Integration of nested cross-validation, automated hyperparameter optimization, high-performance computing to reduce and quantify the variance of test performance estimation of deep learning models

2025-03-11

AI Summary

Reliable performance evaluation of AI models for medical imaging in real-world deployment remains challenging, primarily because single-test-set evaluation fails to quantify estimator variance, undermining clinical trustworthiness. To address this, we propose NACHOS: a novel framework integrating nested cross-validation (NCV) and Bayesian hyperparameter optimization (AHPO) in a tightly coupled architecture on high-performance computing platforms. We further introduce the DACHOS paradigm, enabling joint optimization of full-data modeling and deployment performance. Our analysis systematically demonstrates NCV’s quantifiable suppression of estimator variance. Evaluated on multi-center chest X-ray (CXR) and optical coherence tomography (OCT) datasets, NACHOS reduces AUC estimation standard deviation by 47% and improves mean deployed-model AUC by 1.8–3.2 percentage points. The entire pipeline is open-sourced, ensuring full reproducibility and compliance with clinical AI validation standards for trustworthy model assessment.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The variability and biases in the real-world performance benchmarking of deep learning models for medical imaging compromise their trustworthiness for real-world deployment. The common approach of holding out a single fixed test set fails to quantify the variance in the estimation of test performance metrics. This study introduces NACHOS (Nested and Automated Cross-validation and Hyperparameter Optimization using Supercomputing) to reduce and quantify the variance of test performance metrics of deep learning models. NACHOS integrates Nested Cross-Validation (NCV) and Automated Hyperparameter Optimization (AHPO) within a parallelized high-performance computing (HPC) framework. NACHOS was demonstrated on a chest X-ray repository and an Optical Coherence Tomography (OCT) dataset under multiple data partitioning schemes. Beyond performance estimation, DACHOS (Deployment with Automated Cross-validation and Hyperparameter Optimization using Supercomputing) is introduced to leverage AHPO and cross-validation to build the final model on the full dataset, improving expected deployment performance. The findings underscore the importance of NCV in quantifying and reducing estimation variance, AHPO in optimizing hyperparameters consistently across test folds, and HPC in ensuring computational feasibility. By integrating these methodologies, NACHOS and DACHOS provide a scalable, reproducible, and trustworthy framework for DL model evaluation and deployment in medical imaging.

Problem

• Reduces variance in deep learning model performance estimation
• Quantifies test performance metrics using advanced validation techniques
• Enhances trustworthiness of models for medical imaging deployment

Innovation

• Nested Cross-Validation reduces performance estimation variance
• Automated Hyperparameter Optimization ensures consistent model tuning
• High-performance computing enables scalable and efficient model evaluation