Artificial intelligence to assess body composition on routine abdominal CT scans and predict mortality in pancreatic cancer- A recipe for your local application.

BACKGROUND: Body composition is associated with mortality; however its routine assessment is too time-consuming. PURPOSE: To demonstrate the value of artificial intelligence (AI) to extract body composition measures from routine studies, we aimed to develop a fully automated AI approach to measure fat and muscles masses, to validate its clinical discriminatory value, and to provide the code, training data and workflow solutions to facilitate its integration into local practice. METHODS: We developed a neural network that quantified the tissue components at the L3 vertebral body level using data from the Liver Tumor Challenge (LiTS) and a pancreatic cancer cohort. We classified sarcopenia using accepted skeletal muscle index cut-offs and visceral fat based its median value. We used Kaplan Meier curves and Cox regression analysis to assess the association between these measures and mortality. RESULTS: Applying the algorithm trained on LiTS data to the local cohort yielded good agreement [>0.8 intraclass correlation (ICC)]; when trained on both datasets, it had excellent agreement (>0.9 ICC). The pancreatic cancer cohort had 136 patients (mean age: 67 ± 11 years; 54% women); 15% had sarcopenia; mean visceral fat was 142 cm2. Concurrent with prior research, we found a significant association between sarcopenia and mortality [mean survival of 15 ± 12 vs. 22 ± 12 (p < 0.05), adjusted HR of 1.58 (95% CI: 1.03-3.33)] but no association between visceral fat and mortality. The detector analysis took 1 ± 0.5 s. CONCLUSIONS: AI body composition analysis can provide meaningful imaging biomarkers from routine exams demonstrating AI's ability to further enhance the clinical value of radiology reports.

Use of CEUS LI-RADS for the Accurate Diagnosis of Nodules in Patients at Risk for Hepatocellular Carcinoma: A Validation Study.

Purpose: To validate the contrast agent-enhanced US Liver Imaging Reporting and Data System (CEUS LI-RADS) algorithm for accurate diagnosis of hepatocellular carcinoma (HCC) and categorization of all nodules encountered in patients at risk for HCC. Materials and Methods: A single-center retrospective review of 196 nodules in 184 patients at risk for HCC (consisting of 139 HCCs, 18 non-HCC malignancies, and 39 benign nodules) was performed in a three-reader blinded read format, with the use of the CEUS LI-RADS algorithm. Pathologic confirmation was available for 143 nodules (122 HCCs, 18 non-HCC malignancies, and three benign nodules). Nodule sizes ranged between 1.0 and 16.2 cm. Nodules assessed with contrast-enhanced US were assigned various CEUS LI-RADS categories by three blinded readers. CEUS LI-RADS categorization was then compared against histopathologic findings, concurrent CT, and/or MR images or follow-up imaging to assess diagnostic accuracy of CEUS LI-RADS. In addition, the proportion of HCC in all LI-RADS (LR) categories, univariable and multivariable feature analysis, and interrater agreement using Light κ were determined. Results: The LR-5 category, determined through radiologist categorization of nodules using the CEUS LI-RADS criteria, showed sensitivity, specificity, positive predictive value, and negative predictive value of 86% (119 of 139), 96% (55 of 57), 98% (119 of 121), and 73% (55 of 75), respectively, for the diagnosis of HCC. Two false-positive cases of LR-5 included a cholangiocarcinoma and a combined hepatocellular and cholangiocarcinoma. The remainder of the cholangiocarcinomas in the sample (n = 8) were appropriately categorized as LR-M. Multivariable logistic regression analysis showed that washout of greater than 60 seconds was the contrast-enhanced US feature most predictive of HCC diagnosis, whereas washout of less than 60 seconds was the feature most predictive of nonhepatocellular malignancy. The proportion of HCC nodules categorized in the LR-M and LR-4 categories was 35% and 20%, respectively. Light κ agreement between readers for LI-RADS categorization was 90%. Conclusion: This study showed excellent specificity for the CEUS LI-RADS LR-5 category, allowing for confident imaging diagnosis of HCC, without necessity for pathologic confirmation. In addition, there was accurate differentiation of HCC from non-HCC malignancies and benign nodules. Only a single cholangiocarcinoma was misdiagnosed as category LR-5, with the remainder of the cholangiocarcinomas in the sample appropriately characterized as category LR-M.Keywords: Abdomen/GI, Evidence Based Medicine, Liver, Neoplasms-Primary, Ultrasound-Contrast© RSNA, 2020.

Bipolar radiofrequency ablation of aneurysm remnants after coil embolization can improve endovascular treatment of experimental bifurcation aneurysms.

OBJECTIVE Endovascular treatment of aneurysms may result in incomplete initial occlusion and aneurysm recurrence at angiographic follow-up studies. This study aimed to assess the feasibility and efficacy of bipolar radiofrequency ablation (RFA) of aneurysm remnants after coil embolization. METHODS Bipolar RFA was accomplished using the coil mass as 1 electrode, while the second electrode was a stent placed across the aneurysmal neck. After preliminary experiments and protocol approval from the Animal Care committee, wide-necked bifurcation aneurysms were constructed in 24 animals. Aneurysms were allocated to 1 of 3 groups: partial intraoperative coil embolization, followed by RFA (n = 12; treated group) or without RFA (n = 6; control group 1); or attempted complete endovascular coil embolization 2-4 weeks later (n = 6; control group 2). Angiographic results were compared at baseline, immediately after RFA, and at 12 weeks, using an ordinal scale. Pathological results and neointima formation at the neck were compared using a semiquantitative grading scale. RESULTS Bipolar RFA was able to reliably target the aneurysm neck when the coil mass and stent were used as electrodes. RFA improved angiographic results immediately after partial coiling (p = 0.0024). Two RFA-related complications occurred, involving transient occlusion of 1 carotid artery and 1 hemorrhage from an adventitial arterial blister. At 12 weeks, angiographic results were improved with RFA (median score of 0), when compared with controls (median score of 2; p = 0.0013). Neointimal closure of the aneurysm neck was better with RFA compared with controls (p = 0.0003). CONCLUSIONS Bipolar RFA can improve results of embolization in experimental models by selectively ablating residual lesions after coil embolization.

Flow diversion of giant curved sidewall and bifurcation experimental aneurysms with very-low-porosity devices.

OBJECTIVE: Flow diverters (FDs) are increasingly used to treat difficult intracranial aneurysms. The objective of this study was to test whether treatment challenges posed by giant curved sidewall (cSW) and endwall bifurcation (EwB) aneurysms can be overcome with the use of very-low-porosity devices. METHODS: Large and giant EwB (n = 12) and cSW aneurysms (n = 5) were constructed in 17 dogs. EwB aneurysms were treated with 48 (n = 4), 64 (n = 4), or two overlapping 64-wire low-porosity devices (n = 4), whereas all cSW aneurysms were treated with single 64-wire devices. Angiographic results were recorded immediately and at 12 weeks before euthanasia. Pathologic specimens were photographed and neointimal coverage of devices measured and scored. RESULTS: By 12 weeks, 1 of 12 EwB and 1 of 5 cSW aneurysms were occluded. All other aneurysms were patent. Device-related arterial stenoses occurred in 13 of 17 animals, hemodynamically significant in two. All branches jailed by the FDs remained patent. There was a significant correlation between angiographic scores and the degree of neointima formation on the device (Rho = 0.527; P = 0.04). Failures of aneurysm occlusion could be explained by holes, sometimes barely visible, in the neointima that formed over FDs. CONCLUSION: Low-porosity FDs fail to reliably occlude experimental giant EwB and cSW aneurysms.

In vitro reproduction of device deformation leading to thrombotic complications and failure of flow diversion.

Flow diverters (FDs) are increasingly used for complex intracranial aneurysms. As these self-expanding devices are deployed across an aneurysm neck, they can undergo deformations. The potential clinical consequences of FD deformations remain unclear. We describe an immediate thrombotic complication attributed to a stereotypical stenotic deformation of an FD extremity that can occur when landing zones are of insufficient length. This case is supplemented with in vitro studies showing the relationship between i) the length of the landing zones and ii) discrepancies between the diameter of the device and recipient vessel, and the severity of FD stenosis. In vitro, a shorter landing zone was associated with a progressive stenotic deformation of the terminal ends of all FDs studied. This deformation was more pronounced when the diameter of the device was oversized compared to the size of the recipient tube. In our clinical case, the presence of this deformation led to an immediate thrombotic complication, requiring deployment of a second stent to correct the observed stenosis. In addition, treatment failure ultimately led to a fatal rupture, a failure that can be explained by residual flows through a more porous transition zone, another characteristic FD deformation which occurs when they are oversized as compared to the parent vessel, but free to expand at the level of the aneurysm. Proper selection of device diameter and length of the landing zone is important, and may decrease the incidence of deformation-related complications.

Y-crossing of braided stents with stents and flow diverters does not cause significant stenosis in bench-top studies.

Y-stent placement to treat bifurcation aneurysms requires the second device to cross the confines of the first stent, with concerns regarding the formation of stenosis of the second device at the site of crossing. Various braided stents and flow diverters (FDs) were deployed to cross through a high porosity braided stent, in a Y configuration, with the ends of the devices inserted in plastic tubes of various diameters, leaving the mid-portion free to expand. The ensuing constructs were photographed, paying attention to the degree of stenosis, if any, created where the second device crosses the first stent. Experiments were repeated selecting different zones of the first stent as the site of crossing for the second device, different tube diameters, and changing the angle of the bifurcation. Crossing the first stent did not cause the second stent to become significantly stenosed in any case. Crossing through the transition or expansion zone of the first device had no influence on results. Different bifurcation angles had no influence on the occurrence of stenosis. Y-stent placement to treat arterial bifurcations using braided self-expanding stents and FDs does not lead to significant stenosis in bench-top studies.

Endovascular treatment with flow diverters may fail to occlude experimental bifurcation aneurysms.

INTRODUCTION: Animal models may explain how stents and flow diverters (FDs) may succeed or fail to treat bifurcation aneurysms. METHODS: In vitro studies were designed to anticipate device deformations in bifurcations. Large, wide-necked bifurcation aneurysms were constructed in 21 animals and treated 4-8 weeks later using stents and FDs in various combinations, forming four main groups: parent artery to right branch flow diversion (RBFD, n = 6), parent artery to left branch flow diversion (LBFD, n = 6), Y flow diversion (YFD, n = 4), and compared with high-porosity Y-stenting (YHPS, n = 4). The results include immediate and follow-up angiography at 3 months, followed by grading of the extent of neointimal coverage of devices at pathology. RESULTS: In vitro, all braided devices showed varying porosities according to characteristic zones. FDs can be compacted to decrease porosities, but a limiting factor is the constant presence of a more porous transition zone. In vivo, 3/6 RBFD, 4/6 LBFD, and 2/4 YFD treated aneurysms had decreased in size by 3 months, while those treated with YHPS increased in size (P = 0.15). There was a significant correlation between device porosities and extent of neointimal coverage (r = 0.639, P = 0.002), and between porosities and angiographic evolution (r = -0.655, P = 0.002), but not between neointima formation and angiographic evolution (r = -0.278, P = 0.235). Failures could be explained in all cases by the presence of leaks or holes in the neointima at the level of the transition zones. CONCLUSION: FDs did not successfully treat most bifurcation aneurysms, at least in this animal model.

Stents and flow diverters in the treatment of aneurysms: device deformation in vivo may alter porosity and impact efficacy.

INTRODUCTION: High-porosity (HP) and flow-diverting (FD) stents are increasingly used to treat intracranial aneurysms. In vivo device deformations and their impact on the porosity of the segment of device lying over the aneurysm neck remain inadequately characterized. METHODS: Porosities of different braided FDs were studied in straight and 90° curved glass tubes. In vivo, 11 experimental lateral wall aneurysms were treated with FD (n = 7) or HP (n = 4) stents. At 3 months, the segment of FDs and HP stents over the aneurysm neck was analyzed, paying attention to changes in device diameter, metallic porosity, and neointimal closure of pores over the aneurysm or branch ostia. Device deformations were reproduced with benchtop experiments. RESULTS: In 90° curved tubes, FD porosity was higher (P = 0.025) and pore density was lower (P = 0.01) on convex compared to concave surfaces, but variations remained within 5-10 %. After in vivo deployment, a spindle-shaped deformation of FDs occurred, with focal expansion at the level of the aneurysm neck (P = 0.004). This deformation translated into an accordion-like distribution of stent struts across the aneurysm neck, where porosity was not uniform. The midsection of the aneurysm ostium had more metal coverage than adjacent ostial areas (P = 0.002). Mean porosity over the aneurysm neck was 78 ± 9.4 and 32.6 ± 12.1 % for HP and FD stents, respectively (P = 0.008), decreasing to 13.0 ± 10.1 and 1.4 ± 0.6 % (P = 0.022) following neointimal coverage, respectively. Spindle-shaped deformations and accordion effects were reproduced with benchtop manipulations; fluctuations in porosity and diameter changes correlated closely (R = 0.81; P = 0.005). CONCLUSION: Alterations in porosity may occur following in vivo implantation.

Flow diversion to treat aneurysms: the free segment of stent.

PURPOSE: Flow diverters (FDs) have led to spectacular results in otherwise untreatable aneurysm cases, but complications can occur. There is a pressing need to study factors that might predict their safety and efficacy. METHODS: The anatomical constraints that may impact on the ability of FDs to redirect blood flow and provide a scaffold for neointima formation across the aneurysm or branch ostia are explored and classified. A nomenclature is needed to identify the key factors that should be taken into account before contemplating the use of FDs in clinical aneurysms, and that should be reproduced in experimental models, if they are to guide safe clinical use. RESULTS: The free stent segment (FSS), the portion of the device that covers an aneurysm or branch origin, dictates whether aneurysms or branches will remain patent. Three levels of increasing complexity must be taken into account to anticipate what will occur at the FSS level. (1) Virtual models can provide basic principles; (2) in vitro studies allow testing FSS deformations that may occur in various anatomical circumstances and impact on efficacy and safety; (3) but only in vivo studies can provide key information on neointimal closure following implantation that will differentiate success from failure. CONCLUSIONS: A nomenclature is necessary to determine the optimal or suboptimal conditions for FDs and to design the virtual, in vitro and in vivo studies that will allow a better understanding of the factors involved in the success or failure of this novel treatment.