This study evaluated the performance of automated machine-learning to diagnose non-alcoholic fatty liver disease (NAFLD) by ultrasound and compared these findings to radiologist performance.Methods: 96 patients with histologic (33) or proton density fat fraction MRI (63) diagnosis of NAFLD and 100 patients without evidence of NAFLD were retrospectively identified. The "Fatty Liver" label included 96 patients with 405 images and the "Not Fatty Liver" label included 100 patients with 500 images. These 905 images made up a "Comprehensive Image" group.A "Radiology Selected Image" group was then created by selecting only images considered diagnostic by a blinded radiologist, resulting in 649 images. Cloud AutoML Visionbeta (Google LLC, Mountain View, CA) was used for machine learning. The models were evaluated against three blinded radiologists. Results:The "Comprehensive Image" group model demonstrated a sensitivity of 88.6% (73.3-96.8%) and a specificity of 95.3% (84.2-99.4%). Radiologist performance on this image group included a sensitivity of 81.0% (74.3-87.6%) and specificity of 86.0% (72.6-99.5%). The model's overall accuracy was 92.3% (84.0-97.1%), compared with mean individual performance (83.8%, 78.4-89.1%). The "Radiology Selected Image" group model demonstrated a sensitivity of 88.6% (73.3 -96.8%) and specificity of 87.9% (71.8-96.6%). Mean radiologist sensitivity was 92.4% (86.9-97.9%) and specificity was 91.9% (83.4-100%). The model's overall accuracy was 88.2% (78.1-94.8%) which was comparable to the individual radiologist performance (92.2%, 90.1-94.2%) and consensus performance (95.6%, 87.6-99.1%).Conclusions: An automated machine-learning algorithm may accurately detect NAFLD on ultrasound.
We report a rare case of diffuse replacement of the pancreas with neuroendocrine tumour mimicking chronic pancreatitis. A 55-year-old female with no significant past medical history initially presented with abdominal pain in 2006. A CT of the abdomen and pelvis was performed, revealing diffuse pancreatic parenchymal calcifications with mild pancreatic ductal dilatation and no discrete mass. She was diagnosed with chronic pancreatitis and followed clinically until 2015, where she presented with recurrent abdominal pain. A repeat CT and MRI of the abdomen were performed which revealed new hypoenhancing masses within the pancreas, particularly in the pancreatic tail. There was a persistent background of pancreatic parenchymal calcifications. The possibility of pancreatic neuroendocrine tumour was raised, and an indium-111 Octreotide scan was recommended. Diffuse intense uptake was identified throughout the pancreas on the indium-111 imaging. Given the concern for neuroendocrine tumour, a total pancreatectomy was performed, with histopathology revealing replacement of the pancreas with coalescing well-circumscribed nodules. Many of the nodules had numerous calcifications and localized amyloid deposition. Immunohistochemical stains of the neoplastic cells were strong for neuroendocrine markers chromogranin A and synaptophysin. Overall the findings were consistent with numerous neuroendocrine tumours of the pancreas, Grade II, as per the 2010 WHO criteria for neuroendocrine tumours of the pancreas. Neuroendocrine tumours of the pancreas are lesions that arise from the islet cells, with an approximate incidence of five cases per million people per year. Only one other case report has been documented in the literature by Singh et al demonstrating diffuse pancreatic neuroendocrine tumour replacing the entire pancreas. As diffuse pancreatic neuroendocrine tumour can look similar on imaging to chronic pancreatitis or other infiltrative processes, we wanted to present this case and some of the more specific imaging findings in distinguishing these entities.
The purpose of this pictorial essay is to review different etiologies for lower extremity pain encountered on lower extremity venous sonography including acute deep venous thrombosis, chronic postthrombotic change, central venous disease, common arterial pathologies, and nonvascular abnormalities.
BackgroundFocal nodular hyperplasia (FNH) and hepatic adenoma (HA) are two common benign liver lesions with different management options. In particular, resection is considered for large HA lesions to avoid possible bleeding complications or rarely malignant degeneration.PurposeTo determine whether early enhancement of a draining hepatic vein (EDHV) and absence of perilesional enhancement (PLE) on arterial phase MR images are useful for distinguishing FNH from HA.Study TypeRetrospective.PopulationA total of 34 patients: 16 with FNH and 18 with HA lesions.Field Strength/SequenceA1.5 T, axial T1 fat‐suppressed arterial postcontrast.AssesmentFour abdominal radiologists blinded to pathologic diagnosis assessed for the presence or absence of EDHV in association with the lesion, definitively characterized by pathology. This was considered present if contrast could be identified in a hepatic vein contiguous with the lesion in question. Secondarily, PLE was evaluated.Statistical TestsFleiss's multirater kappa statistic, Chi‐squared statistic, Phi‐coefficient. Significance level P < 0.05.ResultsConsidering all observations obtained from the four readers, an EDHV was identified with FNH 48.5% of the time. EDHV was seen with HA in 8.8% of cases. PLE was seen with significantly greater frequency in HA. The presence of an EDHV was associated with the absence of PLE.Data ConclusionIn a lesion that may be either an FNH or HA, confident identification on arterial phase images of an EDHV should lead the reader to favor FNH, while the presence PLE should dissuade the reader from FNH.Evidence Level4.Technical EfficacyStage 2.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.