Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4533–4538 June 6–11, 2021. ©2021 Association for Computational Linguistics 4533 Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality Hyun Gi Lee1 Evan Sholle2 Ashley Beecy3 Subhi Al’Aref4 Yifan Peng1 1Department of Population Health Sciences, Weill Cornell Medicine 2Information Technologies and Services, Weill Cornell Medicine 3Department of Medicine, Weill Cornell Medicine 4Department of Internal Medicine, UAMS College of Medicine Abstract Utilizing clinical texts in survival analysis is difficult because they are largely unstructured. Current automatic extraction models fail to capture textual information comprehensively since their labels are limited in scope. Fur- thermore, they typically require a large amount of data and high-quality expert annotations for training. In this work, we present a novel method of using BERT-based hidden layer representations of clinical texts as covariates for proportional hazards models to predict pa- tient survival outcomes. We show that hid- den layers yield notably more accurate pre- dictions than predefined features, outperform- ing the previous baseline model by 5.7% on average across C-index and time-dependent AUC. We make our work publicly available at https://github.com/bionlplab/ heart_failure_mortality. 1 Introduction Survival analysis estimates the expected time until an event of interest occurs (Ranganath et al., 2016). In clinical research, it is used to understand the relationship between prognostic covariates (e.g., age and treatment) and patient survival time for important use cases such as predicting mortality of heart failure patients and providing management recommendations for intensive care units during a public health crisis like the COVID-19 pandemic (Pandey et al., 2020; Sprung et al., 2020; Nielsen et al., 2019). Clinical texts such as radiology reports contain rich information about patients that is used to diag- nose disease, plan treatments and monitor progress. It also contains the high-level reasoning of human experts that requires years of knowledge accumula- tion and professional training (Langlotz, 2015). De- spite their clinical relevance, it is challenging to use them in survival analysis since they are largely un- structured. Automatic labelers are often unable to capture detailed information to distinguish between patients, especially ones with similar conditions, because they mostly rely on a small set of manually selected labels (Lao et al., 2017). Developing meth- ods for accessing the critical information embedded in unstructured clinical texts holds the potential to meaningfully benefit clinical research. To bridge this gap, we propose a deep learning method to predict the survival probability of heart failure (HF) patients based on the high-dimensional feature representations of their radiology reports. Concretely, we extract hidden features from the texts with BERT-based (Devlin et al., 2019) mod- els and apply a recurrent neural network (RNN) to model sequences of reports and estimate the log-risk function for the overall mortality predic- tion. This approach can encapsulate more textual information than hand-crafted features and incorpo- rate higher-order temporal information from report sequences. We find that our model improves on average 5.7% in both C-index and …

Content truncated to 500 words