Selecting cardiac magnetic resonance images suitable for annotation of pulmonary arteries using an active-learning based deep learning model

An increasing and aging patient population poses a growing burden on healthcare professionals. Automation of medical imaging diagnostics holds promise for enhancing patient care and reducing manpower required to accommodate an increasing patient-population. Deep learning, a subset of machine learning, has the potential to facilitate automated diagnostics, but commonly requires large-scaled labeled datasets. In medical domains, data is often abundant but labeling is a laborious and costly task. Active learning provides a method to optimize the selection of unlabeled samples that are most suitable for improvement of the model and incorporate them into the model training process. This approach proves beneficial when only a small number of labeled samples are available. Various selection methods currently exist, but most of them employ fixed querying schedules. There is limited research on how the timing of a query can impact performance in relation to the number of queried samples. This paper proposes a novel approach called dynamic querying, which aims to optimize the timing of queries to enhance model development while utilizing as few labeled images as possible. The performance of the proposed model is compared to a model trained utilizing a fully-supervised training method, and its effectiveness is assessed based on dataset size requirements and loss rates. Dynamic querying demonstrates a considerably faster learning curve in relation to the number of labeled samples used, achieving an accuracy of 70% using only 24 samples, compared to 82% for a fully-supervised model trained on the complete training dataset of 1017 images.

Automated summary

The increasing and aging patient population presents challenges for healthcare professionals. Automation of medical imaging diagnostics using deep learning has the potential to improve patient care and reduce manpower requirements. However, labeled datasets are often laborious and costly to obtain in the medical field. Active learning, particularly dynamic querying, provides a method to optimize the selection of unlabeled samples for model improvement while minimizing the use of labeled images. This approach shows a faster learning curve and lower dataset size requirements compared to fully-supervised training methods.