Large Language Models for Automated Synoptic Reports and Resectability Categorization in Pancreatic Cancer.

Background Structured radiology reports for pancreatic ductal adenocarcinoma (PDAC) improve surgical decision-making over free-text reports, but radiologist adoption is variable. Resectability criteria are applied inconsistently. Purpose To evaluate the performance of large language models (LLMs) in automatically creating PDAC synoptic reports from original reports and to explore performance in categorizing tumor resectability. Materials and Methods In this institutional review board-approved retrospective study, 180 consecutive PDAC staging CT reports on patients referred to the authors' European Society for Medical Oncology-designated cancer center from January to December 2018 were included. Reports were reviewed by two radiologists to establish the reference standard for 14 key findings and National Comprehensive Cancer Network (NCCN) resectability category. GPT-3.5 and GPT-4 (accessed September 18-29, 2023) were prompted to create synoptic reports from original reports with the same 14 features, and their performance was evaluated (recall, precision, F1 score). To categorize resectability, three prompting strategies (default knowledge, in-context knowledge, chain-of-thought) were used for both LLMs. Hepatopancreaticobiliary surgeons reviewed original and artificial intelligence (AI)-generated reports to determine resectability, with accuracy and review time compared. The McNemar test, t test, Wilcoxon signed-rank test, and mixed effects logistic regression models were used where appropriate. Results GPT-4 outperformed GPT-3.5 in the creation of synoptic reports (F1 score: 0.997 vs 0.967, respectively). Compared with GPT-3.5, GPT-4 achieved equal or higher F1 scores for all 14 extracted features. GPT-4 had higher precision than GPT-3.5 for extracting superior mesenteric artery involvement (100% vs 88.8%, respectively). For categorizing resectability, GPT-4 outperformed GPT-3.5 for each prompting strategy. For GPT-4, chain-of-thought prompting was most accurate, outperforming in-context knowledge prompting (92% vs 83%, respectively; P = .002), which outperformed the default knowledge strategy (83% vs 67%, P < .001). Surgeons were more accurate in categorizing resectability using AI-generated reports than original reports (83% vs 76%, respectively; P = .03), while spending less time on each report (58%; 95% CI: 0.53, 0.62). Conclusion GPT-4 created near-perfect PDAC synoptic reports from original reports. GPT-4 with chain-of-thought achieved high accuracy in categorizing resectability. Surgeons were more accurate and efficient using AI-generated reports. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Chang in this issue.

The multidisciplinary pancreatic cyst clinic: A collaborative approach to streamlined management.

BACKGROUND: Pancreatic cysts are often incidentally detected on routine imaging studies. Of these, mucinous cysts have a malignant potential. Several guidelines propose different management strategies, and implementation in patient care is inconsistent in the absence of dedicated infrastructure. METHODS: To address the challenges of pancreatic cyst diagnosis and management, we established a multidisciplinary pancreas cyst clinic (PCC) within our health system. This clinic encompasses both tertiary care academic centers and community hospitals, with leadership from surgical oncology, gastroenterology, and radiology. Our PCC's primary goal is to provide accurate diagnosis and tailored management recommendations for all patients with pancreatic cysts. Additionally, we maintain a prospective database to study the disease's natural history and the outcomes of various treatment strategies. CLINIC INFRASTRUCTURE: The clinic meets once per week for 45 min virtually via Zoom in the mornings. Patients are referred via electronic medical record (EMR) order, telephone call, or email from patient or referring provider. A dedicated advanced practice provider reviews referrals several times per day, calls patients to gather clinical data, ensures imaging is uploaded, and coordinates logistical aspects of the meeting during the dedicated time. Conferences are attended by representatives from surgery, radiology, medical pancreatology, and interventional gastroenterology. Each patient case is reviewed in detail and recommendations are submitted to referring providers and patients via an EMR message and letter. For patients requiring imaging surveillance, patients are followed longitudinally by the referring provider, gastroenterology team, or surgical team. For patients requiring endoscopic ultrasound (EUS) or surgical consultation, expedited referral to these services is made with prompt subsequent evaluation. RESULTS: A total of 1052 patients from our health system were evaluated between 2020 and 2021. Of these, 196 (18.6 %) underwent EUS, 41 (3.9 %) underwent upfront surgical resection, and the remainder were referred to gastroenterology (141-13.4 %), surgery (314-29.8 %), or back to their referring provider (597-56.7 %) for ongoing surveillance in collaboration with their primary care provider (PCP). Of cysts under surveillance, 61.3 % remained stable, 13.2 % increased in size, and 2 % decreased in size. A total of 2.3 % of patients were recommended to discontinue surveillance. CONCLUSIONS: The PCC provides infrastructure that has served to provide multidisciplinary review and consensus recommendations to patients with pancreatic cysts. This has served to improve the application of guidelines while providing individualized recommendations to each patient, while aiding non-expert referring providers throughout the region.

Intra-patient variability of iodine quantification across different dual-energy CT platforms: assessment of normalization techniques.

OBJECTIVES: To investigate intra-patient variability of iodine concentration (IC) between three different dual-energy CT (DECT) platforms and to test different normalization approaches. METHODS: Forty-four patients who underwent portal venous phase abdominal DECT on a dual-source (dsDECT), a rapid kVp switching (rsDECT), and a dual-layer detector platform (dlDECT) during cancer follow-up were retrospectively included. IC in the liver, pancreas, and kidneys and different normalized ICs (NICPV:portal vein; NICAA:abdominal aorta; NICALL:overall iodine load) were compared between the three DECT scanners for each patient. A longitudinal mixed effects analysis was conducted to elucidate the effect of the scanner type, scan order, inter-scan time, and contrast media amount on normalized iodine concentration. RESULTS: Variability of IC was highest in the liver (dsDECT vs. dlDECT 28.96 (14.28-46.87) %, dsDECT vs. rsDECT 29.08 (16.59-62.55) %, rsDECT vs. dlDECT 22.85 (7.52-33.49) %), and lowest in the kidneys (dsDECT vs. dlDECT 15.76 (7.03-26.1) %, dsDECT vs. rsDECT 15.67 (8.86-25.56) %, rsDECT vs. dlDECT 10.92 (4.92-22.79) %). NICALL yielded the best reduction of IC variability throughout all tissues and inter-scanner comparisons, yet did not reduce the variability between dsDECT vs. dlDECT and rsDECT, respectively, in the liver. The scanner type remained a significant determinant for NICALL in the pancreas and the liver (F-values, 12.26 and 23.78; both, p < 0.0001). CONCLUSIONS: We found tissue-specific intra-patient variability of IC across different DECT scanner types. Normalization mitigated variability by reducing physiological fluctuations in iodine distribution. After normalization, the scanner type still had a significant effect on iodine variability in the pancreas and liver. CLINICAL RELEVANCE STATEMENT: Differences in iodine quantification between dual-energy CT scanners can partly be mitigated by normalization, yet remain relevant for specific tissues and inter-scanner comparisons, which should be taken into account at clinical routine imaging. KEY POINTS: • Iodine concentration showed the least variability between scanner types in the kidneys (range 10.92-15.76%) and highest variability in the liver (range 22.85-29.08%). • Normalizing tissue-specific iodine concentrations against the overall iodine load yielded the greatest reduction of variability between scanner types for 2/3 inter-scanner comparisons in the liver and for all (3/3) inter-scanner comparisons in the kidneys and pancreas, respectively. • However, even after normalization, the dual-energy CT scanner type was found to be the factor significantly influencing variability of iodine concentration in the liver and pancreas.

Implementation of an Institution-Wide Rules-Based Automated CT Protocoling System.

BACKGROUND. Examination protocoling is a noninterpretive task that increases radiologists' workload and can cause workflow inefficiencies. OBJECTIVE. The purpose of this study was to evaluate effects of an automated CT protocoling system on examination process times and protocol error rates. METHODS. This retrospective study included 317,597 CT examinations (mean age, 61.8 ± 18.1 [SD] years; male, 161,125; female, 156,447; unspecified sex, 25) from July 2020 to June 2022. A rules-based automated protocoling system was implemented institution-wide; the system evaluated all CT orders in the EHR and assigned a protocol or directed the order for manual radiologist protocoling. The study period comprised pilot (July 2020 to December 2020), implementation (January 2021 to December 2021), and postimplementation (January 2022 to June 2022) phases. Proportions of automatically protocoled examinations were summarized. Process times were recorded. Protocol error rates were assessed by counts of quality improvement (QI) reports and examination recalls and comparison with retrospectively assigned protocols in 450 randomly selected examinations. RESULTS. Frequency of automatic protocoling was 19,366/70,780 (27.4%), 68,875/163,068 (42.2%), and 54,045/83,749 (64.5%) in pilot, implementation, and postimplementation phases, respectively (p < .001). Mean (± SD) times from order entry to protocol assignment for automatically and manually protocoled examinations for emergency department examinations were 0.2 ± 18.2 and 2.1 ± 69.7 hours, respectively; mean inpatient examination times were 0.5 ± 50.0 and 3.5 ± 105.5 hours; and mean outpatient examination times were 361.7 ± 1165.5 and 1289.9 ± 2050.9 hours (all p < .001). Mean (± SD) times from order entry to examination completion for automatically and manually protocoled examinations for emergency department examinations were 2.6 ± 38.6 and 4.2 ± 73.0 hours, respectively (p < .001); for inpatient examinations were 6.3 ± 74.6 and 8.7 ± 109.3 hours (p = .001); and for outpatient examinations were 1367.2 ± 1795.8 and 1471.8 ± 2118.3 hours (p < .001). In the three phases, there were three, 19, and 25 QI reports and zero, one, and three recalls, respectively, for automatically protocoled examinations, versus nine, 19, and five QI reports and one, seven, and zero recalls for manually protocoled examinations. Retrospectively assigned protocols were concordant with 212/214 (99.1%) of automatically protocoled versus 233/236 (98.7%) of manually protocoled examinations. CONCLUSION. The automated protocoling system substantially reduced radiologists' protocoling workload and decreased times from order entry to protocol assignment and examination completion; protocol errors and recalls were infrequent. CLINICAL IMPACT. The system represents a solution for reducing radiologists' time spent performing noninterpretive tasks and improving care efficiency.

Society of Radiologists in Ultrasound Consensus Conference Recommendations for Incidental Gallbladder Polyp Management: Interreader Agreement Among 10 Radiologists.

BACKGROUND. The 2022 Society of Radiologists in Ultrasound (SRU) consensus conference recommendations for small gallbladder polyps support management that is less aggressive than earlier approaches and may help standardize evaluation of polyps by radiologists. OBJECTIVE. The purpose of the present study was to assess the interreader agreement of radiologists in applying SRU recommendations for management of incidental gallbladder polyps on ultrasound. METHODS. This retrospective study included 105 patients (75 women and 30 men; median age, 51 years) with a gallbladder polyp on ultrasound (without features highly suspicious for invasive or malignant tumor) who underwent cholecystectomy between January 1, 2003, and January 1, 2021. Ten abdominal radiologists independently reviewed ultrasound examinations and, using the SRU recommendations, assessed one polyp per patient to assign risk category (extremely low risk, low risk, or indeterminate risk) and make a possible recommendation for surgical consultation. Five radiologists were considered less experienced (< 5 years of experience), and five were considered more experienced (≥ 5 years of experience). Interreader agreement was evaluated. Polyps were classified pathologically as nonneoplastic or neoplastic. RESULTS. For risk category assignments, interreader agreement was substantial among all readers (k = 0.710), less-experienced readers (k = 0.705), and more-experienced readers (k = 0.692). For surgical consultation recommendations, inter-reader agreement was substantial among all readers (k = 0.795) and more-experienced readers (k = 0.740) and was almost perfect among less-experienced readers (k = 0.811). Of 10 readers, a median of 5.0 (IQR, 2.0-8.0), 4.0 (IQR, 2.0-7.0), and 0.0 (IQR, 0.0-0.0) readers classified polyps as extremely low risk, low risk, and indeterminate risk, respectively. Across readers, the percentage of polyps classified as extremely low risk ranged from 32% to 72%; as low risk, from 24% to 65%; and as indeterminate risk, from 0% to 8%. Of 10 readers, a median of zero change to 0 (IQR, 0.0-1.0) readers recommended surgical consultation; the percentage of polyps receiving a recommendation for surgical consultation ranged from 4% to 22%. Of a total of 105 polyps, 102 were nonneo-plastic and three were neoplastic (all benign). Based on readers' most common assessments for nonneoplastic polyps, the risk category was extremely low risk for 53 polyps, low risk for 48 polyps, and indeterminate risk for one polyp; surgical consultation was recommended for 16 polyps. CONCLUSION. Ten abdominal radiologists showed substantial agreement for polyp risk categorizations and surgical consultation recommendations, although areas of reader variability were identified. CLINICAL IMPACT. The findings support the overall reproducibility of the SRU recommendations, while indicating opportunity for improvement.

Imaging for Early Detection of Pancreatic Ductal Adenocarcinoma: Updates and Challenges in the Implementation of Screening and Surveillance Programs.

Pancreatic ductal adenocarcinoma (PDA) is one of the most aggressive cancers. It has a poor 5-year survival rate of 12%, partly because most cases are diagnosed at advanced stages, precluding curative surgical resection. Early-stage PDA has significantly better prognoses due to increased potential for curative interventions, making early detection of PDA critically important to improved patient outcomes. We examine current and evolving early detection concepts, screening strategies, diagnostic yields among high-risk individuals, controversies, and limitations of standard-of-care imaging.

Part 2: Current Concepts in Radiologic Imaging & Intervention in Acute Biliary Tract Diseases.

Acute cholangitis is encountered commonly in critically ill, often elderly, patients. The most common causes of cholangitis include choledocholithiasis, biliary strictures, and infection from previous endoscopic, percutaneous, or surgical intervention of the biliary tract. Rare causes of acute cholangitis in the United States include sclerosing cholangitis and recurrent pyogenic cholangitis, the latter predominantly occurring in immigrants of Asian descent. Multidisciplinary management of these conditions is essential, with intensivists, surgeons, diagnostic radiologists, interventional radiologists, gastroenterologists, endoscopists, and infectious disease physicians typically involved in the care of these patients. In this paper intended for intensivists predominantly, we will review the imaging findings and radiologic interventional management of critically ill patients with acute cholangitis, primary and secondary sclerosing cholangitis, and recurrent pyogenic cholangitis.

Part 1: Current Concepts in Radiologic Imaging and Intervention in Acute Cholecystitis.

Acute calculous cholecystitis and acute acalculous cholecystitis are encountered commonly among critically ill, often elderly, patients. Multidisciplinary management of these conditions is essential, with intensivists, surgeons, diagnostic radiologists, interventional radiologists, infectious disease physicians, gastroenterologists, and endoscopists able to contribute to patient care. In this article intended predominantly for intensivists, we will review the imaging findings and radiologic treatment of critically ill patients with acute calculous cholecystitis and acute acalculous cholecystitis.

Diagnostic Anatomic Imaging for Neuroendocrine Neoplasms: Maximizing Strengths and Mitigating Weaknesses.

ABSTRACT: Neuroendocrine neoplasms are a heterogeneous group of gastrointestinal and lung tumors. Their diverse clinical manifestations, variable locations, and heterogeneity present notable diagnostic challenges. This article delves into the imaging modalities vital for their detection and characterization. Computed tomography is essential for initial assessment and staging. At the same time, magnetic resonance imaging (MRI) is particularly adept for liver, pancreatic, osseous, and rectal imaging, offering superior soft tissue contrast. The article also highlights the limitations of these imaging techniques, such as MRI's inability to effectively evaluate the cortical bone and the questioned cost-effectiveness of computed tomography and MRI for detecting specific gastric lesions. By emphasizing the strengths and weaknesses of these imaging techniques, the review offers insights into optimizing their utilization for improved diagnosis, staging, and therapeutic management of neuroendocrine neoplasms.

Pancreatic cancer detection with dual-energy CT: diagnostic performance of 40 keV and 70 keV virtual monoenergetic images.

PURPOSE: To compare the diagnostic performance of 40 keV and 70 keV virtual monoenergetic images (VMIs) generated from dual-energy CT in the detection of pancreatic cancer. METHODS: This retrospective study included patients who underwent pancreatic protocol dual-energy CT from January 2019 to August 2022. Four radiologists (1-11 years of experience), who were blinded to the final diagnosis, independently and randomly interpreted 40 keV and 70 keV VMIs and graded the presence or absence of pancreatic cancer. For each image set (40 keV and 70 keV VMIs), the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated. The diagnostic performance of each image set was compared using generalized estimating equations. RESULTS: Overall, 137 patients (median age, 71 years; interquartile range, 63-78 years; 77 men) were included. Among them, 62 patients (45%) had pathologically proven pancreatic cancer. The 40 keV VMIs had higher specificity (75% vs. 67%; P < .001), PPV (76% vs. 71%; P < .001), and accuracy (85% vs. 81%; P = .001) than the 70 keV VMIs. On the contrary, 40 keV VMIs had lower sensitivity (96% vs. 98%; P = .02) and NPV (96% vs. 98%; P = .004) than 70 keV VMIs. However, the diagnostic confidence in patients with (P < .001) and without (P = .001) pancreatic cancer was improved in 40 keV VMIs than in 70 keV VMIs. CONCLUSIONS: The 40 keV VMIs showed better diagnostic performance in diagnosing pancreatic cancer than the 70 keV VMIs, along with higher reader confidence.

Uncinate Duct Dilatation Predicts Additional Risk for High-Grade Dysplasia or Invasive Carcinoma Among Fukuoka-Positive Intraductal Papillary Mucinous Neoplasms.

OBJECTIVE: To determine whether uncinate duct dilatation (UDD) increases the risk of high-grade dysplasia or invasive carcinoma (HGD/IC) in Fukuoka-positive intraductal papillary mucinous neoplasms (IPMNs). BACKGROUND: Though classified as a branch duct, the uncinate duct is the primary duct of the pancreatic ventral anlage. We hypothesized that UDD, like main duct dilatation, confers additional risk for HGD/IC. METHODS: A total of 467 patients met inclusion criteria in a retrospective cohort study of surgically resected IPMNs at the Massachusetts General Hospital. We used multivariable logistic regression to analyze the association between UDD (defined as ≥4 mm) and HGD/IC, controlling for Fukuoka risk criteria. In a secondary analysis, the modeling was repeated in the 194 patients with dorsal branch duct IPMNs (BD-IPMNs) in the pancreatic neck, body, or tail. RESULTS: Mean age at surgery was 70, and 229 (49%) patients were female. In total, 267 (57%) patients had only worrisome features and 200 (43%) had at least 1 high-risk feature. UDD was present in 164 (35%) patients, of whom 118 (73%) had HGD/IC. On multivariable analysis, UDD increased the odds of HGD/IC by 2.8-fold, even while controlling for Fukuoka risk factors (95% CI: 1.8-4.4, P <0.001). Prevalence of HGD/IC in all patients with UDD was 73%, compared with 74% in patients with high-risk stigmata and 73% in patients with main duct IPMNs. In the secondary analysis, UDD increased the odds of HGD/IC by 3.2-fold in patients with dorsal BD-IPMNs (95% CI: 1.3-7.7, P =0.010). CONCLUSIONS: UDD confers additional risk for HGD/IC unaccounted for by current Fukuoka criteria. Further research can extend this study to Fukuoka-negative patients, including unresected patients.

"Evolving Trends in Pancreatic Cystic Tumors: A 3-Decade Single-Center Experience With 1290 Resections".

OBJECTIVE: The aim of this study was to describe our institutional experience with resected cystic tumors of the pancreas with emphasis on changes in clinical presentation and accuracy of preoperative diagnosis. SUMMARY BACKGROUND DATA: Incidental discovery of pancreatic cystic lesions has increased and has led to a rise in pancreatic resections. It is important to analyze surgical outcomes from these procedures, and the prevalence of malignancy, pre-malignancy and resections for purely benign lesions, some of which may be unintended. METHODS: Retrospective review of a prospective database spanning 3 decades. Presence of symptoms, incidental discovery, diagnostic studies, type of surgery, postoperative outcomes, and concordance between presumptive diagnosis and final histopathology were recorded. RESULTS: A total of 1290 patients were identified, 62% female with mean age of 60 years. Fifty-seven percent of tumors were incidentally discovered. Ninety-day operative mortality was 0.9% and major morbidity 14.4%. There were 23 different diagnosis, but IPMN, MCN, and serous cystadenoma comprised 80% of cases. Concordance between preoperative and final histopathological diagnosis increased by decade from 45%, to 68%, and is presently 80%, rising in parallel with the use of endoscopic ultrasound, cytology, and molecular analysis. The addition of molecular analysis improved accuracy to 91%. Of misdiagnosed cases, half were purely benign and taken to surgery with the presumption of malignancy or premalignancy. The majority of these were serous cystadenomas. CONCLUSIONS: Indications and diagnostic work-up of cystic tumors of the pancreas have changed over time. Surgical resection can be performed with very low mortality and acceptable morbidity and diagnostic accuracy is presently 80%. About 10% of patients are still undergoing surgery for purely benign lesions that were presumed to be malignant or premalignant. Further refinements in diagnostic tests are required to improve accuracy.

Resection of NAFLD/NASH-related Hepatocellular Carcinoma (HCC): Clinical Features and Outcomes Compared with HCC Due to Other Etiologies.

BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH) are the leading causes of hepatocellular carcinoma (HCC) worldwide. Limited data exist on surgical outcomes for NAFLD/NASH-related HCC compared with other HCC etiologies. We evaluated differences in clinicopathological characteristics and outcomes of patients undergoing surgical resection for NAFLD/NASH-associated HCC compared with other HCC etiologies. METHODS: Demographic, clinicopathological features, and survival outcomes of patients with surgically resected HCC were collected. NAFLD activity score (NAS) and fibrosis score were assessed by focused pathologic review in a subset of patients. RESULTS: Among 492 patients screened, 260 met eligibility (NAFLD/NASH [n = 110], and other etiologies [n = 150]). Median age at diagnosis was higher in the NAFLD/NASH HCC cohort compared with the other etiologies cohort (66.7 vs. 63.4 years, respectively, P = .005), with an increased percentage of female patients (36% vs. 18%, P = .001). NAFLD/NASH-related tumors were more commonly >5 cm (66.0% vs. 45%, P = .001). There were no significant differences in rates of lymphovascular or perineural invasion, histologic grade, or serum AFP levels. The NAFLD/NASH cohort had lower rates of background liver fibrosis, lower AST and ALT levels, and higher platelet counts (P < .01 for all). Median overall survival (OS) was numerically shorter in NAFLD/NASH vs other etiology groups, however, not statistically significant. CONCLUSIONS: Patients with NAFLD/NASH-related HCC more commonly lacked liver fibrosis and presented with larger HCCs compared with patients with HCC from other etiologies. No differences were seen in rates of other high-risk features or survival. With the caveat of sample size and retrospective analysis, this supports a similar decision-making approach regarding surgical resection for NAFLD/NASH and other etiology-related HCCs.

A Multicenter Assessment of Interreader Reliability of LI-RADS Version 2018 for MRI and CT.

Background Various limitations have impacted research evaluating reader agreement for Liver Imaging Reporting and Data System (LI-RADS). Purpose To assess reader agreement of LI-RADS in an international multicenter multireader setting using scrollable images. Materials and Methods This retrospective study used deidentified clinical multiphase CT and MRI and reports with at least one untreated observation from six institutions and three countries; only qualifying examinations were submitted. Examination dates were October 2017 to August 2018 at the coordinating center. One untreated observation per examination was randomly selected using observation identifiers, and its clinically assigned features were extracted from the report. The corresponding LI-RADS version 2018 category was computed as a rescored clinical read. Each examination was randomly assigned to two of 43 research readers who independently scored the observation. Agreement for an ordinal modified four-category LI-RADS scale (LR-1, definitely benign; LR-2, probably benign; LR-3, intermediate probability of malignancy; LR-4, probably hepatocellular carcinoma [HCC]; LR-5, definitely HCC; LR-M, probably malignant but not HCC specific; and LR-TIV, tumor in vein) was computed using intraclass correlation coefficients (ICCs). Agreement was also computed for dichotomized malignancy (LR-4, LR-5, LR-M, and LR-TIV), LR-5, and LR-M. Agreement was compared between research-versus-research reads and research-versus-clinical reads. Results The study population consisted of 484 patients (mean age, 62 years ± 10 [SD]; 156 women; 93 CT examinations, 391 MRI examinations). ICCs for ordinal LI-RADS, dichotomized malignancy, LR-5, and LR-M were 0.68 (95% CI: 0.61, 0.73), 0.63 (95% CI: 0.55, 0.70), 0.58 (95% CI: 0.50, 0.66), and 0.46 (95% CI: 0.31, 0.61) respectively. Research-versus-research reader agreement was higher than research-versus-clinical agreement for modified four-category LI-RADS (ICC, 0.68 vs 0.62, respectively; P = .03) and for dichotomized malignancy (ICC, 0.63 vs 0.53, respectively; P = .005), but not for LR-5 (P = .14) or LR-M (P = .94). Conclusion There was moderate agreement for LI-RADS version 2018 overall. For some comparisons, research-versus-research reader agreement was higher than research-versus-clinical reader agreement, indicating differences between the clinical and research environments that warrant further study. © RSNA, 2023 Supplemental material is available for this article. See also the editorials by Johnson and Galgano and Smith in this issue.

Renal lesion characterization: clinical utility of single-phase dual-energy CT compared to MRI and dual-phase single-energy CT.

OBJECTIVES: To assess the impact of dual-energy CT (DECT) utilization in practice by measuring the readers' confidence, the need for additional image requests, and diagnostic performance in renal lesion assessment, compared to single-energy CT (SECT) using contrast-enhanced MRI to establish the reference standard. MATERIALS AND METHODS: Sixty-nine patients (M/F = 47/22) who underwent a dual-phase renal SECT (n = 34) or DECT (n = 35) and had a contrast-enhanced MRI within 180 days were retrospectively collected. Three radiologists assessed images on different sessions (SECT, DECT, and MRI) for (1) likely diagnosis (enhancing/non-enhancing); (2) diagnostic confidence (5-point Likert scale); (3) need for additional imaging test (yes/no); and (4) need for follow-up imaging (yes/no). Diagnostic accuracy was compared using AUC; p value < 0.05 was considered significant. RESULTS: One hundred fifty-six lesions consisting of 18% enhancing (n = 28/156, mean size: 30.37 mm, range: 9.9-94 mm) and 82% non-enhancing (n = 128/156, mean size: 23.91 mm, range: 5.0-94.2 mm) were included. The confidence level was significantly lower for SECT than their MRI (4.50 vs. 4.80, p value < 0.05) but not significantly different for DECT and the corresponding MRI (4.78 vs. 4.78, p > 0.05). There were significantly more requests for additional imaging in the SECT session than the corresponding MRI (20% vs. 4%), which was not significantly different between DECT and their MRI counterpart session (5.7% vs. 4.9%). Inter-reader agreement was almost perfect for DECT and MRI (kappa: 0.8-1) and substantial in SECT sessions (kappa: 0.6-0.8) with comparable diagnostic accuracy between SECT, DECT, and MRI (p value > 0.05). CONCLUSION: Single-phase DECT allows confident and reproducible characterization of renal masses with fewer recommendation for additional and follow-up imaging tests than dual-phase SECT and a performance similar to MRI. KEY POINTS: • DECT utilization leads to similar additional image requests to MRI (5.7% vs. 4.9%, p value > 0.05), whereas single-energy CT utilization leads to significantly higher image requests (20% vs. 4%, p value < 0.05). • DECT and MRI utilization bring highly reproducible results with almost perfect inter-reader agreement (kappa: 0.8-1), better than the inter-reader agreement in SECT utilization (kappa: 0.6-0.8). • Readers' confidence was not significantly altered between DECT and their MRI readout session (p value > 0.05). In contrast, confidence in the diagnosis was significantly lower in the SECT session than their MRI readout (p value < 0.05).

CT image quality evaluation in the age of deep learning: trade-off between functionality and fidelity.

OBJECTIVE: To quantitatively compare DLIR and ASiR-V with realistic anatomical images. METHODS: CT scans of an anthropomorphic phantom were acquired using three routine protocols (brain, chest, and abdomen) at four dose levels, with images reconstructed at five levels of ASiR-V and three levels of DLIR. Noise power spectrum (NPS) was estimated using a difference image by subtracting two matching images from repeated scans. Using the max-dose FBP reconstruction as the ground truth, the structure similarity index (SSIM) and gradient magnitude (GM) of difference images were evaluated. Image noise magnitude (σ), frequency location of the NPS peak (fpeak), mean SSIM (MSSIM), and mean GM (MGM) were used as quantitative metrics to compare image quality, for each anatomical region, protocol, algorithm, dose level, and slice thickness. RESULTS: Image noise had a strong (R2 > 0.99) power law relationship with dose for all algorithms. For the abdomen and chest, fpeak shifted from 0.3 (FBP) down to 0.15 mm-1 (ASiR-V 100%) with increasing ASiR-V strength but remained 0.3 mm-1 for all DLIR levels. fpeak shifted down for the brain protocol with increasing DLIR levels. Three levels of DLIR produced similar image noise levels as ASiR-V 40%, 80%, and 100%, respectively. DLIR had lower MSSIM but higher MGM than ASiR-V while matching imaging noise. CONCLUSION: Compared to ASiR-V, DLIR presents trade-offs between functionality and fidelity: it has a noise texture closer to FBP and more edge enhancement, but reduced structure similarity. These trade-offs and unique protocol-dependent behaviors of DLIR should be considered during clinical implementation and deployment. KEY POINTS: • DLIR reconstructed images demonstrate closer noise texture and lower structure similarity to FBP while producing equivalent noise levels comparable to ASiR-V. • DLIR has additional edge enhancement as compared to ASiR-V. • DLIR has unique protocol-dependent behaviors that should be considered for clinical implementation.

Determination of arterial invasion in pancreatic ductal adenocarcinoma: what is the best diagnostic criterion on CT?

OBJECTIVES: To investigate the diagnostic performance and interobserver variability in the determination of arterial invasion in pancreatic ductal adenocarcinoma (PDAC) and determine the best CT imaging criterion. METHODS: We retrospectively evaluated 128 patients with PDAC (73 men and 55 women) who underwent preoperative contrast-enhanced CT. Five board-certified radiologists (expert) and four fellows (non-expert]) independently assessed the arterial invasion (celiac, superior mesenteric, splenic, and common hepatic arteries) using a 6-point score: 1, no tumor contact; 2, hazy attenuation ≤ 180°; 3, hazy attenuation > 180°; 4, solid soft tissue contact ≤ 180°; 5, solid soft tissue contact > 180°; and 6, contour irregularity. ROC analysis was performed to evaluate the diagnostic performance and determine the best diagnostic criterion for arterial invasion, with pathological or surgical findings as references. Interobserver variability was assessed using Fleiss's ĸ statistics. RESULTS: Among the 128 patients, 35.2% (n = 45/128) received neoadjuvant treatment (NTx). Solid soft tissue contact ≤ 180° was the best diagnostic criterion for arterial invasion as defined by the Youden Index both in patients who did and did not receive NTx (sensitivity, 100% vs. 100%; specificity, 90% vs. 93%; and AUC, 0.96 vs. 0.98, respectively). Interobserver variability among the non-expert was not inferior to that among the expert (ĸ = 0.61 vs 0.61; p = .39 and ĸ = 0.59 vs 0.51; p < .001 in patients treated with and without NTx, respectively). CONCLUSIONS: Solid soft tissue contact ≤ 180° was the best diagnostic criterion for the determination of arterial invasion in PDAC. Considerable interobserver variability was seen among the radiologists. KEY POINTS: • Solid soft tissue contact ≤ 180° was the best diagnostic criterion for the determination of arterial invasion in pancreatic ductal adenocarcinoma. • Interobserver agreement among non-expert radiologists was almost comparable to that among expert radiologists.

Abdominal Imaging Findings on Computed Tomography as a Tool for COVID-19 Mortality Risk Assessment: Comparison With Chest Radiograph Severity Scores.

OBJECTIVE: To quantify the association between computed tomography abdomen and pelvis with contrast (CTAP) findings and chest radiograph (CXR) severity score, and the incremental effect of incorporating CTAP findings into predictive models of COVID-19 mortality. METHODS: This retrospective study was performed at a large quaternary care medical center. All adult patients who presented to our institution between March and June 2020 with the diagnosis of COVID-19 and had a CXR up to 48 hours before a CTAP were included. Primary outcomes were the severity of lung disease before CTAP and mortality within 14 and 30 days. Logistic regression models were constructed to quantify the association between CXR score and CTAP findings. Penalized logistic regression models and random forests were constructed to identify key predictors (demographics, CTAP findings, and CXR score) of mortality. The discriminatory performance of these models, with and without CTAP findings, was summarized using area under the characteristic (AUC) curves. RESULTS: One hundred ninety-five patients (median age, 63 years; 119 men) were included. The odds of having CTAP findings was 3.89 times greater when a CXR score was classified as severe compared with mild (P = 0.002). When CTAP findings were included in the feature set, the AUCs for 14-day mortality were 0.67 (penalized logistic regression) and 0.71 (random forests). Similar values for 30-day mortality were 0.76 and 0.75. When CTAP findings were omitted, all AUC values were attenuated. CONCLUSIONS: The CTAP findings were associated with more severe CXR score and may serve as predictors of COVID-19 mortality.

Prediction of R Status in Resections for Pancreatic Cancer Using Simplified Radiological Criteria.

OBJECTIVE: Predicting R status before surgery for pancreatic cancer (PDAC) patients with upfront surgery and neoadjuvant therapy. SUMMARY BACKGROUND DATA: Negative surgical margins (R0) are a key predictor of long-term outcomes in PDAC. METHODS: Patients undergoing pancreatic resection with curative intent for PDAC were identified. Using the CT scans from the time of diagnosis, the 2019 NCCN borderline resectability criteria were compared to novel criteria: presence of any alteration of the superior mesenteric-portal vein (SMPV) and perivascular stranding of the superior mesenteric artery (SMA). Accuracy of predicting R status was evaluated for both criteria. Patient baseline characteristics, surgical, histopathological parameters, and long-term overall survival (OS) after resection were evaluated. RESULTS: A total of 593 patients undergoing pancreatic resections for PDAC between 2010 and 2018 were identified. Three hundred and twenty-five (54.8%) patients underwent upfront surgery, whereas 268 (45.2%) received neoadjuvant therapy. In upfront resected patients, positive SMA stranding was associated with 56% margin positive resection rates, whereas positive SMA stranding and SMPV alterations together showed a margin positive resection rate of 75%. In contrast to these criteria, the 2019 NCCN borderline criteria failed to predict margin status. In patients undergoing neoadjuvant therapy, only perivascular SMA stranding remained a predictor of margin positive resection, leading to a rate of 33% R+ resections. Perivascular SMA stranding was related to higher clinical T stage (P = 0.003) and clinical N stage (P = 0.043) as well as perineural invasion (P = 0.022). SMA stranding was associated with worse survival in both patients undergoing upfront surgery (36 vs 22 months, P = 0.002) and neoadjuvant therapy (47 vs 34 months, P = 0.050). CONCLUSIONS: The novel criteria were accurate predictors of R status in PDAC patients undergoing upfront resection. After neoadjuvant treatment, likelihood of positive resection margins is approximately halved, and only perivascular SMA stranding remained a predictive factor.

Dual-Energy CT of the Abdomen: Radiology In Training.

A 61-year-old man with an esophageal cancer diagnosis underwent staging dual-energy CT of the chest and abdomen in the portal venous phase after contrast media administration. Aside from the primary tumor and suspicious local lymph nodes, CT revealed hypoattenuating ambiguous liver lesions, an incidental right adrenal nodule, and a right renal lesion with soft-tissue attenuation. In addition, advanced atherosclerosis of the abdominal aorta and its major branches was noted. This article provides a case-based review of dual-energy CT technologies and their applications in the abdomen. The clinical utility of virtual monoenergetic images, virtual unenhanced images, and iodine maps is discussed.