Pablo Ferri

TL;DR: The authors applied machine learning to arsenic species and metallomics profiles of toenails to evaluate associations of environmental arsenic with incident cancer cases, user satisfaction with an AI system for chest X-ray analysis implemented in a hospital’s emergency setting; scaling AI projects for radiology causes and consequences; ECG classification using combination of linear and non-linear features with neural network;dataset comparison tool: utility and privacy; when context matters for credible measurement of drug-drug interactions based on real-world data; a lightweight and interpretable model to classify bundle branch blocks from ECG signals; analysis of stroke assistance in Covid-19 pandemic by process mining techniques; automatic diagnosis of autism spectrum disorder condition using shape based features extracted from brainstem; using explainable supervised machine learning, and an image based object recognition system for wound detection and classification of diabetic foot and venous leg ulcers.

TL;DR: The results suggest that in the presence of highly missing data, combining translation and encoding imputation-which considers informative missingness-with tree ensemble classifiers-random forest and gradient boosting-is a sensible choice when aiming to maximize performance, in terms of area under curve.

TL;DR: The model captures information present in emergency medical calls not considered by the existing in-house triage protocol, but relevant to carry out incident classification, and the results suggest that most of this information is present in the free text dispatcher observations.

TL;DR: This study develops and evaluates a deep classifier for prioritizing Emergency Medical Call Incidents under dataset shifts, utilizing a large dataset from the Health Services Department of Valencia, Spain, and demonstrates the effectiveness of Continual Learning techniques in adapting to changing data distributions.