Atsushi Higaki¹, Akira Yamamoto¹, Takeshi Fukunaga¹, Hirotake Nishimura², Akihiko Kanki¹, and Tsutomu Tamada^1
1Radiology, Kawasaki Medical School, Kurashiki, Japan, ²Pathology, Kawasaki Medical School, Kurashiki, Japan

We used contrast-enhanced MRI and CT to evaluate the diagnostic potential of histological fibrous capsule and pseudocapsule in HCC. 51.4% (36/70) of HCCs had capsules, and pseudocapsule was found in 35.3% (18/51) of lesions with enhancing capsule imaging. Among the various imaging features, T1WI low signal rim, arterial phase low signal capsule, enhancing capsule on MRI and CT, and tumor margin enlargement sign on HBP were useful in diagnosing fibrous capsules. Among these, tumor margin enlargement sign on HBP showed the highest accuracy for the diagnosis of capsule, and was the only finding that could diagnose pseudocapsule.

Elisabeth Sarah Pickles^1,2, John Connell², Stella Nay Kyi Kin², Alison Telford², Arina Kazimianec², Michael Brady², Daniel Bulte¹, Oi Fong Chong³, Han Jun Choo⁴, Sulaiha Binte Ithnin⁵, Jason Pik Eu Chang⁶, Guan Huei Lee⁷, Marianne Anastasia De Roza⁸, Kok Kiong Ong⁹, Sabrina Yi-Mei Wee¹⁰, Kee Tung Tan¹¹, Ngiap Chuan Tan¹², Han Chong Toh¹³, Yu Jun Wong^14,15, Wei Lyn Yang¹⁶, Xin Yi Yeap¹⁷, Cheryl Min En Chua¹⁸, Jacelyn Siou Sze Chua¹⁸, Jade Shu Qi Goh¹⁸, Lynette Soh Han Lai¹⁸, Yu Ki Sim¹⁸, and Pierce K.H. Chow^19,20
1Institute of Biomedical Engineering, University of Oxford, Oxford, United Kingdom, ²Perspectum Ltd, Oxford, United Kingdom, ³SingHealth Polyclinic - Bedok, Singapore, Singapore, ⁴SingHealth Polyclinic - Sengkang, Singapore, Singapore, ⁵SingHealth Polyclinic - Tampines, Singapore, Singapore, ⁶Department of Gastroenterology and Hepatology, Singapore General Hospital, Singapore, Singapore, ⁷Division of Gastroenterology and Hepatology, National University Hospital, Singapore, Singapore, ⁸Department of Gastroenterology and Hepatology, Sengkang General Hospital, Singapore, Singapore, ⁹SingHealth Polyclinic - Outram, Singapore, Singapore, ¹⁰SingHealth Polyclinic - Bukit Merah, Singapore, Singapore, ¹¹SingHealth Polyclinic - Marine Parade, Singapore, Singapore, ¹²SingHealth Polyclinic - Pasir Ris, Singapore, Singapore, ¹³Division of Medical Oncology, National Cancer Centre Singapore, Singapore, Singapore, ¹⁴Department of Gastroenterology & Hepatology, Changi General Hospital, Singapore, Singapore, ¹⁵Duke-NUS Medical School, Singapore, Singapore, ¹⁶Gastroenterology and Hepatology, Tan Tock Seng Hospital, Singapore, Singapore, ¹⁷SingHealth Polyclinic - Punggol, Singapore, Singapore, ¹⁸Division of Surgery and Surgical Oncology, National Cancer Centre Singapore, Singapore, Singapore, ¹⁹Department of Hepato-pancreato-biliary and Transplant Surgery, Singapore General Hospital and National Cancer Centre Singapore, Singapore, Singapore, ²⁰Surgery Academic Clinical Programme, Duke-NUS Medical School, Singapore, Singapore

T2*, measured using 3D breath-hold chemical shift encoded multi-gradient echo can be used to estimate liver iron concentration. The breath-hold duration (typically 10-20 s) with traditional parallel imaging acceleration techniques such as Sensitivity Encoding (SENSE) can be difficult for patients. Compressed SENSE (C-SENSE) can be used to decrease acquisition time. We compare T2* obtained with SENSE and C-SENSE in 21 patients. We found that C-SENSE acquisition was faster (9 s vs 18 s), with strong correlation and agreement with SENSE. Our findings suggest C-SENSE may potentially improve patient tolerance and turnover speed for T2* acquisitions when compared to SENSE.

Tang Cui¹, Jinming Xu², Moubin Lin³, Shixiong Qiu⁴, Xiaoming Zuo⁴, Mengxiao Liu⁵, and Peijun Wang^6
1Yangpu Hospital, School of Medicine, Tongji University,, shang hai, China, ²Department of Radiology, Yangpu Hospital, Tongji University School of Medicine, Shanghai, shang hai, China, ³Department of General Surgery, Yangpu Hospital, Tongji University School of Medicine, Shanghai, Shanghai, China, ⁴Department of Radiology, Yangpu Hospital, Tongji University School of Medicine, Shanghai, Shanghai, China, ⁵MR Scientific Marketing, Diagnostic Imaging, Siemens Healthcare Ltd., Shanghai, Shanghai, China, ⁶Department of Radiology, Tongji Hospital, Tongji University School of Medicine, Shanghai, Shanghai, China

The results of the present study suggest that there was a significant difference in the occurrence of distant metastasis between RC patients with or without mrEMVI, CRM invasion, regional LN metastasis, and MR peritoneal reflex involvement. And mrEMVI and regional LN metastasis were independent risk factors for distant metastasis of rectal cancer. Therefore, MRI can be used to observe the status of EMVI before surgery, which is helpful for the formulation of individualized treatment plans for rectal cancer and assessment of prognosis.

Chen Qiao¹, Ailian Liu¹, Ying Zhao¹, Qingwei Song¹, Xin Li ¹, Yan Guo², and Tingfan Wu^2
1the First Affiliated Hospital of Dalian Medical University, DaLian, China, ²Philips Healthcare, BeiJing, China

This work aimed at texture analysis based on mean apparent diffusion coefficient (MD) of diffusion kurtosis imaging (DKI) to evaluate the treatment efficacy of transcatheter arterial chemoembolization (TACE) in hepatocellular carcinoma (HCC). The results showed that MD texture features can differentiate objective response and non-response groups. When using the single parameter, the GreyLevelNonuniformity_angle135_offset1 achieved the best result (AUC: 0.714, sensitivity: 68.2%, specificity: 81%). In the combined diagnosis of multiple parameters, FrequencySize,MaxIntensity, GreyLevelNonuniformity_angle0_offset1 and GreyLevelNonuniformity_angle135_offset1 were combined to obtain the best combined diagnosis（AUC：0.760，sensitivity: 81.8%, specificity: 66.7%).

Matthew Harwood¹, Parth Parikh², Alexander Clinkenbeard¹, Sailen Naidu³, Marcela Salomao⁴, and Alvin Silva^1
1Radiology, Mayo Clinic Arizona, Phoenix, AZ, United States, ²Mayo Clinic Alix School of Medicine, Arizona, Scottsdale, AZ, United States, ³Interventional Radiology, Mayo Clinic Arizona, Phoenix, AZ, United States, ⁴Pathology, Mayo Clinic Arizona, Phoenix, AZ, United States

The liver imaging and reporting data system treatment response algorithm (LIRADS TRA) for CT and MRI has been used to predict residual viable tumor after treatment with yttrium-90 (Y-90). This single-center cohort of 20 patients were reviewed using the LIRADS TRA assessing residual viable tumor following Y-90 therapy. Pathology of subsequent liver explants was the reference standard. The LIRADS TRA predicted positive treatment effect (nonviable): NPV 82.4% (95% CI 0.47-1.55), accuracy 75% (72.0-89.4). The results of this preliminary study will need further validation in a larger, similar, pathologic-confirmed cohort.

Xueqin ZHANG¹, Tao ZHANG¹, Lei XU¹, and Xiance ZHAO^2
1Affiliated Nantong Hospital 3 of Nantong University, Nantong, China, ²Philips Healthcare, Shanghai, China

We used Look-Locker sequences to acquire T1 mapping images pre and post-contrast at 20 minutes after gadoxetic acid administration. Liver-to-spleen contrast index (LSI), contrast uptake index (CUI), relative liver enhancement (RLE), hepatocyte fraction (HeF), T1 relaxation times on unenhanced phase (T1_unenh), T1 relaxation times on hepatobiliary phase (T1_HBP), reduction rates of T1 relaxation times (rrT1), and the functional liver imaging score (FLIS) of the liver were measured and calculated by two radiologists independently. Our study showed that LSI, CUI, HeF, T1_HBP, rrT1 and FLIS based on gadoxetic acid–enhanced MRI are help for the evaluation of liver function.

Farzin Mobayyyen^1,2, Behrooz Taghvainia³, Hassan Homayoun², Ali Abbasian Ardakani⁴, Anahita Fathi Kazerooni², Hanieh Mobarak Salari², Nasser Rakhshani⁵, Hamidreza Salighehrad^1,2, and Hamid Reza haghighatkhah^6
1Department of Medical Physics and Biomedical Engineering, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of), Tehran, Iran (Islamic Republic of), ²Quantitative MR Imaging and Spectroscopy Group, Research Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of), ³Shahid Beheshti University of Medical Science, Tehran, Iran (Islamic Republic of), ⁴Department of Radiology Technology, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran (Islamic Republic of), ⁵Gastrointestinal and Liver Disease Research Center, Firoozgar Hospital, Iran University of Medical Sciences, Tehran, Iran (Islamic Republic of), ⁶Department of Diagnostic Imaging, Shohada-e-Tajrish Hospital, Shahid Beheshti University of Medical Science, Tehran, Iran (Islamic Republic of)

Neither biochemical markers nor a qualitative assessment of medical images are reliable to differentiate mild from moderate stages of liver fibrosis. The main purpose of this study is to develop a machine learning model to classify mild and moderate liver fibrosis based on radiomic features extracted from MRI images (T₁-w, T₂^*-map & ADC-map). Nu-SVC classifier was employed as the classification technique, trained by the extracted data from image series of 29 patients with histopathology-confirmed mild and moderate liver fibrosis. Results demonstrate that radiomic analysis of T₂^*-map and ADC-map has high potential in classifying different stages of liver fibrosis.