Revisiting N-CNN for Clinical Practice

TL;DR

This study optimizes hyperparameters of the N-CNN for neonatal pain assessment, evaluates their effects on classification and explainability, and introduces soft labels from a facial coding system to improve neonatal pain evaluation tools.

Contribution

It presents a hyperparameter tuning approach for N-CNN and introduces soft labels from a facial coding system for neonatal pain classification.

Findings

Hyperparameter tuning improved classification metrics and explainability.

Optimizations did not significantly enhance calibration performance.

Soft labels contributed to training facial expression models.

Abstract

This paper revisits the Neonatal Convolutional Neural Network (N-CNN) by optimizing its hyperparameters and evaluating how they affect its classification metrics, explainability and reliability, discussing their potential impact in clinical practice. We have chosen hyperparameters that do not modify the original N-CNN architecture, but mainly modify its learning rate and training regularization. The optimization was done by evaluating the improvement in F1 Score for each hyperparameter individually, and the best hyperparameters were chosen to create a Tuned N-CNN. We also applied soft labels derived from the Neonatal Facial Coding System, proposing a novel approach for training facial expression classification models for neonatal pain assessment. Interestingly, while the Tuned N-CNN results point towards improvements in classification metrics and explainability, these improvements did not directly translate to calibration performance. We believe that such insights might have the potential to contribute to the development of more reliable pain evaluation tools for newborns, aiding healthcare professionals in delivering appropriate interventions and improving patient outcomes.