Interpretable Representation Learning for Healthcare via Capturing Disease Progression through Time

et al. 2020

AAAI

Deep learning-based health status representation learning and clinical prediction have raised much research interest in recent years. Existing models have shown superior performance, but there are still several major issues that have not been fully taken into consideration. First, the historical variation pattern of the biomarker in diverse time scales plays a vital role in indicating the health status, but it has not been explicitly extracted by existing works. Second, key factors that strongly indicate the health risk are different among patients. It is still challenging to adaptively make use of the features for patients in diverse conditions. Third, using prediction models as the black box will limit the reliability in clinical practice. However, none of the existing works can provide satisfying interpretability and meanwhile achieve high prediction performance. In this work, we develop a general health status representation learning model, named AdaCare. It can capture the long and short-term variations of biomarkers as clinical features to depict the health status in multiple time scales. It also models the correlation between clinical features to enhance the ones which strongly indicate the health status and thus can maintain a state-of-the-art performance in terms of prediction accuracy while providing qualitative interpretability. We conduct a health risk prediction experiment on two real-world datasets. Experiment results indicate that AdaCare outperforms state-of-the-art approaches and provides effective interpretability, which is verifiable by clinical experts.

Section: Interpretability Of Emr Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

AdaCare: Explainable Clinical Health Status Representation Learning via Scale-Adaptive Feature Extraction and Recalibration

Gao²,

et al. 2020

AAAI

“…Zheng et al balance the inheritance and update of hidden states based on the time decay function when updating the hidden layer state of GRU [11]. Bai et al propose the Timeline model to model the decay rate of different events affecting patients [9]. These efforts use time decay factors to solve inconsistencies in clinical event sequences but do not consider the short-term out-of-order in clinical event sequence in EHR data.…”

Section: B Deep Sequential Models For Ehrmentioning

confidence: 99%

Predictive Multi-level Patient Representations from Electronic Health Records

2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

Liu

et al. 2019

The advent of the Internet era has led to an explosive growth in the Electronic Health Records (EHR) in the past decades. The EHR data can be regarded as a collection of clinical events, including laboratory results, medication records, physiological indicators, etc, which can be used for clinical outcome prediction tasks to support constructions of intelligent health systems. Learning patient representation from these clinical events for the clinical outcome prediction is an important but challenging step. Most related studies transform EHR data of a patient into a sequence of clinical events in temporal order and then use sequential models to learn patient representations for outcome prediction. However, clinical event sequence contains thousands of event types and temporal dependencies. We further make an observation that clinical events occurring in a short period are not constrained by any temporal order but events in a long term are influenced by temporal dependencies. The multi-scale temporal property makes it difficult for traditional sequential models to capture the short-term co-occurrence and the long-term temporal dependencies in clinical event sequences. In response to the above challenges, this paper proposes a Multilevel Representation Model (MRM). MRM first uses a sparse attention mechanism to model the short-term co-occurrence, then uses interval-based event pooling to remove redundant information and reduce sequence length and finally predicts clinical outcomes through Long Short-Term Memory (LSTM). Experiments on real-world datasets indicate that our proposed model largely improves the performance of clinical outcome prediction tasks using EHR data.

“…Thus, considering the particular condition of the patient, the way of attending the medical features in the whole prediction process should be individualized. Some existing works try to model the relationship between clinical visits (Ma et al 2017), dynamic features (Bai et al 2018;Choi et al 2016) or incorporate the static information (Lee et al 2018). However, none of the existing models explored the interdependencies among dynamic records as well as static baseline information via a global view.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, medical records are produced irregularly in clinical practice. It is assumed by many existing works (Pham et al 2016;Baytas et al 2017;Ma, Xiao, and Wang 2018;Bai et al 2018) that the more recent clinical records weight more than previous records in general on the healthcare prediction. However, under certain circumstance, historical records also contain valuable clinical information, which may not be revealed in the latest record (e.g., the blood glucose level was extremely abnormal).…”

Section: Introductionmentioning

confidence: 99%

ConCare: Personalized Clinical Feature Embedding via Capturing the Healthcare Context

Zhang

et al. 2020

AAAI

104

Predicting the patient's clinical outcome from the historical electronic medical records (EMR) is a fundamental research problem in medical informatics. Most deep learning-based solutions for EMR analysis concentrate on learning the clinical visit embedding and exploring the relations between visits. Although those works have shown superior performances in healthcare prediction, they fail to explore the personal characteristics during the clinical visits thoroughly. Moreover, existing works usually assume that the more recent record weights more in the prediction, but this assumption is not suitable for all conditions. In this paper, we propose ConCare to handle the irregular EMR data and extract feature interrelationship to perform individualized healthcare prediction. Our solution can embed the feature sequences separately by modeling the time-aware distribution. ConCare further improves the multi-head self-attention via the cross-head decorrelation, so that the inter-dependencies among dynamic features and static baseline information can be effectively captured to form the personal health context. Experimental results on two real-world EMR datasets demonstrate the effectiveness of ConCare. The medical findings extracted by ConCare are also empirically confirmed by human experts and medical literature.