Semantically Redundant Training Data Removal and Deep Model Classification Performance: A Study with Chest X-rays
This addresses the issue of reduced performance and generalizability in medical imaging models due to redundant data, offering a practical improvement over conventional training practices.
The study tackled the problem of semantic redundancy in training data for deep learning models, particularly in medical imaging, by proposing an entropy-based method to remove redundant samples, resulting in improved recall scores from 0.6597 to 0.7164 on internal testing and from 0.2589 to 0.3185 on external testing.
Deep learning (DL) has demonstrated its innate capacity to independently learn hierarchical features from complex and multi-dimensional data. A common understanding is that its performance scales up with the amount of training data. Another data attribute is the inherent variety. It follows, therefore, that semantic redundancy, which is the presence of similar or repetitive information, would tend to lower performance and limit generalizability to unseen data. In medical imaging data, semantic redundancy can occur due to the presence of multiple images that have highly similar presentations for the disease of interest. Further, the common use of augmentation methods to generate variety in DL training may be limiting performance when applied to semantically redundant data. We propose an entropy-based sample scoring approach to identify and remove semantically redundant training data. We demonstrate using the publicly available NIH chest X-ray dataset that the model trained on the resulting informative subset of training data significantly outperforms the model trained on the full training set, during both internal (recall: 0.7164 vs 0.6597, p<0.05) and external testing (recall: 0.3185 vs 0.2589, p<0.05). Our findings emphasize the importance of information-oriented training sample selection as opposed to the conventional practice of using all available training data.