Search

Background Point-of-care (POC) MRI is a bedside imaging technology with fewer than five units in clinical use in the United States and a paucity of scientific studies on clinical applications. Purpose To evaluate the clinical and operational impacts of deploying POC MRI in emergency department (ED) and intensive care unit (ICU) patient settings for bedside neuroimaging, including the turnaround time. Materials and Methods In this preliminary retrospective study, all patients in the ED and ICU at a single academic medical center who underwent noncontrast brain MRI from January 2021 to June 2021 were investigated to determine the number of patients who underwent bedside POC MRI. Turnaround time, examination limitations, relevant findings, and potential CT and fixed MRI findings were recorded for patients who underwent POC MRI. Descriptive statistics were used to describe clinical variables. The Mann-Whitney

Hepatocellular carcinoma (HCC) is the most common primary malignant tumor of the liver worldwide. Noninvasive diagnosis of HCC is possible based on imaging features, without the need for tissue diagnosis. Liver Imaging Reporting and Data System (LI-RADS) CT/MRI diagnostic algorithm allows for standardized radiological interpretation and reporting of imaging studies for patients at high risk for HCC. Diagnostic categories of LR-1 to LR-5 designate each liver observation to reflect the probability of overall malignancy, HCC, or benignity based on imaging features, where LR-5 category has 95% probability of HCC. Optimal imaging protocol and scanning technique as described by the technical recommendations for LI-RADS are essential for the depiction of features to accurately characterize liver observations. The LI-RADS MRI technical guidelines recommend the minimum required sequences of T1-weighted out-of-phase and in-phase Imaging, T2-weighted Imaging, and multiphase T1-weighted Imaging. Additional sequences, including diffusion-weighted imaging, subtraction imaging, and the hepatobiliary phase when using gadobenate dimeglumine as contrast, improve diagnostic confidence, but are not required by the guidelines. These optional sequences can help differentiate true lesions from pseudolesions, detect additional observations, identify parenchymal observations when other sequences are suboptimal, and improve observations conspicuity. This manuscript reviews the optional sequences, the advantages they offer, and discusses technical optimization of these sequences to obtain the highest image quality and to avoid common artifacts.

Irreversible electroporation (IRE) is a promising non-thermal ablation technique for the treatment of patients with hepatocellular carcinoma. Early differentiation of the IRE zone from surrounding reversibly electroporated (RE) penumbra is vital for the evaluation of treatment response. In this study, an advanced statistical learning framework was developed by evaluating standard MRI data to differentiate IRE ablation zones, and to correlate with histological tumor biomarkers.Fourteen rabbits with VX2 liver tumors were scanned following IRE ablation and forty-six features were extracted from T1w and T2w MRI. Following identification of key imaging variables through two-step feature analysis, multivariable classification and regression models were generated for differentiation of IRE ablation zones, and correlation with histological markers reflecting viable tumor cells, microvessel density, and apoptosis rate. The performance of the multivariable models was assessed by measuring accuracy, receiver operating characteristics curve analysis, and Spearman correlation coefficients.The classifiers integrating four radiomics features of T1w, T2w, and T1w+T2w MRI data distinguished IRE from RE zones with an accuracy of 97%, 80%, and 97%, respectively. Also, pixelwise classification models of T1w, T2w, and T1w+T2w MRI labeled each voxel with an accuracy of 82.8%, 66.5%, and 82.9%, respectively. Regression models obtained a strong correlation with behavior of viable tumor cells (0.62 ≤ rMRI radiomics features provide descriptive power for early differentiation of IRE and RE zones while observing strong correlations among multivariable MRI regression models and histological tumor biomarkers.

To determine if ergonomic improvements in a radiology department can decrease repetitive stress injuries (RSIs), advance ergonomics knowledge, and improve well-being.Radiologists in an academic institution were surveyed regarding physician wellness, workstations, RSIs, and ergonomics knowledge before and after interventions over 1 year. Interventions included committee formation, education, wrist pads and wireless mice, broken table and chair replacement, and cord organization. Mann-Whitney U test was used for analysis.Survey response was 40% preinterventions (59/147), and 42% (66/157) postinterventions. Preinterventions, of radiologists with RSI history, 17/40 (42%) reported the RSI caused symptoms which can lead to burnout, and 15/40 (37%) responded their RSI made them think about leaving their job. Twenty-three of 59 (39%) radiologists had an active RSI preinterventions. Postinterventions, 9/25 (36%) RSI resolved, 13/25 (52%) RSI improved, and 3/25 (12%) RSI did not improve. RSI improvements were attributed to ergonomic interventions in 19/25 (76%) and therapy in 2/25 (8%). Radiologists who thought their workstation was designed with well-being in mind increased from 9/59 (15%) to 52/64 (81%). The percentage of radiologists knowing little or nothing about ergonomics decreased from 15/59 (25%) to 5/64 (8%). After ergonomics interventions, more radiologists thought the administration cared about safety and ergonomics, equipment was distributed fairly, and radiologists had the ability to ask for equipment (p.01). Fifty-three of 64 (83%) of radiologists after interventions said improving workstation ergonomic design contributed to well-being.Ergonomic improvements in radiology can decrease RSIs, advance ergonomics knowledge, and improve well-being.

Radiologist wellness is important on an individual and group/institutional level and helps to promote a strong and healthy working environment, which can improve radiologist retention and engagement. This paper will discuss case examples of radiologist wellness improvements in a single academic institution over a 3-year period using the DMAIC (Define, Measure, Analyze, Improve, and Control) model. Leveraging this framework led to the implementation of reading room assistants, reduction in work-related injuries by improvements in ergonomics, and the formation of a faculty mentorship program.

Locoregional therapy (LRT) for hepatocellular carcinoma can be used alone or with other treatment modalities to reduce rates of progression, improve survival, or act as a bridge to cure. As the use of LRT expands, so too has the need for systems to evaluate treatment response, such as the World Health Organization and modified Response Evaluation Criteria In Solid Tumors systems and more recently, the Liver Imaging Reporting and Data System (LI-RADS) treatment response algorithm (TRA). Early validation results for LI-RADS TRA have been promising, and as research accrues, the TRA is expected to evolve in the near future.

Overnight radiology coverage for trauma patients is often addressed with a combination of on-call radiology residents (RR) and a teleradiology service; however, the accuracy of these 2 readers has not been studied for trauma. We aimed to compare the accuracy of RR versus teleradiologist interpretations of CT scans for trauma patients.A retrospective analysis (March 2019 through May 2020) of trauma patients presenting to a single American College of Surgeons Level I trauma center was performed. Patients whose CT scans were performed between 10 pm to 8 am were included, because their scans were interpreted by both a RR and teleradiologist. Interpretations were compared with the final attending faculty radiologist's interpretation and graded for accuracy based on the RADPEER scoring system. Discrepancies were characterized as traumatic injury or incidental findings and missed findings or overcalls. Turnaround time was also compared.A total of 1,053 patients and 8,226 interpretations were included. Compared with teleradiologists, RR had a lower discrepancy (7.7% vs 9.0%, p = 0.026) and major discrepancy rate (3.8% vs 5.2%, p = 0.003). Among major discrepancies, RR had a lower rate of traumatic injury discrepancies (3.2% vs 4.4%, p = 0.004) and missed findings (3.4% vs 5.1%, p0.001), but a higher rate of overcalls (0.5% vs 0.1%, p0.001) compared with teleradiologists. The mean turnaround time was shorter for RR (51.3 vs 78.8 minutes, p0.001). The combination of both RR and teleradiologist interpretations had a lower overall discrepancy rate than RR (5.0% vs 7.7%, p0.001).This study identified lower discrepancy rates and a faster turnaround time by RR compared with teleradiologists for trauma CT studies. The combination of both interpreters had an even lower discrepancy rate, suggesting this combination is optimal when an in-house attending radiologist is not available.

Artificial Intelligence (AI) continues to shape the practice of radiology, with imaging of hepatocellular carcinoma (HCC) being of no exception. This article prepared by members of the LI-RADS Treatment Response (TR LI-RADS) work group and associates, presents recent trends in the utility of AI applications for the volumetric evaluation and assessment of HCC treatment response. Various topics including radiomics, prognostic imaging findings, and locoregional therapy (LRT) specific issues will be discussed in the framework of HCC treatment response classification systems with focus on the Liver Reporting and Data System treatment response algorithm (LI-RADS TRA).

There have been many publications detailing imaging features of malignant transformation of intraductal papillary mucinous neoplasms (IPMN), management and recommendations for imaging follow-up of diagnosed or presumed IPMN. However, there is no consensus on several practical aspects of imaging IPMN that could serve as a clinical guide for radiologists and enable future data mining for research. These aspects include how to measure IPMN, define reporting terminology, standardize reporting and unify guidelines for surveillance. The Society of Abdominal Radiology (SAR) created multiple Disease-Focused Panels (DFP) comprised multidisciplinary panel members who focus on a particular disease, with the goal to develop ways for radiologists to improve patient care, education, and research. DFP members met to identify the current controversies and limitations of imaging pancreatic IPMN. This paper aims to provide a practical review of the key imaging characteristics of IPMN for trainees and practicing radiologists, to guide uniformity of performance and interpretation of surveillance imaging studies, and to improve communication with clinicians by providing a lexicon and reporting template based on the experience of the SAR-DFP panel members.

Preoperative prediction of perineural invasion (PNI) and Kirsten RAS (KRAS) mutation in colon cancer is critical for treatment planning and patient management. We developed machine learning models for diagnosis of PNI and KRAS mutation in colon cancer patients by interpreting preoperative CT. This retrospective study included 207 patients who received surgical resection in our institution. The underlying tumor characteristics were described by analyzing CT image texture quantitatively. The key radiomics features were determined with similarity analysis followed by RELIEFF method among 306 CT imaging features. Eight kernel-based support vector machines classifiers were constructed using individual (II, III, or IV) or multi-stage (II + III + IV) patient cohorts for predicting PNI and KRAS mutation. The model performances were evaluated using accuracy, receiver operating curve, and decision curve analyses. Multi-stage classifiers obtained AUC of 0.793 and 0.862 for detecting PNI and KRAS mutation for test cohort. Moreover, individual-stage classifiers demonstrated significantly improved diagnostic performance at all stages (IIAUC: [0.86; 0.99], IIIAUC: [0.99; 0.99], and IVAUC: [1.00; 1.00], respectively, for PNI and KRAS mutation in test cohort). Besides, stage II tumor is better described with coarse texture features while more detailed features are required for better characterization of advanced-stage tumors (III and IV) for diagnoses of PNI or KRAS mutation. Machine learning models developed using preoperative CT data can predict PNI and KRAS mutation in colon cancer patients with satisfactory performance. Individual-stage models better-characterized the relationship between CT features and PNI or KRAS mutation than multi-stage models and demonstrated good prediction scores.

Background Concerns about potential risks of using contrast media in patients with chronic renal insufficiency limit the utilization of CT angiography in this population. Purpose To evaluate the feasibility of abdominopelvic CTA with very low volumes of contrast media. Material and methods In this retrospective study, 20 patients with chronic renal insufficiency underwent high-pitch abdominopelvic (AP) CTA on a third-generation dual-source CT scanner with 30 mL of nonionic iodinated contrast. The homogeneity of intravascular attenuation at the suprarenal aorta, infrarenal aorta, and the right common iliac artery was measured. Image noise, contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR) were used to assess objective image quality. Subjective image quality was evaluated on a 5-point scale (1 = unacceptable; 5 = excellent). Results Twelve male and eight female patients underwent CTA of the abdomen and pelvis at 80 kVp. Five CTAs also included the chest (CAP). The mean scan duration was 0.78 ± 0.19 s for AP and 0.96 ± 0.06 s for CAP CTAs. The mean ± SD of attenuation at suprarenal aorta, infrarenal aorta, and right common iliac artery were 235.1 ± 68.0, 249.2 ± 61.3, and 254.4 ± 67.7 HU, respectively. The attenuation was homogeneous across vascular levels (P = 0.06). All scans had diagnostic subjective image quality with the median (IQR) of 3.5 (1.75). CNR and SNR were homogeneous across vascular levels (P = 0.08 and P = 0.14, respectively). Conclusion Sub-second, high-pitch abdominopelvic CTA with a low volume of contrast in patients with chronic renal insufficiency is technically and clinically feasible with good diagnostic image quality and homogenous attenuation across vascular levels.

Gastrointestinal bleeding is a common cause for hospital admissions and is an important cause of morbidity and mortality. Although endoscopy is accepted as the standard initial diagnostic modality for the evaluation of gastrointestinal bleeding, multiphasic computed tomography (CT) imaging has become an alternative diagnostic tool. Dual-energy CT with post-processing techniques may have additional advantages over single-energy computed tomography in evaluation of gastrointestinal bleeding. In this article, we discuss the role of dual-energy CT in the evaluation of gastrointestinal bleeding with potential advantages over conventional CT and limitations.

Intraluminal pathologic conditions of the bile ducts and gallbladder are common, most frequently consisting of calculi and adenocarcinoma. In recent years, intraductal papillary neoplasm of the bile ducts (IPN-B), which is analogous to intraductal papillary mucinous neoplasm of the pancreas, has been recognized as a distinct pathologic entity and a precursor lesion to adenocarcinoma of the bile ducts. Intraductal tubulopapillary neoplasm (ITPN) of the bile duct was subsequently described as a distinct pathologic entity. With increased awareness and advances in imaging techniques, these lesions are diagnosed with increased frequency at preoperative imaging. A similar neoplasm in the gallbladder is referred to as intracholecystic papillary neoplasm. These lesions are often diagnosed at a preinvasive stage and have a better prognosis than invasive cholangiocarcinoma when treated with curative resection, underscoring the importance of an accurate imaging diagnosis. The most common causes of polypoid lesions of the gallbladder are cholesterol polyps and adenomyomatosis. These lesions need to be differentiated from the less common but clinically important adenocarcinoma of the gallbladder. Imaging is crucial to identify polyps that are at high risk for malignancy so that the appropriate management choice between imaging follow-up and cholecystectomy can be made by the treating physicians. Other less common gallbladder tumors, such as gallbladder adenomas, lymphoma, and metastases to the gallbladder, can manifest as intraluminal tumors; and awareness of these lesions is also important. In this article, the recent literature is reviewed; and the imaging appearances, histopathologic findings, and management of uncommon intraluminal tumors of the bile ducts and gallbladder and their mimics are discussed. ©RSNA, 2019.

The LSL-KrasG12D/+;LSL-Trp53R172H/+;Pdx-1-Cre (KPC) mouse is one of the most widely used transgenic models to evaluate tumor characteristics and to develop novel therapies for pancreatic ductal adenocarcinoma (PDAC). There is no report of the effective systemic evaluation of longitudinal KPC tumor imitation and growth. Therefore, we aimed to characterize the initiation and progression of pancreatic cancer in KPC mice using longitudinal multiparametric magnetic resonance imaging (MRI) approaches and overall survival. Ten KPC mice were used to develop spontaneous PDAC and monitored by MRI. Tumor growth was evaluated using weekly acquired MRI data. The relationship between diffusion-weighted MRI (DW-MRI) imaging biomarkers (apparent diffusion coefficient - ADC) and tumor fibrosis measurement by pathological methods was assessed by Pearson correlation coefficient. Six KPC mice developed spontaneously pancreatic tumors at the age of 20.0 ± 2.9 weeks with a relatively short life span (6.8 ± 1.8 weeks). The tumors could be detected by MRI with a minimum diameter of 3.88 ± 1.18 mm (range, 2.18–5.20 mm), showing a rapid growth curve according to both the longest diameter (1.63 ± 0.52 mm/week) and tumor volume (148.77 ± 80.87 mm3)/week. Pathological results confirmed that the tumors display histopathological features of human pancreatic cancer. A strong correlation between ADC values and fibrosis measurements was observed (R = −0.825, P = .043). Our results show that the initiation and progression of pancreatic tumor in KPC mice can be evaluated by longitudinally non-invasive dynamic MRI approaches. The findings will be the fundamental KPC background data for developing novel therapeutic approaches, in particular for evaluation of response to novel treatments.

PurposeTo evaluate the effect of the anatomic size on 3D radiomic imaging features of the breast cancer hepatic metastases.Materials and methodsCT scans of 81 liver metastases from 54 patients with breast cancer were evaluated. Ten most common 3D radiomic features from the histogram and gray level co-occurrence matrix (GLCM) categories were calculated for the hepatic metastases (HM) and compared to normal liver (NL). The effect of size was evaluated by using linear mixed-effects regression models. The effect of size on different radiomic features was analyzed for both liver lesions and background liver.ResultsThree-dimensional radiomic features from GLCM demonstrate an important size dependence. The texture-feature size dependence was found to be different among feature categories and between the HM and NL, thus demonstrating a discriminatory power for the tissue type. Significant difference in the slope was found for GLCM homogeneity (NL slope = 0.004, slope difference 95% confidence interval [CI] 0.06-0.1, p

To assess current practice patterns with respect to protocols used for incidental pancreatic cyst follow-up, management guidelines, and template reporting.The Society of Abdominal Radiology Disease Focused Panel on intraductal pancreatic neoplasms distributed an anonymous 14-question survey to its members in June 2018 that focused on current utilization of incidental pancreatic cyst guidelines, protocols, and template reporting.Among the 1,390 email invitations, 323 responded, and 94.7% (306 of 323) completed all questions. Respondents were mainly radiologists (93.8%, 303 of 323) from academic institutions (74.7%, 227 of 304) in North America (93.7%, 286 of 305). Of respondents, 42.5% (136 of 320) preferred 2017 ACR recommendations, 17.8% (57 of 320) homegrown systems, 15.0% (48 of 320) Fukuoka guidelines, and 7.8% (25 of 320) American Gastroenterological Association guidelines. The majority (68.7%, 222 of 323) agreed or strongly agreed that developing a single international consensus recommendation for management was important, and most radiologists preferred to include them in reports (231 of 322, 71.7%); yet only half included recommendations in75% of reports (161 of 321). MR cholangiopancreatography was the modality of choice for follow-up of2.5 cm cysts. Intravenous contrast was routinely used by 69.7% (212 of 304). Standardized reporting templates were rarely used in practice (12.8% 39 of 306).Nearly 7 of 10 radiologists desire a unified international consensus recommendation for management of incidental cystic pancreatic lesions; ACR 2017 recommendations are most commonly used, followed by homegrown systems and Fukuoka guidelines. The majority of radiologists routinely use MR cholangiopancreatography with intravenous contrast for follow-up of incidental cystic lesions, but template reporting is rarely used.

Artificial Intelligence (AI) continues to shape the practice of radiology, with imaging of hepatocellular carcinoma (HCC) being of no exception. This article prepared by members of the LI-RADS Treatment Response (TR LI-RADS) work group and associates, presents recent trends in the utility of AI applications for the volumetric evaluation and assessment of HCC treatment response. Various topics including radiomics, prognostic imaging findings, and locoregional therapy (LRT) specific issues will be discussed in the framework of HCC treatment response classification systems with focus on the Liver Reporting and Data System treatment response algorithm (LI-RADS TRA).

Anna J Shangguan,1,* Kang Zhou,2,* Jia Yang,1 Aydin Eresen,1 Bin Wang,3 Chong Sun,4 Liang Pan,5 Su Hu,6 Ali T Khan,1 Samdeep K Mouli,1 Vahid Yaghmai,7 Zhuoli Zhang1 1Department of Radiology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA; 2Department of Radiology, Peking Union Medical College Hospital, Peking Union Medical College, Beijing, People’s Republic of China; 3Department of General Surgery, Nanfang Hospital, Southern Medical University, Guangdong Provincial Engineering Technology Research Center of Minimally Invasive Surgery, Guangzhou, People’s Republic of China; 4Department of Orthopedics, Affiliated Hospital of Qingdao University, Qingdao, Shandong, People’s Republic of China; 5Department of Radiology, Third Affiliated Hospital of Suzhou University, Changzhou, Jiangsu, People’s Republic of China; 6Department of Radiology, First Affiliated Hospital of Soochow University, Suzhou, Jiangsu, People’s Republic of China; 7Department of Radiological Sciences, School of Medicine, University of California Irvine, Irvine, CA, USA*These authors contributed equally to this workCorrespondence: Zhuoli ZhangDepartment of Radiology, Northwestern University, 737 N Michigan Ave, 16 th Floor, Chicago, IL 60611, USATel +1 312 695 5753Fax +1 312 926 5991Email zhuoli-zhang@northwestern.eduPurpose: Irreversible electroporation (IRE) is a promising new ablation method for hepatocellular carcinoma (HCC) treatment with few side-effects; however, tissue perfusion and differentiation between treatment zones have not been sufficiently studied. In this project, we analyzed HCC tumor perfusion changes immediately after IRE treatment using transcatheter intraarterial perfusion (TRIP)-MRI to monitor treatment zone margins.Materials and Methods: All protocols were approved by the institutional animal care and use committee. A total of 34 rabbits were used for this prospective study: tumor liver group (n=17), normal liver group (n=14), and 3 for growing VX2 tumors. All procedures and imaging were performed under anesthesia. VX2 tumors were grown by injection of VX2 cells into rabbit hindlimbs. Liver tumors were induced by percutaneous US-guided injection of VX2 tumor fragments into liver. For digital subtraction angiography (DSA), a 2F catheter was advanced through left hepatic artery via femoral artery access, followed by contrast injection. All rabbits underwent baseline anatomic MRI, then IRE procedure or IRE probe placement only, and lastly post-procedure anatomic and TRIP-MRI. Liver tissues were dissected immediately after imaging for histology. All statistical analyses were performed on GraphPad Prism, with P< 0.05 considered significant.Results: IRE generated central IRE zone and peripheral reversible electroporation (RE) zone on anatomic MRI for both normal liver and liver tumor tissues. The semiquantitative analysis showed that IRE zone had the lowest AUC, PE, WIS, Ktrans, ve, and vp as well as the highest TTP, followed by RE zone, then untreated tissues. Receiver operating characteristic analysis showed that WIS and AUC60 had the highest AUCROC. Histologic analysis showed a positive correlation in viable area fraction between MRI and histologic measurements. IRE zone had the highest %apoptosis and lowest CD31+ staining.Conclusion: Our results demonstrated that intraprocedural TRIP-MRI can effectively differentiate IRE and RE zones after IRE ablation in normal liver and liver tumor tissues.Keywords: hepatocellular carcinoma, irreversible electroporation, liver cancer, perfusion, transcatheter intraarterial perfusion magnetic resonance imaging

It is unknown whether the route of administration impacts dendritic cell (DC)-based immunotherapy for pancreatic ductal adenocarcinoma (PDAC). We compared the effect of intraperitoneal (i.p.), subcutaneous (s.c.), and intratumoral (i.t.) administration of DC vaccine on induction of antitumor responses in a KPC mouse model of PDAC. Histological analysis and flow cytometry were used to evaluate tumor progression and antitumor immunity after different routes of DC vaccination. Using a flank mouse model of PDAC, we found that the i.t. route of DC vaccination had no significant effect on tumor growth rates compared with i.p. and s.c. routes (i.p. 6.66 ± 2.58% vs s.c. 6.79 ± 1.36% vs i.t. 8.57 ± 2.36%; P = 0.33). However, in an orthotopic PDAC model, i.p. injection of DC vaccine effectively suppressed tumor growth, inhibited tumor progression, and increased antitumor immunity compared with s.c. vaccination (tumor weight: i.p. 71.60 ± 15.55 mg vs control 200.40 ± 53.04 mg; P = 0.048; s.c. 151.40 ± 41.64 mg vs control 200.40 ± 53.04 mg; P = 0.49). Our study suggests that immunization via an i.p. route results in superior antitumor immune response and tumor suppression when compared with other routes.

There have been many publications detailing imaging features of malignant transformation of intraductal papillary mucinous neoplasms (IPMN), management and recommendations for imaging follow-up of diagnosed or presumed IPMN. However, there is no consensus on several practical aspects of imaging IPMN that could serve as a clinical guide for radiologists and enable future data mining for research. These aspects include how to measure IPMN, define reporting terminology, standardize reporting and unify guidelines for surveillance. The Society of Abdominal Radiology (SAR) created multiple Disease-Focused Panels (DFP) comprised multidisciplinary panel members who focus on a particular disease, with the goal to develop ways for radiologists to improve patient care, education, and research. DFP members met to identify the current controversies and limitations of imaging pancreatic IPMN. This paper aims to provide a practical review of the key imaging characteristics of IPMN for trainees and practicing radiologists, to guide uniformity of performance and interpretation of surveillance imaging studies, and to improve communication with clinicians by providing a lexicon and reporting template based on the experience of the SAR-DFP panel members.

Rationale and objectivesTo develop classification and regression models interpreting tumor characteristics obtained from structural (T1w and T2w) magnetic resonance imaging (MRI) data for early detection of dendritic cell (DC) vaccine treatment effects and prediction of long-term outcomes for LSL-KrasG12D; LSL-Trp53R172H; Pdx-1-Cre (KPC) transgenic mice model of pancreatic ductal adenocarcinoma.Materials and methodsEight mice were treated with DC vaccine for 3 weeks while eight KPC mice were used as untreated control subjects. The reproducibility of the computed 264 features was evaluated using the intraclass correlation coefficient. Key variables were determined using a three-step feature selection approach. Support vector machines classifiers were generated to differentiate treatment-related changes on tumor tissue following first- and third weeks of the DC vaccine therapy. The multivariable regression models were generated to predict overall survival (OS) and histological tumor markers of KPC mice using quantitative features.ResultsThe quantitative features computed from T1w MRI data have better reproducibility than T2w MRI features. The KPC mice in treatment and control groups were differentiated with a longitudinally increasing accuracy (first- and third weeks: 87.5% and 93.75%). The linear regression model generated with five features of T1w MRI data predicted OS with a root-mean-squared error (RMSE)

Preoperative prediction of perineural invasion (PNI) and Kirsten RAS (KRAS) mutation in colon cancer is critical for treatment planning and patient management. We developed machine learning models for diagnosis of PNI and KRAS mutation in colon cancer patients by interpreting preoperative CT.This retrospective study included 207 patients who received surgical resection in our institution. The underlying tumor characteristics were described by analyzing CT image texture quantitatively. The key radiomics features were determined with similarity analysis followed by RELIEFF method among 306 CT imaging features. Eight kernel-based support vector machines classifiers were constructed using individual (II, III, or IV) or multi-stage (II + III + IV) patient cohorts for predicting PNI and KRAS mutation. The model performances were evaluated using accuracy, receiver operating curve, and decision curve analyses.Multi-stage classifiers obtained AUC of 0.793 and 0.862 for detecting PNI and KRAS mutation for test cohort. Moreover, individual-stage classifiers demonstrated significantly improved diagnostic performance at all stages (IIMachine learning models developed using preoperative CT data can predict PNI and KRAS mutation in colon cancer patients with satisfactory performance. Individual-stage models better-characterized the relationship between CT features and PNI or KRAS mutation than multi-stage models and demonstrated good prediction scores.

Background Preoperative detection of lymph node (LN) metastasis is critical for planning treatments in colon cancer (CC). The clinical diagnostic criteria based on the size of the LNs are not sensitive to determine metastasis using CT images. In this retrospective study, we investigated the potential value of CT texture features to diagnose LN metastasis using preoperative CT data and patient characteristics by developing quantitative prediction models. Methods A total of 390 CC patients, undergone surgical resection, were enrolled in this monocentric study. 390 histologically validated LNs were collected from patients and randomly separated into training (312 patients, 155 metastatic and 157 normal LNs) and test cohorts (78 patients, 39 metastatic and 39 normal LNs). Six patient characteristics and 146 quantitative CT imaging features were analyzed and key variables were determined using either exhaustive search or least absolute shrinkage algorithm. Two kernel-based support vector machine classifiers (patient-characteristic model and radiomic-derived model), generated with 10-fold cross-validation, were compared with the clinical model that utilizes long-axis diameter for diagnosis of metastatic LN. The performance of the models was evaluated on the test cohort by computing accuracy, sensitivity, specificity, and area under the receiver operating curve (AUC). Results The clinical model had an overall diagnostic accuracy of 64.87%; specifically, accuracy of 65.38% and 62.82%, sensitivity of 83.87% and 84.62%, and specificity of 47.13% and 41.03% for training and test cohorts, respectively. The patient-demographic model obtained accuracy of 67.31% and 73.08%, the sensitivity of 62.58% and 69.23%, and specificity of 71.97% and 76.23% for training and test cohorts, respectively. Besides, the radiomic-derived model resulted in an accuracy of 81.09% and 79.49%, sensitivity of 83.87% and 74.36%, and specificity of 78.34% and 84.62% for training and test cohorts, respectively. Furthermore, the diagnostic performance of the radiomic-derived model was significantly higher than clinical and patient-demographic models (p Conclusions The texture of the LNs provided characteristic information about the histological status of the LNs. The radiomic-derived model leveraging LN texture provides better preoperative diagnostic accuracy for the detection of metastatic LNs compared to the clinically accepted diagnostic criteria and patient-demographic model.

Background There is a lack of well-established clinical tools for predicting dendritic cell (DC) vaccination response of pancreatic ductal adenocarcinoma (PDAC). DC vaccine treatment efficiency was demonstrated using histological analysis in pre-clinical studies; however, its usage was limited due to invasiveness. In this study, we aimed to investigate the potential of MRI texture features for detection of early immunotherapeutic response as well as overall survival (OS) of PDAC subjects following dendritic cell (DC) vaccine treatment in LSL-KrasG12D;LSL-Trp53R172H;Pdx-1-Cre (KPC) transgenic mouse model of pancreatic ductal adenocarcinoma (PDAC). Materials and methods KPC mice were treated with DC vaccines, and tumor growth was dynamically monitored. A total of a hundred and fifty-two image features of T2-weighted MRI images were analyzed using a kernel-based support vector machine model to detect treatment effects following the first and third weeks of the treatment. Moreover, univariate analysis was performed to describe the association between MRI texture and survival of KPC mice as well as histological tumor biomarkers. Results OS for mice in the treatment group was 54.8 ± 22.54 days while the control group had 35.39 ± 17.17 days. A subset of three MRI features distinguished treatment effects starting from the first week with increasing accuracy throughout the treatment (75% to 94%). Besides, we observed that short-run emphasis of approximate wavelet coefficients had a positive correlation with the survival of the KPC mice (r = 0.78, p r = 0.84, p r = − 0.97, p r = 0.81, p Conclusion Our results demonstrate that MRI texture features can be used as imaging biomarkers for early detection of therapeutic response following DC vaccination in the KPC mouse model of PDAC. Besides, MRI texture can be utilized to characterize tumor microenvironment reflected with histology analysis.

RATIONALE AND OBJECTIVES: To investigate whether transcatheter intraarterial perfusion (TRIP) magnetic resonance imaging (MRI) can differentiate reversible electroporation (RE) zones from irreversible electroporation (IRE) zones immediately after IRE procedure in the rabbit liver. MATERIALS AND METHODS: All studies were approved by the institutional animal care and use committee and performed in accordance with institutional guidelines. A total of thirteen healthy New Zealand White rabbits were used. After selective catheterization of the hepatic artery under X-ray fluoroscopy, we acquired TRIP-MRI at 20 minutes post-IRE using 3 mL of 5% intraarterial gadopentetate dimeglumine. Semi-quantitative (peak enhancement, PE; time to peak, TTP; wash-in slope, WIS; areas under the time-intensity curve, AUT, over 30, 60, 90, 120, 150, and 180 seconds after the initiation of enhancement) and quantitative (K(trans), v(e), and v(p)) TRIP-MRI parameters were calculated. The relationships between TRIP-MRI parameters and histological measurements and the differential ability of TRIP-MRI parameters was assessed. RESULTS: PE, AUT(60), AUT(90), AUT(120), AUT(150), AUT(180), K(trans), and v(e) were significantly higher in RE zones than in IRE zones (all P < 0.05), and AUC for these parameters ranged from 0.91(95% CI, 0.80, 1.00) to 0.99 (95% CI, 0.98, 1.00). There was no significant difference in AUC between any two parameters (Z, 0–1.47; P, 0.14–1.00). Hepatocyte apoptosis strongly correlated with PE, AUT(60), AUT(90), AUT(120), AUT(150), AUT(180), K(trans), and v(p) (the absolute value r, 0.6–0.7, all P < 0.0001). CONCLUSION: AUT(150) or AUT(180) could be a potential imaging biomarker to differentiate RE from IRE zones, and TRIP-MRI permits to differentiate RE from IRE zones immediately after IRE procedure in the rabbit liver.

RATIONALE AND OBJECTIVES This study sought to more definitely illustrate the impact and feasibility of implementing a low-dose protocol for computed tomography (CT)-guided biopsies using size-specific dose estimates and multivariate analyses. MATERIALS AND METHODS Fifty consecutive CT-guided lung and extrapulmonary biopsies were reviewed before and after implementation of a low-dose protocol (200 patients total, mean age 61 ± 15 years, 128 women). Analyses of variance with Bonferroni correction were used to compare standard and low-dose protocols in terms of patient demographics, physician experience, target lesion size, total dose-length product, total acquisitions, size-specific dose estimate, signal-to-noise ratio, contrast-to-noise ratio, and lesion conspicuity ratings. All procedures were performed on the same 16-slice CT scanner. RESULTS Voluntary protocol adherence was 100% (lung) and 89% (extrapulmonary). The low-dose protocol achieved significantly lower total average dose-length product [(lung) 735.6 ± 599.4 mGy × cm to 252.1 ± 101.9 mGy × cm, P

The objective or our study was to assess the incidence rate and clinical characteristics of allergiclike reactions in patients who received both nonionic iodinated contrast medium (ICM) and gadolinium-based contrast medium (GBCM).Acute allergiclike reactions in patients who received both ICM and GBCM to nonionic ICM or GBCM injections during a 5-year period were analyzed. Allergy preparation was not administered when patients received a different type of contrast material. Acute allergiclike reactions to both ICM and GBCM were evaluated.Of 302,858 contrast injections (155,234 ICM and 147,624 GBCM) during a 5-year period, 1006 (752 ICM and 254 GBCM) acute allergiclike contrast reactions were reported. The overall rate of reaction to ICM was 0.48% (95% CI, 0.45-0.52%), and the overall rate of reaction to GBCM was 0.17% (95% CI, 0.15-0.19%). A total of 19,237 patients received at least one ICM injection and one GBCM injection, with a total of 56,310 injections (19,237 initial injections and 37,073 subsequent injections). Nine patients had reactions to both ICM and GBCM with the primary reaction rate of 9/19,237 (incidence rate, 0.047%; 95% CI, 0.044-0.050%), and the secondary reaction rate of 9/37,073 (incidence rate, 0.024%; 95% CI, 0.023-0.026%). All secondary reactions in patients who had a reaction to both ICM and GBCM were mild. None of the patients required medication for the treatment of the secondary reaction.An allergiclike reaction to both nonionic ICM and GBCM was an extremely rare event that presented as a mild acute reaction without significant clinical consequences despite the fact that an allergy preparation was not administered.

While natural killer (NK) cell‐based adoptive transfer immunotherapy (ATI) provides only modest clinical success in cancer patients. This study was hypothesized that MRI‐guided transcatheter intra‐hepatic arterial (IHA) infusion permits local delivery to liver tumors to improve outcomes during NK‐based ATI in a rat model of hepatocellular carcinoma (HCC). Mouse NK cells were labeled with clinically applicable iron nanocomplexes. Twenty rat HCC models were assigned to three groups: transcatheter IHA saline infusion as the control group, transcatheter IHA NK infusion group, and intravenous (IV) NK infusion group. MRI studies were performed at baseline and at 24 h, 48 h, and 8 days postinfusion. There was a significant difference in tumor R2* values between baseline and 24 h following the selective transcatheter IHA NK delivery to the tumors (P = 0.039) when compared to IV NK infusion (P = 0.803). At 8 days postinfusion, there were significant differences in tumor volumes between the control, IV, and IHA NK infusion groups (control vs. IV, P = 0.196; control vs. IHA, P

Does imaging response predict survival in hepatocellular carcinoma (HCC)? We studied the ability of posttherapeutic imaging response to predict overall survival. Over 14 years, 948 patients with HCC were treated with radioembolization. Patients with baseline metastases, vascular invasion, multifocal disease, Child-Pugh B7, and transplanted/resected were excluded. This created our homogeneous study cohort of 134 patients with Child-Pugh ≤ B7 and solitary HCC. Response (using European Association for Study of the Liver [EASL] and Response Evaluation Criteria in Solid Tumors 1.1 [RECIST 1.1] criteria) was associated with survival using Landmark and risk-of-death methodologies after reviewing 960 scans. In a subanalysis, survival times of responders were compared to those of patients with stable disease (SD) and progressive disease (PD). Uni/multivariate survival analyses were performed at each Landmark. At the 3-month Landmark, responders survived longer than nonresponders by EASL (hazard ratio [HR], 0.46; confidence interval [CI], 0.26-0.82; P = 0.002) but not RECIST 1.1 criteria (HR, 0.70; CI, 0.37-1.32; P = 0.32). At the 6-month Landmark, responders survived longer than nonresponders by EASL (HR, 0.32; CI, 0.15-0.77; P 0.001) and RECIST 1.1 criteria (HR, 0.50; CI, 0.29-0.87; P = 0.021). At the 12-month Landmark, responders survived longer than nonresponders by EASL (HR, 0.34; CI, 0.15-0.77; P 0.001) and RECIST 1.1 criteria (HR, 0.52; CI 0.27-0.98; P = 0.049). At 6 months, risk of death was lower for responders by EASL (P 0.001) and RECIST 1.1 (P = 0.0445). In subanalyses, responders lived longer than patients with SD or PD. EASL response was a significant predictor of survival at 3-, 6-, and 12-month Landmarks on uni/multivariate analyses.Response to radioembolization in patients with solitary HCC can prognosticate improved survival. EASL necrosis criteria outperformed RECIST 1.1 size criteria in predicting survival. The therapeutic objective of radioembolization should be radiologic response and not solely to prevent progression. (Hepatology 2018;67:873-883).

Hepatocellular carcinoma (HCC) is the third most frequent source of deaths associated with cancer after lung cancer in the world despite recent innovative treatment techniques. Liver transplantation, hepatic resection, and percutaneous ablation techniques hold great promise as potentially curative treatments for patients at early stages. Nevertheless, most of the patients are not suitable for these curative treatments due to their advanced disease stages at the time of diagnosis. Food and Drug Administration (FDA) approved tyrosine kinase inhibitor, sorafenib is a standard therapy for advanced-stage HCC patients which extends overall survival for several months. However, its therapeutic efficacy is restricted by adverse events and drug resistance which limits the number of patients benefiting from this systemic chemotherapeutic drug. During the last decade, novel approaches including but not limited to immunotherapies, ablation methods, and chemotherapeutic drugs were proposed to enhance sensitivity to sorafenib, improve therapeutic efficacy, and prohibit adverse events through novel delivery routes, utilization of nanoparticle carriers, and combination with other therapeutic agents. However, studies are still being conducted to optimize the efficiency of sorafenib and reduce its adverse events. In this review paper, we examine research studies evaluating novel delivery methods to reduce drug-related cytotoxicity to improve patient tolerance to sorafenib and its therapeutic efficacy in patients with HCC. Moreover, therapeutic approaches with the synergistic potential to combine with sorafenib are briefly summarized.

Hepatocellular carcinoma (HCC) is among the most lethal cancer types despite great advancement in overall survival of the patients over the last decades. Surgical resection or partial hepatectomy has been approved as the curative treatment for early-stage HCC patients however only up to 30% of them are eligible for the procedures. Natural killer (NK) cells are cytotoxic lymphocytes recognized for killing virally infected cells and improving immune functions for defending the body against malignant cells. Although autologous NK cells failed to demonstrate significant clinical benefit, transfer of allogeneic adoptive NK cells arises as a promising approach for the treatment of solid tumors. The immunosuppressive tumor microenvironment and inadequate homing efficiency of NK cells to tumors can inhibit adoptive transfer immunotherapy (ATI) efficacy. However, potential of the NK cells is challenged by the transfection efficiency. The local ablation techniques that employ thermal or chemical energy have been investigated for the destruction of solid tumors for three decades and demonstrated promising benefits for individuals not eligible for surgical resection or partial hepatectomy. Irreversible electroporation (IRE) is one of the most recent minimally invasive ablation methods that destruct the cell within the targeted region through non-thermal energy. IRE destroys the tumor cell membrane by delivering high-frequency electrical energy in short pulses and overcomes tumor immunosuppression. The previous studies demonstrated that IRE can induce immune changes which can facilitate activation of specific immune responses and improve transfection efficiency. In this review paper, we have discussed the mechanism of NK cell immunotherapy and IRE ablation methods for the treatment of HCC patients and the combinatorial benefits of NK cell immunotherapy and IRE ablation.

Transarterial radioembolization is a novel therapy that has gained rapid clinical acceptance for the treatment of hepatocellular carcinoma (HCC). Segmental radioembolization [also termed radiation segmentectomy (RS)] is a technique that can deliver high doses (> 190 Gy) of radiation selectively to the hepatic segment(s) containing the tumor. The aim of this comprehensive review is to provide an illustrative summary of the most relevant imaging findings encountered after radiation segmentectomy. A 62-patient cohort of Child-Pugh A patients with solitary HCC

Radiologists play a central role in the assessment of patient response to locoregional therapies for hepatocellular carcinoma (HCC). The identification of viable tumor following treatment guides further management and potentially affects transplantation eligibility. Liver Imaging Reporting and Data Systems (LI-RADS) first introduced the concept of LR-Treated in 2014, and a new Treatment Response Algorithm is included in the 2017 update to assist radiologists in image interpretation of HCC after locoregional therapy. In addition to offering imaging criteria for viable and nonviable HCC, new concepts of nonevaluable tumors as well as tumors with equivocal viability are introduced. Existing guidelines provided by Response Evaluation Criteria In Solid Tumors (RECIST) and modified RECIST address patient-level assessments and are routinely used in clinical trials but do not address the variable appearances following different locoregional therapies. The new LI-RADS Treatment Response Algorithm addresses this gap and offers a comprehensive approach to assess treatment response for individual lesions after a variety of locoregional therapies, using either contrast-enhanced CT or MRI.

There is a lack of a well-established approach for assessment of early treatment outcomes for modern therapies for pancreatic ductal adenocarcinoma (PDAC) e.g. dinaciclib or dendritic cell (DC) vaccination. Here, we developed multivariate models using MRI texture features to detect treatment effects following dinaciclib drug or DC vaccine therapy in a transgenic mouse model of PDAC including 21 LSL-Kras(G12D); LSL-Trp53(R172H); Pdx-1-Cre (KPC) mice used as untreated control subjects (n=8) or treated with dinaciclib (n=7) or DC vaccine (n=6). Support vector machines (SVM) technique was performed to build a linear classifier with three variables for detection of tumor tissue changes following drug or vaccine treatments. Besides, multivariate regression models were generated with five variables to predict survival behavior and histopathological tumor markers (Fibrosis, CK19, and Ki67). The diagnostic performance was evaluated using accuracy, area under the receiver operating characteristic curve (AUC) and decision curve analyses. The regression models were evaluated with adjusted r-squared (R(adj) (2)). SVM classifier successfully distinguished changes in tumor tissue with an accuracy of 95.24% and AUC of 0.93. The multivariate models generated with five variables were strongly associated with histopathological tumor markers, fibrosis (R(adj) (2)=0.82, P

OBJECTIVE. The purpose of this study is to retrospectively assess the impact of iterative metal artifact reduction (IMAR) with iterative reconstruction (IR) on the image quality and diagnostic performance of CT urography in the evaluation of patients with hip prostheses, compared with IR alone. MATERIALS AND METHODS. CT urography examinations that were reconstructed using IR with and without IMAR were analyzed for 57 patients (29 women and 28 men; mean age, 74 years [range, 22-94 years]) with hip prostheses (40 unilateral and 17 bilateral). For quantitative analysis, image noise within the bladder was measured. Two radiologists (radiologist 1 [RAD1] and radiologist 2 [RAD2]) qualitatively evaluated the images using both a 5-point scale to assess the degree of visualization of artifacts and a 6-point scale to determine diagnostic confidence in visualization of the bladder, ureters, prostate or uterus, pelvic calcifications, and genitourinary abnormalities involving the bladder, distal ureters, prostate, uterus, and ovaries. RESULTS. The combination of IMAR and an IR technique provided improvement in quantitative and qualitative measurements (p < 0.05). Forty-three genitourinary abnormalities were detected in 29 patients. Quantitative and qualitative comparisons of scans obtained with and without the use of IMAR, respectively, revealed image noise of 99.6 versus 173.3 HU and the following radiologist scores: for improvement of artifacts, 3.2 versus 1.6 (for RAD1) and 3.1 versus 1.6 (for RAD2); for visualization of the bladder, 3.6 versus 1.5 (RAD1) and 3.8 versus 1.6 (RAD2); visualization of the ureters, 3.8 versus 1.6 (RAD1) and 3.9 versus 1.7 (RAD2); visualization of the uterus, 4.3 versus 2.8 (RAD1) and 4.3 versus 2.6 (RAD2); visualization of the prostate, 4.5 versus 2.3 (RAD1) and 4.5 versus 2.2 (RAD2); diagnostic confidence for calcifications, 4.7 versus 3.5 (RAD1) and 4.7 versus 3.3 (RAD2); and diagnostic confidence for genitourinary abnormalities, 5.0 versus 3.2 (RAD1) and 4.8 versus 2.9 (RAD2), respectively. CONCLUSION. The addition of IMAR to IR led to statistically significant improvement in the retrospective diagnostic performance and image quality of CT urography for patients with hip prostheses, compared with IR alone.

The aim of this study was to develop and validate a new non-invasive artificial intelligence (AI) model based on preoperative computed tomography (CT) data to predict the presence of liver metastasis (LM) in colon cancer (CC). A total of forty-eight eligible CC patients were enrolled, including twenty-four patients with LM and twenty-four patients without LM. Six clinical factors and one hundred and fifty-two tumor image features extracted from CT data were utilized to develop three models: clinical, radiomics, and hybrid (a combination of clinical and radiomics features) using support vector machines with 5-fold cross-validation. The performance of each model was evaluated in terms of accuracy, specificity, sensitivity, and area under the curve (AUC). For the radiomics model, a total of four image features utilized to construct the model resulting in an accuracy of 83.87% for training and 79.50% for validation. The clinical model that employed two selected clinical variables had an accuracy of 69.82% and 69.50% for training and validation, respectively. The hybrid model that combined relevant image features and clinical variables improved accuracy of both training (90.63%) and validation (85.50%) sets. In terms of AUC, hybrid (0.96; 0.87) and radiomics models (0.91; 0.85) demonstrated a significant improvement compared with the clinical model (0.71; 0.69), and the hybrid model had the best prediction performance. In conclusion, the AI model developed using preoperative conventional CT data can accurately predict LM in CC patients without additional procedures. Furthermore, combining image features with clinical characteristics greatly improved the model’s prediction performance. We have thus generated a promising tool that allows guidance and individualized surveillance of CC patients with high risks of LM.

PURPOSE Irreversible electroporation (IRE) is a nonthermal tissue ablation technique that represents a promising treatment option for unresectable liver tumors, but the effectively treated zone cannot be reliably predicted. We investigate the potential benefit of transcatheter intra-arterial perfusion (TRIP) -MRI for the early noninvasive differentiation of IRE zone from surrounding reversibly electroporated (RE) zone. METHODS Seventeen rabbits with VX2 liver tumors were scanned with morphological and contrast-enhanced MRI sequences approximately 30 min after IRE tumor ablation. Quantitative TRIP-MRI perfusion parameters were evaluated in IRE zone and RE zone, defined according to histology. MRI and histology results were compared among zones using Wilcoxon rank-sum tests and correlations were evaluated by Pearson's correlation coefficient. RESULTS There were significant differences in area under the curve, time to peak, maximum and late enhancement, wash-in and wash-out rates in the tumor IRE zones compared with the boundary tumor RE zones and untreated tumors. Histology showed significantly fewer tumor cells, microvessels and significantly more apoptosis in tumor IRE zones compared with tumor RE zones (-51%, -66% and +185%, respectively) and untreated tumors (-60%, -67%, and +228%, respectively). A strong correlation was observed between MRI and histology measurements of IRE zones (r = 0.948) and RE zones (r = 0.951). CONCLUSION TRIP-MRI demonstrated the potential to detect immediate perfusion changes following IRE liver tumor ablation and effectively differentiate the IRE zone from the surrounding tumor RE zone.

To assess the feasibility of implementing fully automated computer-aided diagnosis (CAD) for detection of pulmonary nodules on CT pulmonary angiography (CTPA) studies in emergency setting. CTPA of 48 emergency patients was retrospectively reviewed. Fully automated CAD nodule detection was performed at the scanner and results were automatically submitted to PACS. A third-year radiology resident (RAD1) and a cardiothoracic radiologist with 6 years’ experience (RAD2) reviewed the scans independently to detect pulmonary nodules in two different sessions 8 weeks apart: session 1, CAD was reviewed first and then all images were reviewed; session 2, CAD was reviewed last after all images were reviewed. Time spent by RAD to evaluate image sets was measured for each case. Fisher’s exact test and t test were used. There were 17 male and 31 female patients with mean ± SD age of 48.7 ± 16.4 years. Using CAD at the beginning was associated with lower average reading time for both readers. However, difference in reading time did not reach statistical significance for RAD1 (RAD1 94.6 s vs. 102.7 s, P > 0.05; RAD2 61.1 s vs. 76.5 s, P

The purpose of this study was to develop a radiomics signature for distinguishing stage in advanced colon cancer (CC). 195 colon cancer patients were enrolled in this study (stage III, n = 146 vs. stage IV, n = 49) and divided into training cohort (n = 136) and validation cohort (n = 59). A total of 286 radiomic features were extracted from tumor and LN images. A radiomics signature was generated using the least absolute shrinkage and selection operator (LASSO) technique. The relationship between radiomics signature and CC staging was explored using a kernel-based support vector machine (SVM) classifier model. The classification performance was assessed by accuracy and the receiver operating characteristics (ROC) curve. A total of 5 features (2 for tumor and 3 for LN) were selected among 286 features. Radiomics signature built from extracted features successfully differentiated stage III from stage IV CC with no known distant metastases on imaging preoperatively. Furthermore, the SVM classifier model generated using tumor and LN images together achieved better performance than the tumor alone, with accuracies of 86.03% vs. 78.68% and 83.05% vs. 76.27% in training and validation cohorts, respectively. In ROC analysis, the model showed a significant improvement for training (AUC 89.16% vs. 69.5%) and validation cohorts (AUC 75.15% vs. 55%) in comparison with the combined analysis and the tumor alone. In conclusion, the radiomics signature based on preoperative CT may distinguish stage III from stage IV CC with no known distant metastases. In addition, the radiomic features from combined images achieved better classification performance than tumor alone.