Picture for Giovanni Cinà

Giovanni Cinà

Pacmed

Mitigating Overconfidence in Out-of-Distribution Detection by Capturing Extreme Activations

Add code
May 21, 2024
Viaarxiv icon

Causal prediction models for medication safety monitoring: The diagnosis of vancomycin-induced acute kidney injury

Add code
Nov 15, 2023
Viaarxiv icon

Fixing confirmation bias in feature attribution methods via semantic match

Add code
Jul 03, 2023
Viaarxiv icon

Semantic match: Debugging feature attribution methods in XAI for healthcare

Add code
Jan 06, 2023
Viaarxiv icon

Why we do need Explainable AI for Healthcare

Add code
Jun 30, 2022
Figure 1 for Why we do need Explainable AI for Healthcare
Figure 2 for Why we do need Explainable AI for Healthcare
Viaarxiv icon

Out-of-Distribution Detection for Medical Applications: Guidelines for Practical Evaluation

Add code
Sep 30, 2021
Figure 1 for Out-of-Distribution Detection for Medical Applications: Guidelines for Practical Evaluation
Figure 2 for Out-of-Distribution Detection for Medical Applications: Guidelines for Practical Evaluation
Figure 3 for Out-of-Distribution Detection for Medical Applications: Guidelines for Practical Evaluation
Figure 4 for Out-of-Distribution Detection for Medical Applications: Guidelines for Practical Evaluation
Viaarxiv icon

A pragmatic approach to estimating average treatment effects from EHR data: the effect of prone positioning on mechanically ventilated COVID-19 patients

Add code
Sep 14, 2021
Figure 1 for A pragmatic approach to estimating average treatment effects from EHR data: the effect of prone positioning on mechanically ventilated COVID-19 patients
Figure 2 for A pragmatic approach to estimating average treatment effects from EHR data: the effect of prone positioning on mechanically ventilated COVID-19 patients
Figure 3 for A pragmatic approach to estimating average treatment effects from EHR data: the effect of prone positioning on mechanically ventilated COVID-19 patients
Figure 4 for A pragmatic approach to estimating average treatment effects from EHR data: the effect of prone positioning on mechanically ventilated COVID-19 patients
Viaarxiv icon

Know Your Limits: Monotonicity & Softmax Make Neural Classifiers Overconfident on OOD Data

Add code
Dec 11, 2020
Figure 1 for Know Your Limits: Monotonicity & Softmax Make Neural Classifiers Overconfident on OOD Data
Figure 2 for Know Your Limits: Monotonicity & Softmax Make Neural Classifiers Overconfident on OOD Data
Viaarxiv icon

Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data

Add code
Nov 06, 2020
Figure 1 for Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data
Figure 2 for Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data
Figure 3 for Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data
Figure 4 for Trust Issues: Uncertainty Estimation Does Not Enable Reliable OOD Detection On Medical Tabular Data
Viaarxiv icon

Uncertainty estimation for classification and risk prediction in medical settings

Add code
Apr 13, 2020
Figure 1 for Uncertainty estimation for classification and risk prediction in medical settings
Figure 2 for Uncertainty estimation for classification and risk prediction in medical settings
Figure 3 for Uncertainty estimation for classification and risk prediction in medical settings
Figure 4 for Uncertainty estimation for classification and risk prediction in medical settings
Viaarxiv icon