Longitudinal detection of radiological abnormalities with time-modulated LSTM
This work addresses the challenge of irregularly sampled medical data for clinicians, though it is incremental as it modifies an existing LSTM architecture.
The study tackled the problem of detecting radiological abnormalities in medical images by modeling sequences of radiographs with time-modulated LSTMs, resulting in improved classification performance for abnormalities like cardiomegaly and pleural effusion.
Convolutional neural networks (CNNs) have been successfully employed in recent years for the detection of radiological abnormalities in medical images such as plain x-rays. To date, most studies use CNNs on individual examinations in isolation and discard previously available clinical information. In this study we set out to explore whether Long-Short-Term-Memory networks (LSTMs) can be used to improve classification performance when modelling the entire sequence of radiographs that may be available for a given patient, including their reports. A limitation of traditional LSTMs, though, is that they implicitly assume equally-spaced observations, whereas the radiological exams are event-based, and therefore irregularly sampled. Using both a simulated dataset and a large-scale chest x-ray dataset, we demonstrate that a simple modification of the LSTM architecture, which explicitly takes into account the time lag between consecutive observations, can boost classification performance. Our empirical results demonstrate improved detection of commonly reported abnormalities on chest x-rays such as cardiomegaly, consolidation, pleural effusion and hiatus hernia.