Corpus ID: 16979767

Clinical Tagging with Joint Probabilistic Models

@inproceedings{Halpern2016ClinicalTW,
  title={Clinical Tagging with Joint Probabilistic Models},
  author={Yoni Halpern and Steven Horng and David A. Sontag},
  booktitle={MLHC},
  year={2016}
}
We describe a method for parameter estimation in bipartite probabilistic graphical models for joint prediction of clinical conditions from the electronic medical record. The method does not rely on the availability of gold-standard labels, but rather uses noisy labels, called anchors, for learning. We provide a likelihood-based objective and a moments-based initialization that are effective at learning the model parameters. The learned model is evaluated in a task of assigning a heldout… Expand
Identifiable Phenotyping using Constrained Non-Negative Matrix Factorization
TLDR
Qualitative assessment of the empirical results by clinical experts suggests that the proposed model learns clinically interpretable phenotypes while being predictive of 30 day mortality. Expand
A latent topic model for mining heterogenous non-randomly missing electronic health records data
TLDR
MixEHR is presented, an unsupervised generative model integrating collaborative filtering and latent topic models, which jointly models the discrete distributions of data observation bias and actual data using latent disease-topic distributions. Expand
Towards deep anchor learning
TLDR
This paper proposes the deep anchor learning framework, which uses a deep autoencoder to obtain stronger features for use in anchor learning, and achieves state-of-the art performance on the task of postoperative delirium prediction from anchors. Expand
Diversity-Promoting and Large-Scale Machine Learning for Healthcare
TLDR
This thesis develops machine learning models and systems for distilling highvalue patterns from unstructured clinical data and making informed and real-timemedical predictions and recommendations, to aid physicians in improving the efficiency of workflow and the quality of patient care and proposes a rich set of optimization algorithms to solve the regularized problems. Expand
Constrained Labeling for Weakly Supervised Learning
TLDR
This paper proposes a simple data-free approach for combining weak supervision signals by defining a constrained space for the possible labels of the weak signals and training with a random labeling within this constrained space, and proves theoretical conditions under which the worst-case error of the randomized label decreases with the rank of the linear constraints. Expand
A General Framework for Adversarial Label Learning
We consider the task of training classifiers without fully labeled data. We propose a weakly supervised method—adversarial label learning—that trains classifiers to perform well when noisy andExpand
Predicting Discharge Medications at Admission Time Based on Deep Learning
TLDR
A convolutional neural network is built which takes an admission note as input and predicts the medications placed on the patient at discharge time and is able to distill semantic patterns from unstructured and noisy texts, and is capable of capturing the pharmacological correlations among medications. Expand
Deep learning for health informatics: Recent trends and future directions
TLDR
Recent trends and applications of deep learning applied to the healthcare domain are reviewed and possible future directions that could be pursued further in this domain are suggested. Expand
Facebook’s Cyber–Cyber and Cyber–Physical Digital Twins
TLDR
It is shown how the unique properties of cyber–cyber digital twins open up exciting opportunities for research and development and have practical ramifications for the design, implementation and deployment of digital twins at Facebook. Expand
...
1
2
...

References

SHOWING 1-10 OF 27 REFERENCES
Using Anchors to Estimate Clinical State without Labeled Data
TLDR
A novel framework for learning to estimate and predict clinical state variables without labeled data and develops a user interface to enable experts to choose anchor variables in an informed manner to enable real-time decision support in the emergency department. Expand
Learning statistical models of phenotypes using noisy labeled training data
TLDR
The feasibility of utilizing semi-automatically labeled training sets to create phenotype models via machine learning, using a comprehensive representation of the patient medical record is demonstrated, providing an alternative to manual labeling for creating training sets for statistical models of phenotypes. Expand
A Tractable Inference Algorithm for Diagnosing Multiple Diseases
TLDR
A probabilistic model for the diagnosis of multiple diseases, in which diseases and findings are represented as binary variables, and an algorithm for computing the posterior probability of each disease, given a set of observed findings, called quickscore is presented. Expand
Electronic medical record phenotyping using the anchor and learn framework
TLDR
A phenotype library that uses both structured and unstructured data from the EMR to represent patients for real-time clinical decision support is developed and the resulting phenotypes are interpretable and fast to build, and perform comparably to statistically learned phenotypes developed with 5000 manually labeled patients. Expand
Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base. I. The probabilistic model and inference algorithms.
TLDR
The design of a probabilistic reformulation of the Quick Medical Reference (QMR) diagnostic decision-support tool is reported and a two-level multiply connected belief-network representation of the QMR knowledge base of internal medicine is described. Expand
Anchored Discrete Factor Analysis
TLDR
A semi-supervised learning algorithm for learning discrete factor analysis models with arbitrary structure on the latent variables is presented, showing that it is possible to consistently recover moments of the latent variable and use these moments to learn complete models. Expand
A Simple Algorithm for Identifying Negated Findings and Diseases in Discharge Summaries
TLDR
It is concluded that with little implementation effort a simple regular expression algorithm for determining whether a finding or disease mentioned within narrative medical reports is present or absent can identify a large portion of the pertinent negatives from discharge summaries. Expand
Learning with Noisy Labels
TLDR
The problem of binary classification in the presence of random classification noise is theoretically studied—the learner sees labels that have independently been flipped with some small probability, and methods used in practice such as biased SVM and weighted logistic regression are provably noise-tolerant. Expand
Semi-supervised Learning with Deep Generative Models
TLDR
It is shown that deep generative models and approximate Bayesian inference exploiting recent advances in variational methods can be used to provide significant improvements, making generative approaches highly competitive for semi-supervised learning. Expand
Learning classifiers from only positive and unlabeled data
TLDR
This paper shows that models trained using the new methods perform better than the current state-of-the-art biased SVM method for learning from positive and unlabeled examples, and applies them to solve a real-world problem: identifying protein records that should be included in an incomplete specialized molecular biology database. Expand
...
1
2
3
...