Search

Introduction: The administration of intravenous (IV) contrast media during computed tomography (CT) examinations is essential to enhance diagnostic accuracy in various clinical scenarios. Traditionally, older age is considered a risk factor for the development of post-contrast Acute Kidney Injury (PC-AKI); however, there is limited information available for the super-elderly population (aged ≥ 85). This study aims to investigate the incidence and risk factors associated with PC-AKI in individuals aged 85 and older undergoing CT scans with IV contrast. Methods: A retrospective cohort study, including all hospitalized patients aged 85 or older who underwent CT scans between the years 2005 and 2021. Patients were categorized into IV contrast and non-IV contrast groups. Baseline demographic and clinical data, along with kidney function parameters, were collected. Results: The final cohort included 7,078 patients who underwent CT scans, with 40% receiving IV contrast. The overall AKI occurrence within 72 h post-CT was 5.72%, slightly elevated in the non-IV contrast group (6.25% vs. 4.94%, p = 0.02). However, multivariate analysis revealed no significant difference between the groups (OR 1, CI 0.8–1.2, p = 0.92), even after stratifying by kidney function. A secondary analysis, using a less strict AKI definition, supported these findings. Baseline creatinine levels emerged as prominent risk factor associated with PC- AKI. Conclusion: The current study provides reassurance regarding the safety of contrast-enhanced CT scans in super-elderly patients, particularly those with baseline normal to mild kidney dysfunction. These findings may contribute to the ongoing discussion on the risk–benefit balance of contrast-enhanced CT scans in the super-elderly population. [ABSTRACT FROM AUTHOR]

Purpose: Advanced imaging may augment the diagnostic milieux for presumed acute appendicitis (AA) during pregnancy, however it is not clear when such imaging modalities are indicated. The aim of this study was to assess the sensitivity and specificity of clinical scoring systems with the findings on magnetic resonance imaging (MRI) of AA in pregnant patients. Methods: A retrospective cohort study between 2019 and 2021 was performed in two tertiary level centers. Pregnant patients presenting with suspected AA and non-diagnostic trans-abdominal ultrasound who underwent MRI as part of their evaluation were identified. Patient demographics, parity, gestation, presenting signs, and symptoms were documented. The Alvarado and Appendicitis Inflammatory Response (AIR) score for each patient were calculated and correlated with clinical and MRI findings. Univariate analysis was used to identify factors associated with AA on MRI. Results: Of the 255 pregnant patients who underwent MRI, 33 (13%) had findings of AA. On univariate analysis, presentation during the second/third trimester, migration of pain, vomiting and RLQ tenderness correlated with MRI findings of AA. Whilst 5/77 (6.5%) of patients with an Alvarado score ≤4 had signs of AA on MRI, a score of ≥5 had a sensitivity, specificity, negative and positive predictive value of 84.8%, 36.6%, 94.0% and 17.2%. For an AIR score ≥ 5, this was 78.8%, 41.5%, 93.0%, and 16.7%, respectively. Conclusions: Whilst clinical scoring systems may be useful in identifying which pregnant patients require MRI to be performed when AA is suspected, the low sensitivity implies further research is needed to refine the use of this valuable resource. [ABSTRACT FROM AUTHOR]

Objectives: In the midst of the coronavirus disease 2019 (COVID-19) outbreak, chest X-ray (CXR) imaging is playing an important role in diagnosis and monitoring of patients with COVID-19. We propose a deep learning model for detection of COVID-19 from CXRs, as well as a tool for retrieving similar patients according to the model's results on their CXRs. For training and evaluating our model, we collected CXRs from inpatients hospitalized in four different hospitals. Methods: In this retrospective study, 1384 frontal CXRs, of COVID-19 confirmed patients imaged between March and August 2020, and 1024 matching CXRs of non-COVID patients imaged before the pandemic, were collected and used to build a deep learning classifier for detecting patients positive for COVID-19. The classifier consists of an ensemble of pre-trained deep neural networks (DNNS), specifically, ReNet34, ReNet50¸ ReNet152, and vgg16, and is enhanced by data augmentation and lung segmentation. We further implemented a nearest-neighbors algorithm that uses DNN-based image embeddings to retrieve the images most similar to a given image. Results: Our model achieved accuracy of 90.3%, (95% CI: 86.3–93.7%) specificity of 90% (95% CI: 84.3–94%), and sensitivity of 90.5% (95% CI: 85–94%) on a test dataset comprising 15% (350/2326) of the original images. The AUC of the ROC curve is 0.96 (95% CI: 0.93–0.97). Conclusion: We provide deep learning models, trained and evaluated on CXRs that can assist medical efforts and reduce medical staff workload in handling COVID-19. Key Points: • A machine learning model was able to detect chest X-ray (CXR) images of patients tested positive for COVID-19 with accuracy and detection rate above 90%. • A tool was created for finding existing CXR images with imaging characteristics most similar to a given CXR, according to the model's image embeddings. [ABSTRACT FROM AUTHOR]