Browsing by Author "Coelho, Raquel"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Benchmarking computer-vision-based facial emotion classification algorithms while wearing surgical masksPublication . Coelho, Luis; Reis, Sara; Moreira, Cristina; Cardoso, Helena; Sequeira, Miguela; Coelho, RaquelEffective human communication relies heavily on emotions, making them a crucial aspect of interaction. As technology progresses, the desire for machines to exhibit more human-like characteristics, including emotion recognition, grows. DeepFace has emerged as a widely adopted library for facial emotion recognition. However, the widespread use of surgical masks after the COVID-19 pandemic presents a considerable obstacle to its performance. To assess this issue, we conducted a benchmark using the FER2013 dataset. The results revealed a substantial performance decline when individuals wore surgical masks. “Disgust” suffers a 22.6% F1-score reduction, while “Surprise” is least affected with a 48.7% reduction. Addressing these issues improves human–machine interfaces and paves the way for more natural machine communication.
- The effectiveness of NIRS technology to the early diagnosis of lower limb ischemia in patients on peripheral VA ECMO: A systematic review and meta-analysisPublication . Coelho, Raquel; Tavares, Joana; Marinheiro, Catarina; Costa, Carina; Ferreira, Simão; Gregório, Tiago; Ferreira, SimãoAcute lower limb ischemia is a major complication of peripheral venoarterial ECMO, significantly impacting patient outcomes and survival rates. Traditional methods for assessing limb perfusion, such as physical exams and Doppler ultrasound, are often unreliable and do not provide continuous monitoring. Near-infrared spectroscopy (NIRS), a non-invasive technique, shows promise for perfusion monitoring in venoarterial ECMO patients, but its effectiveness in the early detection of limb hypoperfusion remains unreviewed. Evaluate the effectiveness of NIRS technology in the early diagnosis of lower limb ischemia in patients undergoing peripheral VA ECMO. The search strategy covered five databases. Inclusion criteria included studies in Portuguese, English, Spanish, or German involving participants aged 18 or older dependent on peripheral VA ECMO. The intervention assessed was limb perfusion monitoring using NIRS in VA ECMO patients. The primary outcome was the effectiveness of NIRS in the early diagnosis of limb ischemia. Exclusion criteria included review articles, book chapters, books, editorials, conference papers, and studies on pediatric patients, central VA ECMO, or venovenous ECMO. Study quality was evaluated using the ROBINS-I tool. Meta-analysis was performed using R package meta. Narrative synthesis was applied when meta-analysis was unfeasible. Of 180 studies, 164 were excluded after initial screening. Of the remaining 16 studies, eight were removed for irrelevance, high bias risk, or pediatric focus, leaving eight studies. The results revealed a pooled sensitivity of the diagnostic method of 0.71 (95% CI: [0.67, 0.74]) and a pooled specificity of 0.68 (95% CI: [0.61, 0.74]). NIRS technology is an effective diagnostic tool for reliably detecting true positive cases of limb ischemia.