Related Experiment Video
Updated: Feb 8, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
Patient representation learning and interpretable evaluation using clinical notes
Madhumita Sushil1, Simon Šuster2, Kim Luyckx3
1Antwerp University Hospital, ICT Department, Wilrijkstraat 10, Edegem 2650, Belgium; Computational Linguistics and Psycholinguistics (CLiPS) Research Center, University of Antwerp, Prinsstraat 13, Antwerp 2000, Belgium.
Abstract:
We have three contributions in this work: 1. We explore the utility of a stacked denoising autoencoder and a paragraph vector model to learn task-independent dense patient representations directly from clinical notes. To analyze if these representations are transferable across tasks, we evaluate them in multiple supervised setups to predict patient mortality, primary diagnostic and procedural category, and gender. We compare their performance with sparse representations obtained from a bag-of-words model. We observe that the learned generalized representations significantly outperform the sparse representations when we have few positive instances to learn from, and there is an absence of strong lexical features. 2. We compare the model performance of the feature set constructed from a bag of words to that obtained from medical concepts. In the latter case, concepts represent problems, treatments, and tests. We find that concept identification does not improve the classification performance. 3. We propose novel techniques to facilitate model interpretability. To understand and interpret the representations, we explore the best encoded features within the patient representations obtained from the autoencoder model. Further, we calculate feature sensitivity across two networks to identify the most significant input features for different classification tasks when we use these pretrained representations as the supervised input. We successfully extract the most influential features for the pipeline using this technique.
More Related Videos
Related Concept Videos
State Space Representation
Consider an RLC circuit, a...
Graphical Representation of Inequalities
Control Volume and System Representations
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface. For instance, in the case of water...
Vector Representation of Complex Numbers
Consider a function defined as the product of the complex factors in the numerator divided by the product of the complex factors in the...
Graphical and Analytic Representation of Sinusoids
The first step is measuring the peak-to-peak value, which is twice the amplitude of the sinusoid. This provides information about the maximum voltage swing of the waveform.
Secondly, the period and angular frequency are determined. The period is the time taken for one complete cycle of the waveform, while...
Interpreting R Charts
An R chart plots the range of subsets of measurements collected from a process. Each point on the chart represents the range—defined as the difference between the maximum and minimum...

