RESUMEN
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.
Asunto(s)
Neoplasias Pancreáticas , Sarcopenia , Anciano , Inteligencia Artificial , Composición Corporal , Femenino , Humanos , Masculino , Persona de Mediana Edad , Músculo Esquelético/patología , Neoplasias Pancreáticas/diagnóstico por imagen , Neoplasias Pancreáticas/patología , Sarcopenia/patología , Tomografía Computarizada por Rayos XRESUMEN
PURPOSE: To assess the ability of the ring-enhancing sign and focal necrosis to diagnose adenosquamous carcinoma (ASqC), a variant of pancreatic ductal adenocarcinoma (PDAC), on MRI and CT. METHODS: The following features of ASqC and conventional PDAC were evaluated on CT and MRI: tumor size, location, margins, borders (non-exophytic, exophytic), and T1 signal intensity. Two readers, blinded to histopathology results, rated their confidence in detecting ring-enhancement and focal necrosis (FN) on a 5-point Likert scale on both MRI and CT. Inter-reader agreement was assessed with Cohen's kappa (k). RESULTS: A total of 24 patients were included: eight patients with treatment naïve and histologically proven ASqC (six women, mean age: 63, range: 40-75) and 16 patients with PDAC (eight women, mean age: 67, range: 47-83). Statistically significant differences between ASqC and PDAC were seen in tumor size, location, presence of FN, and ring enhancement (p = 0.01-0.037). The readers were more confident in depicting the key differentiating feature ring-enhancement in ASqC on MRI compared to CT (confidence 1.71 ± 0.49 vs. 0.88 ± 0.35, p = 0.017) with moderate inter-reader agreement (k = 0.46 and 0.5, respectively). FN showed substantial inter-reader agreement on MR and moderate agreement on CT (k = 0.67 and 0.5, respectively). CONCLUSIONS: Compared to CT, MRI depicts ring-enhancement in ASqC with greater reader confidence and FN in ASqC with higher inter-reader agreement. The concurrent presence of these two imaging features should raise high suspicion for ASqC.
Asunto(s)
Carcinoma Adenoescamoso , Carcinoma Ductal Pancreático , Neoplasias Pancreáticas , Anciano , Carcinoma Adenoescamoso/diagnóstico por imagen , Carcinoma Ductal Pancreático/diagnóstico por imagen , Femenino , Humanos , Imagen por Resonancia Magnética , Persona de Mediana Edad , Necrosis/diagnóstico por imagen , Neoplasias Pancreáticas/diagnóstico por imagen , Estudios Retrospectivos , Tomografía Computarizada por Rayos XRESUMEN
PURPOSE: This study examined the impact of the COVID-19 pandemic on emergency department CT use for acute nontraumatic abdominal pain, to better understand why imaging volume so drastically decreased during the COVID-19 pandemic. METHODS: This was a retrospective review of emergency imaging volumes from January 5 to May 30, 2020. Weekly volume data were collected for total imaging studies, abdominopelvic CT, and abdominopelvic CTs positive for common causes of acute nontraumatic abdominal pain. Two emergency radiology attendings scored all diverticulitis cases independently, and weekly volume data for uncomplicated and complicated diverticulitis cases was also collected. Volume data prior to and during the COVID-19 pandemic was compared, using 2019 volumes as a control. RESULTS: During the COVID-19 pandemic, overall emergency imaging volume decreased 30% compared to 2019 (p = 0.002). While the number of emergency abdominopelvic CTs positive for appendicitis and small bowel obstruction did not significantly change during the COVID-19 pandemic, the number of cases of diverticulitis decreased significantly compared to 2019 (p = 0.001). This reduction can be specifically attributed to decreased uncomplicated diverticulitis cases, as the number of uncomplicated diverticulitis cases dropped significantly (p = 0.002) while there was no significant difference in the number of complicated diverticulitis cases (p = 0.09). CONCLUSIONS: Reduced emergency abdominopelvic CT volume during the COVID-19 pandemic can partially be explained by decreased imaging of lower acuity patients. This data may help formulate future strategies for imaging resource utilization with an improved understanding of the relationship between perceived imaging risk and symptom acuity.
Asunto(s)
Infecciones por Coronavirus/epidemiología , Diverticulitis/diagnóstico por imagen , Servicio de Urgencia en Hospital , Neumonía Viral/epidemiología , Tomografía Computarizada por Rayos X/estadística & datos numéricos , Betacoronavirus , COVID-19 , Femenino , Humanos , Masculino , Massachusetts/epidemiología , Pandemias , Estudios Retrospectivos , SARS-CoV-2 , Revisión de Utilización de RecursosRESUMEN
Intrathoracic accessory lobes of the liver are exceedingly rare and usually found incidentally in asymptomatic patients. Its diagnosis poses a real challenge for radiologists due to its rarity, location and rounded solid mass appearance. Herein, we describe the case of a supradiaphragmatic caudate lobe of the liver in a 43-year-old African American woman presenting to the hospital for evaluation of an inferior vena cava (IVC) thrombus with CT. Final diagnosis was achieved by MRI using intravenous contrast material, showing a 4.7â¯cm by 2.7â¯cm oval shaped mass, with similar signal intensity to the main liver on all sequences. Appropriate diagnosis of this intrathoracic mass is important to negate the need for unnecessary procedures and set a proper follow up after clinical diagnosis.