EAEduardo Pimenta Ribeiro Pontes Almeida

Eduardo Pimenta Ribeiro Pontes Almeida

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Eduardo Pimenta Ribeiro Pontes Almeida has a total of 83 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evaluation of a deep learning model applied to chest X-rays of patients with suspected pneumonia presenting to the emergency department designed to predict admission risk: a retrospective training and prospective non-intervention validation

    by Eduardo Pimenta Ribeiro Pontes Almeida, Sameer Masood, Felix Baldauf-Lenschen, Thomas Jay Hennessy, Shazia Akbar, Yangqing Deng, Felipe Soares Torres

    Objectives To develop, train and test a deep learning model Image-based PROgnostication applied to Chest X Rays (IPRO-X) tool that predicts the inpatient (IP) admission risk in patients with suspected pneumonia presenting to the emergency department (ED). Design The study consists of a retrospective (training) and prospective non-interventional shadow deployment (validation) of a deep learning model. Setting Three-hospital tertiary care system with two EDs. Participants Consecutive adult patients (18 years old or older) who presented to the ED from December 2022 to February 2023 with a chief clinical complaint potentially related to pneumonia were reviewed for eligibility (n=5567), with a final number of 3677 included in the validation study. Primary and secondary outcome measures IPRO-X was developed using standard two-dimensional convolutional neural network InceptionNet architecture and processes chest radiographs (CXRs) acquired at admission generating a continuous value from 0 (IP-negative) to 1 (IP-positive). We examined IPRO-X’s ability to predict IP admission using accuracy, area under the curve (AUC), receiver operating characteristic, sensitivity, specificity, positive predictive value and negative predictive value. IPRO-X scores were compared against observed outcomes (admission and discharge) to determine clinical utility in an ED setting. Four thresholds were defined from the retrospective phase: Youden optimal, highest specificity and thresholds equivalent to 30-day mortality rates of Pneumonia Severity Index (PSI) Risk class IV and Confusion, Respiratory Rate, Blood Pressure and Age 65 or older Score (CRB-65) scores 1 and 2. Performance was also analysed per group of chief complaints. Results In the validation set, 3677 patients (1777 (48%) female, median age 56 years (min 18, max 99)) were included. IPRO-X predicted IP admissions with an AUC of 0.795, and in the pneumonia-specific chief complaints, AUC increased to 0.828, whereas in non-related pneumonia chief complaints, the AUC decreased to 0.755. IPRO-X score was significantly higher in admitted patients compared with discharged patients (pConclusions IPRO-X applied to CXR of patients with signs and symptoms commonly related to pneumonia in the ED can accurately predict IP admission.

  • Automated perfusion imaging did not improve reliability of ASPECTS or collateral score in late-window large vessel occlusion stroke

    by Anass Benomar, Robert Fahed, Abdelsimar Tan Omar, Christine Hawkes, Eef J Hendriks, Clare A Enriquez, Muneer Eesa, Grant Stotts, Hubert Lee, Shashank Nagendra, et al.

    Background: The Alberta Stroke Program Early CT Score (ASPECTS) and CT angiography (CTA) Collateral Score (CS) are routinely used to estimate ischemic core burden and collateral status in acute ischemic stroke due to large vessel occlusion, but suffer from limited reliability. Whether automated CT perfusion (CTP) maps improve scoring consistency remains unclear. Objectives: To assess the reliability of these scores on non-contrast computed tomography (NCCT) and CTA among numerous raters, with and without access to automated perfusion maps, and their correlations with endovascular thrombectomy (EVT) decisions. Methods: Twenty-nine clinicians independently rated 60 imaging cases from 30 late-window stroke patients with large vessel occlusion, each presented twice: once using NCCT and CTA alone (‘non-CTP cases’), and once with the addition of perfusion maps (‘CTP cases’). Twenty-eight raters (97%) completed the survey twice. Inter- and intra-reliability was assessed using Gwet’s AC1/2 coefficients (κ G ). Correlations with EVT decisions were evaluated using Cramer V. Results: Inter-rater reliability for ASPECTS was substantial and similar between non-CTP (κ G = 0.65, 95% CI: 0.53 to 0.78) and CTP cases (κ G = 0.64, 95% CI: 0.51 to 0.77). CS showed an almost perfect reliability in both scenarios (κ G = 0.82 and 0.83). Intra-rater agreement was at least substantial. Dichotomizing ASPECTS as 0-2/3-10 improved reliability to almost perfect. ASPECTS strongly correlated with EVT decisions (V=0.80), whereas CS showed moderate correlation (V=0.56), both remaining similar with the addition of CTP (V=0.75 and 0.54, respectively). Conclusion: Automated perfusion imaging did not improve the reliability of ASPECTS or CS. Strong reliability was achieved with NCCT and CTA alone.

  • Top Cited Papers

    Environmental Sustainability and AI in Radiology: A Double-Edged Sword

    by Florence X. Doo, Jan Vosshenrich, Tessa S. Cook, Linda Moy, Eduardo P.R.P. Almeida, Sean A. Woolen, Judy Wawira Gichoya, Tobias Heye, Kate Hanneman

    According to the World Health Organization, climate change is the single biggest health threat facing humanity. The global health care system, including medical imaging, must manage the health effects of climate change while at the same time addressing the large amount of greenhouse gas (GHG) emissions generated in the delivery of care. Data centers and computational efforts are increasingly large contributors to GHG emissions in radiology. This is due to the explosive increase in big data and artificial intelligence (AI) applications that have resulted in large energy requirements for developing and deploying AI models. However, AI also has the potential to improve environmental sustainability in medical imaging. For example, use of AI can shorten MRI scan times with accelerated acquisition times, improve the scheduling efficiency of scanners, and optimize the use of decision-support tools to reduce low-value imaging. The purpose of this Radiology in Focus article is to discuss this duality at the intersection of environmental sustainability and AI in radiology. Further discussed are strategies and opportunities to decrease AI-related emissions and to leverage AI to improve sustainability in radiology, with a focus on health equity. Co-benefits of these strategies are explored, including lower cost and improved patient outcomes. Finally, knowledge gaps and areas for future research are highlighted.


  • Open Access

    Perfusion vs non-perfusion computed tomography imaging in the late window of emergent large vessel ischemic stroke: A systematic review and meta-analysis

    by Jose Danilo B Diestro, Abdelsimar T Omar, Yu-Qing Zhang, Teruko Kishibe, Alexander Mastrolonardo, Melissa Mary Lannon, Katrina Ignacio, Eduardo Pimenta Ribeiro Pontes Almeida, Anahita Malvea, Ange Diouf, et al.

    Background: Guidelines recommend the treatment of emergent large vessel ischemic stroke (ELVIS) patients presenting beyond 6 hours of last known well time with endovascular thrombectomy (EVT) based on perfusion computed tomography (CT) neuroimaging. We compared the outcomes (long-term good clinical outcomes, symptomatic intracranial hemorrhage (sICH), and mortality) of ELVIS patients according to the type of CT neuroimaging they underwent. Methods: We searched the following databases: Medline, Embase, CENTRAL, and Scopus from January 1, 2015, to June 14, 2023. We included studies of late-presenting ELVIS patients undergoing EVT that had with data for non-perfusion and perfusion CT neuroimaging. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Data were pooled using a random effects model. Results: We found 7 observational cohorts. Non-perfusion versus perfusion CT was not statistically significantly different for both long-term clinical (n = 3,224; RR: 0.96; 95% CI 0.86 to 1.06; I2 = 18%) and sICH (n = 3,724; RR: 1.08 95% CI 0.60 to 1.94; I2 = 76%). Perfusion CT had less mortality (n = 3874; RR: 1.22; 95% CI 1.07 to 1.40; I2 = 0%). The certainty of these findings is very low because of limitations in the risk of bias, indirectness, and imprecision domains of the Grading of Recommendations, Assessment, Development and Evaluations. Conclusion: The use of either non-perfusion or perfusion CT neuroimaging may have little to no effect on long-term clinical outcomes and sICH for late-presenting EVT patients. Perfusion CT neuroimaging may be associated with a reduced the risk of mortality. Evidence uncertainty warrants randomized trial data.