple lung cysts, careful evaluation of cyst characteristics, distribution, and ancillary radiologic findings is paramount.
  • Cystic lung cancers are an underrecognized condition on CT, often missed at early stages. - Suspicious cyst features include thick or asymmetric walls, associated nodules, internal septations, and growth.
  • A change in the morphologic features of a cyst and surrounding tissues should raise suspicion for cancer.

Educational Poster Sunday, Dec 1 |7:30 AM - 6:00 PM CST| LEARNING CENTER

ENHANCING THE PROCESS FOR ENHANCED CT SCANS IN EMERGENCY SETTINGS - A REVIEW OF PRACTICE AND PROCESSES | R2-QI-1


Hayley Panet – UHN

Discussion: Enhancing the IV contrast policy to meet the needs of the fast-paced emergency department in alignment with updated national standards has enhanced CT scanner efficiency, scan turnaround times, advanced technologist practice, and has improved communication and partnership between medical imaging and emergency department staff.

Poster Presentation Thursday, Dec 5 | 9:00 AM - 9:30 AM CST | LEARNING CENTER

CANON Ultrasound

DEVELOPMENT OF AN ARTIFICIAL INTELLIGENCE-BASED SUPPORT SYSTEM FOR FETAL CARDIAC ULTRASOUND SCREENING | S3A-SPOB-2


Masaaki Komatsu, MD, PhD, Presenter | Naoaki Harada, Co-author | Reina Komatsu, Co-author | Ryu Matsuoka, Co-author | Kazuki Iwamoto, Co-author | Akihiko Sekizawa, Co-author | Ryuji Hamamoto, MD, Co-author

*Purpose: Fetal cardiac ultrasound screening is generally performed on all fetuses, however, the global prenatal detection rate for congenital heart disease (CHD) is still insufficient due to technical differences among examiners. Here, we have developed an artificial intelligence (AI)-based support system for this ultrasound screening toward clinical application.

*Methods and Materials: We enrolled fetal cardiac ultrasound videos of 262 normal cases and 38 CHD cases who underwent fetal cardiac ultrasound screening in the second trimester. This dataset included 3 vendors; Voluson® P8/E8/E10/Expert22 (GE Healthcare, Chicago, IL, US), Aplio® i700/i900 (Canon Medical Systems, Otawara, Tochigi, Japan), and Arietta® 70 (Fujifilm Healthcare, Tokyo, Japan). The correct positions of 18 cardiac substructures were annotated with bounding boxes. The time-series information of these detection results was displayed using barcodes and detection rate graphs. First, the diagnostic accuracy of the automatic normal substructure detection was assessed. Then, we attempted to verify whether our AI system could enhance the screening performance of examiners. A comparative study was performed on the screening performance of 50 examiners (6 experts, 30 fellows, and 14 residents) with and without the AI system. In each situation, the examiners were given 60 randomly numbered videos of the normal and CHD cases and determined normal or abnormal. The multiplicity of statistical analyses was controlled by a fixed sequence procedure.

*Results: The sensitivity for the normal substructure detection was 93.5%, showing superiority to the sensitivity threshold of 80% based on preliminary test results (p < 0.001). The specificity was 95.9%, indicating superiority to the specificity threshold of 80% (p < 0.001). In the subgroup analyses, both the sensitivity and specificity for the normal substructure detection of every vendor and machine achieved over 90%. Furthermore, the sensitivity of the screening of unskilled doctors (fellows and residents) using the AI system was 78.4%, showing superiority over the sensitivity of the examiner alone (p = 0.005). The specificity was 86.5%, showing superiority to the specificity of the examiner alone (p < 0.001). On the other hand, the screening performance of experts showed no significant difference with and without the AI system.

*Conclusions: Our AI system demonstrated sufficient accuracy in the automatic normal substructure detection and could enhance the screening performance of unskilled doctors in fetal cardiac ultrasound screening.

*Clinical Relevance/Application: This system aims to support fetal cardiac ultrasound screening in a clinical scenario.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=S3A-SPOB-2

INITIAL EVALUATION OF SENTINEL LYMPH NODES IN PATIENTS WITH CERVICAL AND VULVAR CANCER USING LYMPHOSONOGRAPHY COMPARED TO SURGICAL PATHOLOGY | S3B-SPOB-3


Priscilla Machado, MD, Presenter | Trang Vu, Co-author | Norman G. Rosenblum, MD, PhD, Co-author | Aaron Shafer, Co-author | Flemming Forsberg, PhD, Co-author

*Purpose: The identification of the sentinel lymph node (SLN) is an important aspect of predicting outcomes for patients with cancer. Lymphosonography is a contrast-enhanced US modality that uses subcutaneous injection of US contrast agents to localize the SLNs and it has been studied as a novel technique to lymphatic mapping. The objective of this study was to evaluate this technique in subjects diagnosed with cervical or vulvar cancer scheduled for surgery with SLN excision comparing the lymphosonography findings with pathology.

*Methods and Materials: Subjects diagnosed with cervical or vulvar cancer scheduled for surgery with SLN excision were eligible to be enrolled in this ongoing, IRB-approved study. The US contrast agent Sonazoid (GE Healthcare, Oslo, Norway) was administrated in 4 aliquots at 12, 3, 6, and 9 o’clock positions around the primary tumor by the clinical surgical team, with 0.25 ml for each aliquot for a total dose of 1.0 ml. Real time contrast enhanced US imaging (CEUS) was performed to identify SLNs that demonstrated contrast-enhancement. Results were recorded and compared to the pathology results obtained after the surgical procedure. An Aplio i800 Prism scanner (Canon Medical Systems USA, Tustin, CA) with curvilinear (8C1) and linear (18L5) probes with CEUS capabilities were used during the study.

*Results: A total of 11 SLNs in 5 subjects were surgically excised. Three subjects had cervical cancer with iliac SLNs being excised, 8 SLNS were sent to pathology. From the 8 SLNs, 6 were determined to be benign by pathology and 2 were determined to be malignant. Two subjects had vulvar cancer with inguinal SLNs being the ones excised, 3 SLNS were sent to pathology. From the 3 SLNs, 2 were determined to be benign by pathology, while 1 was determined to be malignant by pathology. Lymphosonography identified 17 SLNs in the 5 subjects. Comparing the surgically excised SLNs and the ones identified by lymphosonography showed that10 of the 11 SLNs surgically excised or 91% were identified by lymphosonography. Lymphosonography identified all the benign SLNs and 2 out of the 3 malignant SLNs.

*Conclusions: This initial evaluation on the use of lymphosonography to identify SLNs in subjects with cervical or vulvar cancer undergoing surgery with SLN excision demonstrate that lymphosonography identified 10 out of the 11 SLNs surgically excised, including all the benign SLNs and 2 malignant SLNs.

*Clinical Relevance/Application: Lymphosonography is a US modality that uses US contrast agents to identify SLNs, which is an important aspect of predicting outcomes for patients with cancer.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=S3B-SPOB-3

THE DIFFERENCE BETWEEN POWER DOPPLER AND SUPERB MICROVASCULAR IMAGING IN THE ASSESSMENT OF RHEUMATOID ARTHRITIS AND OTHER ARTHROPATHIES | S4-SSMK02-4


Marcos Jimenez Vazquez, MD, Presenter | Carmen Mbongo, MD, Co-author | Pablo Del Nido Recio, Co-author | Manuel Rafael Lopez De La Torre Carretero, Co-author | Alberto Paternain, MD, Co-author | Jesus Damaso Aquerreta, MD, PhD, Co-author

*Purpose: Superb Microvascular imaging (SMI) has been proven to be a reliable Doppler technique and to have more sensitivity when detecting active synovitis, compared to Power (conventional) Doppler (PD). SMI upgrades Doppler activity in comparison to PD. This study aims to analyze whether the upgrade Doppler activity, when comparing SMI and PD, is significantly different in Rheumatoid Arthritis (RA), compared to other arthropathies.

*Methods and Materials: Between May 2023 and April 2024, we prospectively analyzed a cohort of 57 joints of 21 different patients who underwent ultrasound exams at our center. Most of them were previously diagnosed with RA and other arthropathies, such as Osteoarthritis, Psoriatic Arthritis, Gout, etc. PD and SMI imaging were obtained in all joints and the individual grades for Doppler Activity were registered for each joint with active synovitis, raging from 0 to 3, according to the EULAR - OMERACT US Score (European League Against Rheumatism - Outcome Measure in Rheumatology):

  • Grade 0: no Doppler activity.
  • Grade 1: up to three single Doppler spots.
  • Grade 2: greater than Grade 1 but <50% of Doppler Signals in the total affected hypoechoic region.
  • Grade 3: >50% of Doppler signals in the affected region.

Means of each grading for PD, SMI and the difference between PD and SMI were registered. Mann-Whitney U test was applied to calculate means in independent samples. Two-tailed p-values of <0.05 were considered statistically significant.

*Results: 21 joints of 8 patients with RA and 36 joints of 13 patients with other arthropathies were studied. Metacarpophalangical joints were the most frequently analyzed in the RA group (11), and interphalangeal joints were the most frequently analyzed in the other group (18). No statistically significant differences were found between the group with RA and the group with other arthropathies in terms of age (60.38 vs. 54.17 years, p = 0.41) or gender (5 women vs. 7 women, p = 1). When comparing Doppler activity, the mean PD was significantly higher in the RA group compared to the other group (1.14 vs. 0.75, p = 0.043). When upgrading the Doppler activity with SMI, we did not find statistically significant differences (2.52 vs 2.69, p = 1.04). Nevertheless, when comparing the mean of the difference between SMI and PD, it was significantly lower in the RA group (1.38 vs. 1.94, p = 0.016).

*Conclusions: Our results indicate that a higher upgrade between PD and SMI exams is more likely seen in arthropathies such as oligoarthritis, Psoriatic Arthritis or Gout, rather than in RA.

*Clinical Relevance/Application: A higher difference between the PD and SMI grading could be used as a feasible technique to assess the diagnosis of an unknown arthropathy, considering the location of the affected joint and other clinical and systemic features.

THE EFFECT OF BREATH CONTROL ON THE EVALUATION OF HEPATIC SHEAR WAVE ELASTICITY AND DISPERSION IN PEDIATRIC PATIENTS | M5B-SPPD-2


Chanyoung Rhee, MD, Presenter

*Purpose: This study aims to assess the impact of breath control on the reliability and accuracy of 2D Shear Wave Elastography (SWE) and Shear Wave Dispersion (SWD) in diagnosing pediatric liver disease, focusing on the optimization of elastography protocols.

*Methods and Materials: The study cohort comprised 225 pediatric patients who underwent 2D-SWE and SWD assessments between September 2021 and February 2023, with ultrasound evaluations performed in both gentle free-breath and breath-hold states. Within this group, 30 patients also underwent MR elastography and were subjected to a second ultrasound examination by two radiologists, spaced one week apart. The study assessed correlations between breathing conditions, compared the Area Under the Curve (AUC) with MRE as the reference standard, and conducted subgroup analyses to evaluate inter-rater agreement and test-retest reliability.

*Results: Liver stiffness (LS) and liver dispersion (LD) in free-breath and breath-hold states showed a marginally significant difference (5.31 ± 1.28 vs. 5.09 ± 1.27 kPa, p = .08; 12.17 ± 2.04 vs. 11.75 ± 2.01 m/s/kHz, p = .04), with high correlations between each breathing condition (LS, r = .69, P < .001; LD, r = .64, P < .001). The AUC for LS showed no significant difference between free-breath (0.67) and breath-hold (0.71) (p = .175). Inter-rater agreement revealed ICCs of 0.83 (free-breath) and 0.94 (breath-hold) for LS, and 0.70 (free-breath) and 0.85 (breath-hold) for LD, all significant (P < .001). Test-retest reliability for LS was 0.88 in both conditions, while for LD, it was 0.70 in free-breath and 0.74 in breath-hold, all significant (P < .001), underscoring higher but not substantial reliability in breath-hold conditions.

*Conclusions: While maintaining breath-hold during ultrasound elastography slightly increases reliability, the minimal difference from results measured in free-breath conditions renders the latter acceptable, particularly when a breath-hold proves to be difficult.

*Clinical Relevance/Application: In pediatric patients, where breath control is relatively challenging, breath-holding is not essential for reliably measuring 2D-SWE and SWD.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=M5B-SPPD-2

ERECTILE DYSFUNCTION EVALUATION USING SMI AND SWE US TECHNIQUES | T2-SPGU-8


Priscilla Machado, MD, Presenter | Ji-Bin Liu, MD, Co-author | Aaron Hochberg, Co-author | Paul H Chung, MD, Co-author | Flemming Forsberg, PhD, Co-author

*Purpose: Erectile dysfunction (ED) is a highly prevalent condition that affects approximately 30 million men in the United States. While penile duplex ultrasound (DUS) has long been considered standard of care, recent advances in ultrasound (US) technology warrant investigation. Superb microvascular imaging (SMI) is a new microvascular flow imaging mode, which uses adaptative clutter suppression to extract flow signals from large and small vessels represented as a color overlay image or as a grayscale map of flow, named cSMI and mSMI, respectively. Shear wave elastography (SWE) is a technique to measure tissue stiffness. This study investigated the use of SMI and SWE to evaluate ED in subjects undergoing penile US pre and post injection of vasoactive agent.

*Methods and Materials: To date 20 adult male subjects (≥18 years old) with clinical ED defined by International Index of Erectile Function (IIEF) scores were enrolled in this ongoing, IRB-approved study. Subjects underwent spectral Doppler, CDI, PDI, cSMI, mSMI and SWE examination of the penis before and after intracavernosal injection of vasoactive agent using Aplio i800 Prism US system (Canon America Medical Systems, Tustin, CA) with an i18LX5 linear probe. Data was analyzed comparing pre (flaccid penis) and post injection (erect penis) for all images modalities using t-tests.

*Results: The mean age of subjects was 51 years (range: 31 -70). The mean IIEF score was 10 (range: 3-19; lower values indicate more severe ED). The mean spectral Doppler PSV ± SD were 23.64 ± 10.98 cm/s pre injection and 40.47 ± 17.94 cm/s post injection (p < 0.0001), mean RI ± SD were 0.90 ± 0.07 pre injection and 0.88 ± 0.08 post injection (p = 0.28). The mean cSMI pixel counts ± SD were 99.64 ± 6.64 pre injection and 101.10 ± 6.14 post injection (p = 0.69). The mean mSMI pixel counts ± SD were 124.10 ± 3.72 pre injection and 124.0 ± 3.67 post injection (p = 0.89). The mean SWE ± SD were 38.44 ± 20.14 kPa pre injection and 39.63 ± 22.58 kPa post injection (p = 0.82).

*Conclusions: The values of SMI pixel counts showed no significant statistical difference between pre and post injections for cSMI or mSMI (p > 0.65). The evaluation of tissue stiffness using SWE also showed no significant statistical difference pre and post injections (p = 0.20) The results are encouraging with the findings indicating that the use of SMI and SWE without injection of vasoactive agent (flaccid penis) may be a way to diagnose ED.

*Clinical Relevance/Application: Understandably, the sensitive nature of ED, intimate process of penile US and the need to inject medication into the penis to induce erection makes for a less than ideal patient experience. SMI and SWE may be able to identify characteristics consistent with ED in flaccid penis facilitating clinical practice and improving patient experience.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=T2-SPGU-8

COMBINING CEUS AND CT/MRI LI-RADS MAJOR IMAGING FEATURES: DIAGNOSTIC ACCURACY FOR CLASSIFICATION OF INDETERMINATE LIVER OBSERVATIONS IN PATIENTS AT RISK FOR HCC | T5A-SPGI-7


Tania Siu Xiao, MD, Presenter | Cristina Kuon Yeng Escalante, Co-author | Aylin Tahmasebi, MD, Co-author | Yuko Kono, MD, PhD, Co-author | Andrej Lyshchik, MD, PhD, Co-author

*Purpose: This study aimed to determine diagnostic accuracy of combining contrast-enhanced ultrasound (CEUS) and CT/MRI Liver Imaging Reporting and Data System (LI-RADS) major imaging features for the improved classification of indeterminate liver observations.

*Methods and Materials: A retrospective analysis using a database from a prospective study conducted at 11 centers in North America and Europe from 2018 through 2022 included a total of 109 participants at risk for hepatocellular carcinoma (HCC) who had liver observations with indeterminate characterization on both CEUS and CT/MRI (LR3, LR-4, and LR-M). The individual LI-RADS major features for HCC on both CEUS and CT/MRI were extracted and analyzed in various combinations. Biopsy, explant histology, and follow-up CT/MRI were used as the final reference standard. The diagnostic performance of LI-RADS major features combinations for definitive diagnosis of HCC (LR-5) were calculated. A reverse, stepwise logistical regression sub-analysis was also performed.

*Results: This study included 114 observations indeterminate on both CT/MRI and CEUS. These observations were initially classified as LR-3 (n=37), LR-4 (n=41), and LR-M (n=36) on CT/MRI and LR-3 (n= 48), LR-4 (n= 36), LR-M (n= 29), and LR-TIV (n=1) on CEUS. Of them, 43.0% (49/114) were confirmed as HCC, 37.3% (43/114) were non-malignant, and 19.3% (22/114) were non-hepatocellular malignancies. The highest diagnostic accuracy among the combinations of imaging features was achieved in CT/MRI LR-3 observations, where the combination of CEUS arterial phase hyper-enhancement (APHE) + CT/MRI APHE had 96.7% specificity, 75.0% positive predictive value and 86.5% accuracy for HCC. CEUS APHE was the variable that showed synergism between benign versus malignant and HCC versus any other observation.

*Conclusions: The combination of LI-RADS major features on CT/MRI and CEUS showed higher specificity, positive predictive value, and accuracy compared to individual modalities' assessments, particularly for CT/MRI LR-3 observations. These combinations can produce a definitive diagnosis of HCC in a small percentage of cases with inconclusive imaging on individual modalities.

*Clinical Relevance/Application: By using the combination of CEUS and CT/MRI LI-RADS major features, a definitive diagnosis of HCC can be achieved in cases where individual modalities produce inconclusive results, particularly in LR-3 observations. This comprehensive approach may lead to early intervention.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=T5A-SPGI-7

CONFOUNDERS OF ULTRASOUND ATTENUATION IMAGING IN A LINEAR PROBE USING THE CANON APLIO I800 SYSTEM: A PHANTOM STUDY | T6-SSPH08-5


Alexander Martin, DPhil, BSc, Presenter

*Purpose: There have been studies showing attenuation imaging (ATI) with ultrasound as an approach to diagnose liver diseases such as steatosis or cirrhosis. So far, this technique has only been used on a convex probe. The goal of the study was to investigate the feasibility of ATI measurements using the linear array on a canon Aplio i800 scanner on certified phantoms.

*Methods and Materials: Three certified liver tissue attenuation phantoms were measured in five different positions using a linear probe. The effects of positioning and depth were explored and compared. The values were compared to the certified expected value for each phantom as well as the different measurement values for each measurement position.

*Results: The ATI measurements on phantoms significantly affected the different probe positions and region of interest (ROI) depths. Values in the center with the probe perpendicular to the phantom were closest to certified values. Median values at 2.5-4.5 cm depth for phantoms 1 and 2 and 0.5-2.5 cm for phantom 3 were comparable with certified values. Measurements taken at a depth greater than 6 cm in any position were the least representative of the certified values (p-value < 0.01) and had the widest range throughout the sessions.

*Conclusions: ATI measurements can be performed with the linear probe in phantoms; however, careful consideration should be given to depth dependency, as it can significantly affect measurement values. Remaining measurements at various depths within the 0.5-6.0 cm range showed deviation from the certified values of approximately 25%.

*Clinical Relevance/Application: Developing a protocol for using a linear probe for attenuation imaging with potential uses for superficial structures such as MSK, breast or thyroid, as well as possible uses in paediatric medicine.

ENHANCED DIAGNOSTIC PERFORMANCE OF SHEAR-WAVE DISPERSION IMAGING IN BREAST MASSES: COMPARISON WITH SHEAR-WAVE ELASTOGRAPHY ACCORDING TO LESION SIZE, DEPTH, AND BREAST THICKNESS | W3-SSBR08-5


Eunkyung Woo, Co-author | Bo Kyoung Seo, MD, PhD, Presenter | Min Sun Bae, MD, PhD, Co-author | Kyu Ran Cho, MD, PhD, Co-author | Ok Hee Woo, MD, PhD, Co-author | Sung Eun Song, MD, PhD, Co-author

*Purpose: Human breast tissue exhibits viscoelastic properties, but ultrasound (US) assessments have primarily focused on elasticity measurement. Shear wave elastography (SWE) is a commonly used US technique for quantifying tissue elasticity, but its accuracy can be affected by lesion size, depth, and breast thickness. Recent technological advancements allow for the automatic quantification of viscosity through the shear-wave dispersion slope (SWD) in SWE. This study aims to evaluate the diagnostic performance of SWD in distinguishing between benign and malignant breast masses and to explore its supplementary role.

*Methods and Materials: From August 2020 to August 2021, we retrospectively evaluated 606 consecutive patients scheduled for ultrasound-guided biopsy. We performed SWD and SWE before the procedure. A total of 630 breast masses were included in the study. For each mass, we measured the average tumor elasticity (Etumor) and tumor-to-fat elasticity ratio (Eratio) using SWE, as well as the average tumor dispersion (Dtumor) and tumor-to-fat dispersion ratio (Dratio) using SWD. We compared the diagnostic performance of SWE and SWD using the area under the curve (AUC). Additionally, we conducted a Mann-Whitney U test to determine whether lesion size, depth, and breast thickness independently affected the overall diagnostic performance.

*Results: Of the 630 masses, 395 were benign and 235 were malignant. Mean values of Etumor, Eratio, Dtumor, and Dratio were 11.5 kPa, 2.6, 11.2 m/s/kHz, and 2.1 for benign lesions and 79.5 kPa, 16.4, 27.1 m/s/kHz, 5.0 for malignant lesions (p < 0.05). The overall AUC was significantly superior in SWD (Dtumor vs. Etumor, 0.96 vs. 0.93; Dratio vs. Eratio, 0.95 vs. 0.93) (p < 0.05). In small masses ≤ 10 mm, the sensitivity and positive predictive value of Dtumor were higher than Etumor (70% and 85% vs. 38% and 68%) with the preservation of high specificity (98% vs. 96%) (p < 0.05). In lesions of any depth and breast thickness, Dtumor showed the highest sensitivity and specificity.

*Conclusions: SWD can enhance the diagnostic performance for distinguishing between benign and malignant breast masses compared to SWE across various lesion sizes, depths, and breast thicknesses.

*Clinical Relevance/Application: SWD, obtained automatically during SWE examination, could compensate for the shortcomings of SWE regarding variance in diagnostic performance according to lesion size, depth, and breast thickness.

ASSOCIATION BETWEEN US IMAGING MARKERS AND NECROINFLAMMATION, STEATOSIS, AND FIBROSIS IN PATIENTS WITH SUSPECTED METABOLIC DYSFUNCTION-ASSOCIATED STEATOTIC LIVER DISEASE | R3-SSGI17-5


Katsutoshi Sugimoto, MD, PhD, Presenter | Fuminori Moriyasu, Co-author

*Purpose: To assess the relationship between the multiparametric US imaging markers shear-wave dispersion slope (DS), attenuation coefficient (AC), and shear-wave speed (SWS) and liver histopathological necroinflammation in patients with metabolic dysfunction-associated steatotic liver disease (MASLD).

*Methods and Materials: This international, multicenter prospective study enrolled consecutive biopsy-proven MASLD patients between June 2019 and March 2023. Before biopsy, all participants underwent multiparametric US to obtain DS, AC, and SWS. Multivariable linear regression analyses were performed to assess the association between clinical variables or imaging markers with pathological findings. The diagnostic performance of imaging markers for determining inflammation, steatosis, and fibrosis grades was assessed using the area under the receiver operating characteristic curve (AUC).

*Results: A total of 124 participants (mean age, 53 years ± 15 [standard deviation]; 62 men) were evaluated. Lobular inflammation was associated with DS (P = .02), alanine aminotransferase level (P = .002), and Hispanic ethnicity (P = .047), while AC was associated with steatosis (P < .001), and SWS was associated with fibrosis (P < .001) and body mass index (P = .02). DS achieved an AUC of 0.72 (95% confidence interval (CI): 0.63, 0.82) for identifying participants with inflammation grade ≥ A2 (moderate). AC showed excellent performance for identifying participants with steatosis grade ≥ S1 (mild) (AUC, 0.92 [95% CI: 0.87, 0.97]) while SWS showed excellent performance for identifying participants with fibrosis stage ≥ F2 (periportal fibrosis) (AUC, 0.91 [95% CI: 0.86, 0.96]). The combination of DS, AC, and SWS showed an AUC of 0.81 (95% CI:0.73, 0.89) for MASH diagnosis.

*Conclusions: Among the three US imaging markers, DS, AC, and SWS, only DS was associated with lobular inflammation grade on histology and demonstrated fair diagnostic performance in distinguishing moderate lobular inflammation grade.

*Clinical Relevance/Application: Multiparametric US is useful for the management of patients with MASLD.

EARLY DETECTION OF LIVER FIBROSIS IN PATIENTS UNDERGOING PRE- AND POST-HEMATOPOIETIC STEM CELL TRANSPLANTATION: A PRELIMINARY COMPARISON BETWEEN ULTRASOUND ELASTOGRAPHY AND TRANSIENT ELASTOGRAPHY | R5B-SPGI-10


Claudia Gerlotti, MD, Presenter | Maria Luisa Prieto, Co-author | Angela Garcia, Co-author | Enrique de Miguel Campos, Co-author | Paola Lopez-Gomez, MD, Co-author | Joaquim Amorim Sortino, MD, Co-author | Mihail Poida, MD, Co-author | Maria Camarena, Co-author

*Purpose: Veno-occlusive disease (VOD) is a serious complication following post-hematopoietic stem cell transplantation (HSCT), with a mortality rate exceeding 80%, underscoring the critical need for early detection. Ultrasound and/or elastography suggestive of VOD onset are one of the criteria set by the EBMT. A preliminary comparison was done between ultrasound elastography (UE) and transient elastography (TE) to enhance liver fibrosis (LF) detection.

*Methods and Materials: 43 patients who had undergone HSCT were retrospectively studied from January 2023 to February 2024. Pre- and +14-day post-HSCT evaluation was done with TE (FibroScan) and 2D Shear Wave Elastography (2D-SWE) complemented with B-mode ultrasound and Doppler imaging. Inclusion criteria comprised patients aged more than18 years old, undergoing HSCT for an hematologic condition. LF staging utilized Metavir score and Canon Score categorizing TE and 2D-SWE respectively into F0-F1, F2, F3, and F4.

*Results: The pre-HSCT group showed very poor UE-TE correlation with Lin’s coefficient result equal to 0.19 and Kappa’s coefficient result equal to -0.62. The post-HSCT group exhibited moderate correlation with Lin’s coefficient result equal to 0.59 and Kappa’s coefficient equal to 0.55. Significant dispersion was noted, specially, for grades F2, F3, and F4. Also, there was no statistically significant difference between UE and TE before and +14-day after HSCT.

*Conclusions: Post-HSCT UE and TE demonstrated moderate correlation, highlighting the potential advantages of UE in detecting VOD onset. The observed very poor pre-HSCT correlation may stem from varied evaluation times, patient basal conditions, and technique performance, being necessary for more standardized evaluation guidelines. Additionally, no significant difference was noted pre- and post-HSCT UE and TE, suggesting no onset of classical VOD during this period; however, posterior follow-up of these patients was not included in this analysis.

*Clinical Relevance/Application: Early VOD diagnosis in patients undergoing HSCT is crucial. Ultrasound offers additional information, aiding VOD detection through gray-scale B-mode and Doppler imaging, demonstrating features such as hepatomegaly, splenomegaly, portal vein hypertension sings, hepatic venous abnormalities, gallbladder wall thickening and abdominal free fluid, compared to TE's singular fibrosis grade result.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=R5B-SPGI-10

ULTRA-HIGH-FREQUENCY ULTRASOUND EVALUATION OF CUTANEOUS TUMORS: CORRELATION WITH DERMOSCOPY AND HISTOLOGY | R5B-SPMS-2


Ivana Gibbons, Co-author | Soraia Damiao, MD, Presenter | Leticia Cavalcante, Co-author | Mariana Galupo, Co-author | Laura Sales De Carvalho Lima, MD, Co-author | Almir Bitencourt, MD, PhD, Co-author

*Purpose: To describe the sonographic characteristics of cutaneous tumors in ultra-high-frequency (UHF) ultrasound and compare them with dermoscopy and histology.

*Methods and Materials: This IRB-approved, prospective, single-center study included patients with suspicious skin lesions before surgical resection. All lesions were documented with clinical and dermatoscopic photos using a specific dermatoscope. Subsequently, UHF ultrasound was performed with a dedicated 24 MHz device (Aplio i800, Canon Medical Systems), and B-mode and Doppler parameters were described. After the ultrasonographic examination, the lesions were excised and sent for histopathological examination. All examinations were performed by a single observer.

*Results: 289 lesions were included, comprising 108 benign lesions and 181 malignant lesions, of which 105 were basal cell carcinomas (BCCs), 35 squamous cell carcinomas, 24 melanomas, 2 Merkel cell carcinomas, 15 less common malignant tumors and cutaneous metastases from other tumors. The most frequent benign lesions were melanocytic nevus, dermatofibroma, epidermal cysts, actinic and seborrheic keratoses. Each type of skin lesion had specific imaging features. Most lesions (87.6%) were located in the dermis, 267 (92.9%) were hypoechoic, 245 (84.5%) had homogeneous echotexture, 209 (72.1%) were well-defined, and 146 presented some degree of vascularization (50.3%). All malignant lesions were visualized at UHF ultrasound. In B-mode, benign and malignant lesions appeared as variable shapes with no statistically significant differences between the two groups in these criteria; the main difference was observed in Doppler mode. Vascularization was more frequent in malignant lesions, being the most relevant ultrasonographic parameter for differentiating malignant and benign lesions.

*Conclusions: UHF ultrasound proved useful in assessing skin tumors, supporting diagnosis, staging, assessing therapeutic response, and surveillance. However, it's still not possible to differentiate the main skin tumors based only on sonographic criteria. Therefore, the correlation between dermoscopy and histology is necessary for a definitive diagnosis.

*Clinical Relevance/Application: Ultra-high-frequency ultrasound, particularly at 24 MHz or higher, is a valuable tool in the assessment of skin tumors. It aids in the diagnosis process, helps in cancer staging, response evaluation to therapy, and assists in follow-up after treatment.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=R5B-SPMS-2

PREDICTIVE VALUE OF MULTI-PARAMETER ULTRASOUND COMBINED WITH PHYSIOLOGICALINDICATORSFOR HIGH-RISK HEPATITIS IN PATIENTS WITH CHRONIC LIVER DISEASE | R5B-SPGI-12


Xiaohui Qiao, Presenter | Xueqi Li, Co-author | Liyun Xue, PhD, Co-author | Guangwen Cheng, Co-author | Hong Ding, MD, Co-author

*Purpose: To investigate the predictive value of shear wave elastography (SWE), shear wave dispersion (SWD), attenuation imaging (ATI) and blood indexes for high-risk hepatitis in patients with chronic liver disease.

*Methods and Materials: The patients with metabolic dysfunction-associated steatotic liver disease (MASLD) and other chronic liver diseases such as viral, autoimmune and drug-induced liver diseases were enrolled in this prospectively study. All patients underwent liver biopsy and were divided into training set and validation set by the proportion of 8:2 according to the enrollment time. Before liver biopsy, SWE, SWD and ATI were carried out and the obtained measurement was respectively liver stiffness (LS), dispersion slope (DS) and attenuation coefficient (AC). Meanwhile, the baseline data and serological indicators were recorded. High-risk hepatitis was defined as significant inflammation and fibrosis based on pathological results. The logistic regression analysis was used to screen the risk factors and construct predictive model. The receiver operating characteristic (ROC) curve and calibration curve were applied to evaluate the diagnostic efficacy and stability of the model.

*Results: A total of 247 patients with chronic liver disease were included, 197 in the training set and 50 in the validation set, with no significant difference of the baseline characteristics between the two groups. The Logistic regression analysis revealed that LS, DS, ALT and fasting blood glucose (FBG) were the risk factors of high-risk hepatitis (P < 0.05), and a prediction model was developed based on these indicators. In training set, the area under curve (AUC) of LS, DS and the predictive model for identifying high-risk hepatitis was respectively 0.888, 0.887 and 0.926, and in validation set it was 0.743, 0.732 and 0.835, respectively. Calibration curve analysis exhibited good consistency of the predicted and actual high-risk hepatitis in both training and validation sets.

*Conclusions: The prediction model based on SWD, SWD, ALT and FBG can effectively identify high-risk hepatitis in patients with chronic liver disease, providing important reference basis for timely clinical intervention.

*Clinical Relevance/Application: Early identification of high-risk hepatitis in patients with chronic liver disease can reduce the incidence of liver-related events. The invasiveness of liver biopsy makes it unsuitable for universal screening of high-risk hepatitis population. The prediction model can assist doctors in risk stratification of patients with chronic liver disease to reduce unnecessary invasive biopsy.

View Digital Presentation: https://rsna.apprisor.org/portal.cfm?p=R5B-SPGI-12