Self-explaining Hierarchical Model for Intraoperative Time Series
- URL: http://arxiv.org/abs/2210.04417v1
- Date: Mon, 10 Oct 2022 03:24:18 GMT
- Title: Self-explaining Hierarchical Model for Intraoperative Time Series
- Authors: Dingwen Li, Bing Xue, Christopher King, Bradley Fritz, Michael Avidan,
Joanna Abraham, Chenyang Lu
- Abstract summary: We propose a hierarchical model combining the strength of both attention and recurrent models for intraoperative time series.
Experiments on a large dataset of 111,888 surgeries with multiple outcomes and an external high-resolution ICU dataset show that our model can achieve strong predictive performance.
- Score: 6.877686657275981
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Major postoperative complications are devastating to surgical patients. Some
of these complications are potentially preventable via early predictions based
on intraoperative data. However, intraoperative data comprise long and
fine-grained multivariate time series, prohibiting the effective learning of
accurate models. The large gaps associated with clinical events and protocols
are usually ignored. Moreover, deep models generally lack transparency.
Nevertheless, the interpretability is crucial to assist clinicians in planning
for and delivering postoperative care and timely interventions. Towards this
end, we propose a hierarchical model combining the strength of both attention
and recurrent models for intraoperative time series. We further develop an
explanation module for the hierarchical model to interpret the predictions by
providing contributions of intraoperative data in a fine-grained manner.
Experiments on a large dataset of 111,888 surgeries with multiple outcomes and
an external high-resolution ICU dataset show that our model can achieve strong
predictive performance (i.e., high accuracy) and offer robust interpretations
(i.e., high transparency) for predicted outcomes based on intraoperative time
series.
Related papers
- Deep State-Space Generative Model For Correlated Time-to-Event Predictions [54.3637600983898]
We propose a deep latent state-space generative model to capture the interactions among different types of correlated clinical events.
Our method also uncovers meaningful insights about the latent correlations among mortality and different types of organ failures.
arXiv Detail & Related papers (2024-07-28T02:42:36Z) - Feature importance to explain multimodal prediction models. A clinical use case [2.2527696312643637]
Surgery to treat elderly hip fracture patients may cause complications that can lead to early mortality.
We develop a multimodal deep-learning model for post-operative mortality prediction using pre-operative and per-operative data.
arXiv Detail & Related papers (2024-04-29T12:11:26Z) - Prediction of Post-Operative Renal and Pulmonary Complications Using
Transformers [69.81176740997175]
We evaluate the performance of transformer-based models in predicting postoperative acute renal failure, pulmonary complications, and postoperative in-hospital mortality.
Our results demonstrate that transformer-based models can achieve superior performance in predicting postoperative complications and outperform traditional machine learning models.
arXiv Detail & Related papers (2023-06-01T14:08:05Z) - DeepJoint: Robust Survival Modelling Under Clinical Presence Shift [2.9745607433320926]
We propose a recurrent neural network which models three clinical presence dimensions in parallel to the survival outcome.
On a prediction task, explicit modelling of these three processes showed improved performance in comparison to state-of-the-art predictive models.
arXiv Detail & Related papers (2022-05-26T16:42:38Z) - Disentangled Counterfactual Recurrent Networks for Treatment Effect
Inference over Time [71.30985926640659]
We introduce the Disentangled Counterfactual Recurrent Network (DCRN), a sequence-to-sequence architecture that estimates treatment outcomes over time.
With an architecture that is completely inspired by the causal structure of treatment influence over time, we advance forecast accuracy and disease understanding.
We demonstrate that DCRN outperforms current state-of-the-art methods in forecasting treatment responses, on both real and simulated data.
arXiv Detail & Related papers (2021-12-07T16:40:28Z) - Model-Attentive Ensemble Learning for Sequence Modeling [86.4785354333566]
We present Model-Attentive Ensemble learning for Sequence modeling (MAES)
MAES is a mixture of time-series experts which leverages an attention-based gating mechanism to specialize the experts on different sequence dynamics and adaptively weight their predictions.
We demonstrate that MAES significantly out-performs popular sequence models on datasets subject to temporal shift.
arXiv Detail & Related papers (2021-02-23T05:23:35Z) - DeepRite: Deep Recurrent Inverse TreatmEnt Weighting for Adjusting
Time-varying Confounding in Modern Longitudinal Observational Data [68.29870617697532]
We propose Deep Recurrent Inverse TreatmEnt weighting (DeepRite) for time-varying confounding in longitudinal data.
DeepRite is shown to recover the ground truth from synthetic data, and estimate unbiased treatment effects from real data.
arXiv Detail & Related papers (2020-10-28T15:05:08Z) - Temporal Phenotyping using Deep Predictive Clustering of Disease
Progression [97.88605060346455]
We develop a deep learning approach for clustering time-series data, where each cluster comprises patients who share similar future outcomes of interest.
Experiments on two real-world datasets show that our model achieves superior clustering performance over state-of-the-art benchmarks.
arXiv Detail & Related papers (2020-06-15T20:48:43Z) - Dynamic Predictions of Postoperative Complications from Explainable,
Uncertainty-Aware, and Multi-Task Deep Neural Networks [1.7548541038532495]
Multi-task deep learning models outperform random forest models in predicting postoperative complications.
integrated interpretability mechanisms identified potentially modifiable risk factors for each complication.
Monte Carlo dropout methods provided a quantitative measure of prediction uncertainty that has the potential to enhance clinical trust.
arXiv Detail & Related papers (2020-04-27T02:32:32Z) - Adaptive Prediction Timing for Electronic Health Records [3.308743964406688]
We introduce a novel, more realistic, approach to generating patient outcome predictions at an adaptive rate.
We use a Recurrent Neural Network (RNN) and a Bayesian embedding layer with a new aggregation method to demonstrate adaptive prediction timing.
At 48 hours after patient admission, our model achieves equal performance compared to its static-windowed counterparts.
arXiv Detail & Related papers (2020-03-05T12:02:44Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.