Publications by authors named "Dalia Rodriguez-Salas"
Acute ischemic stroke (AIS) is a leading global cause of mortality and morbidity. Improving long-term outcome predictions after thrombectomy can enhance treatment quality by supporting clinical decision-making. With the advent of interpretable deep learning methods in recent years, it is now possible to develop trustworthy, high-performing prediction models.
View Article and Find Full Text PDFArticle Synopsis
- The study critiques the use of AU-ROC as a sole metric for evaluating deep-learning systems, highlighting its limitations in reflecting real-world performance, especially in anomaly detection tasks.
- Traditional methods to address class imbalance in training datasets may not effectively optimize for specific operational contexts, leading to inconsistent performance even with the same AU-ROC values.
- The authors propose a new technique, AUCReshaping, which focuses on improving sensitivity within a defined specificity range, demonstrating significant improvements in detection tasks like Chest X-Ray analysis and credit card fraud detection.