Cohort Bias Adaptation in Aggregated Datasets for Lesion Segmentation
This addresses the challenge of deploying machine learning models in real clinical contexts by improving generalizability across diverse patient cohorts, though it is incremental as it builds on existing normalization techniques.
The paper tackles the problem of poor generalization of lesion segmentation models to new patient cohorts due to cohort biases in aggregated datasets, and shows that their proposed Source-Conditioned Instance Normalization (SCIN) method improves performance on pooled datasets and adapts to new cohorts with only 10 labeled samples.
Many automatic machine learning models developed for focal pathology (e.g. lesions, tumours) detection and segmentation perform well, but do not generalize as well to new patient cohorts, impeding their widespread adoption into real clinical contexts. One strategy to create a more diverse, generalizable training set is to naively pool datasets from different cohorts. Surprisingly, training on this \it{big data} does not necessarily increase, and may even reduce, overall performance and model generalizability, due to the existence of cohort biases that affect label distributions. In this paper, we propose a generalized affine conditioning framework to learn and account for cohort biases across multi-source datasets, which we call Source-Conditioned Instance Normalization (SCIN). Through extensive experimentation on three different, large scale, multi-scanner, multi-centre Multiple Sclerosis (MS) clinical trial MRI datasets, we show that our cohort bias adaptation method (1) improves performance of the network on pooled datasets relative to naively pooling datasets and (2) can quickly adapt to a new cohort by fine-tuning the instance normalization parameters, thus learning the new cohort bias with only 10 labelled samples.