- title: 'Input-Output Non-Linear Dynamical Systems applied to Physiological Condition Monitoring' abstract: 'We present a non-linear dynamical system for modelling the effect of drug infusions on the vital signs of patients admitted in Intensive Care Units (ICUs). More specifically we are interested in modelling the effect of a widely used anaesthetic drug (Propofol) on a patient’s monitored depth of anaesthesia and haemodynamics. We compare our approach with one from the Pharmacokinetics/Pharmacodynamics (PK/PD) literature and show that we can provide significant improvements in performance without requiring the incorporation of expert physiological knowledge in our system.' volume: 56 URL: https://proceedings.mlr.press/v56/Georgatzis16.html PDF: http://proceedings.mlr.press/v56/Georgatzis16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Georgatzis16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Konstantinos family: Georgatzis - given: Chris family: Williams - given: Christopher family: Hawthorne editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 1-16 id: Georgatzis16 issued: date-parts: - 2016 - 12 - 10 firstpage: 1 lastpage: 16 published: 2016-12-10 00:00:00 +0000 - title: 'Identifiable Phenotyping using Constrained Non-Negative Matrix Factorization' abstract: 'This work proposes a new algorithm for automated and simultaneous phenotyping of multiple co-occurring medical conditions, also referred to as comorbidities, using clinical notes from electronic health records (EHRs). A latent factor estimation technique, non-negative matrix factorization (NMF), is augmented with domain constraints from weak supervision to obtain sparse latent factors that are grounded to a fixed set of chronic conditions. The proposed grounding mechanism ensures a one-to-one identifiable and interpretable mapping between the latent factors and the target comorbidities. Qualitative assessment of the empirical results by clinical experts show that the proposed model learns clinically interpretable phenotypes which are also shown to have competitive performance on 30 day mortality prediction task. The proposed method can be readily adapted to any non-negative EHR data across various healthcare institutions.' volume: 56 URL: https://proceedings.mlr.press/v56/Joshi16.html PDF: http://proceedings.mlr.press/v56/Joshi16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Joshi16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Shalmali family: Joshi - given: Suriya family: Gunasekar - given: David family: Sontag - given: Ghosh family: Joydeep editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 17-41 id: Joshi16 issued: date-parts: - 2016 - 12 - 10 firstpage: 17 lastpage: 41 published: 2016-12-10 00:00:00 +0000 - title: 'Predicting Disease Progression with a Model for Multivariate Longitudinal Clinical Data' abstract: 'Accurate prediction of the future trajectory of a disease is an important challenge in personalized medicine and population health management. However, many complex chronic diseases exhibit large degrees of heterogeneity, and furthermore there is not always a single readily available biomarker to quantify disease severity. Even when such a clinical variable exists, there are often additional related biomarkers that may help improve prediction of future disease state. To this end, we propose a novel probabilistic generative model for multivariate longitudinal data that captures dependencies between multivariate trajectories of clinical variables. We use a Gaussian process based regression model for each individual trajectory, and build off ideas from latent class models to induce dependence between their mean functions. We develop a scalable variational inference algorithm that we use to fit our model to a large dataset of longitudinal electronic patient health records. Our model’s dynamic predictions have significantly lower errors compared to a recent state of the art method for modeling disease trajectories, and they are being incorporated into a population health rounding tool to be used by clinicians at our local accountable care organization.' volume: 56 URL: https://proceedings.mlr.press/v56/Futoma16.html PDF: http://proceedings.mlr.press/v56/Futoma16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Futoma16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Joseph family: Futoma - given: Mark family: Sendak - given: Blake family: Cameron - given: Katherine family: Heller editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 42-54 id: Futoma16 issued: date-parts: - 2016 - 12 - 10 firstpage: 42 lastpage: 54 published: 2016-12-10 00:00:00 +0000 - title: 'Using Kernel Methods and Model Selection for Prediction of Preterm Birth' abstract: 'We describe an application of machine learning to the problem of predicting preterm birth. We conduct a secondary analysis on a clinical trial dataset collected by the National Institute of Child Health and Human Development (NICHD) while focusing our attention on predicting different classes of preterm birth. We compare three approaches for deriving predictive models: a support vector machine (SVM) approach with linear and non-linear kernels, logistic regression with different model selection along with a model based on decision rules prescribed by physician experts for prediction of preterm birth. Our approach highlights the pre-processing methods applied to handle the inherent dynamics, noise and gaps in the data and describe techniques used to handle skewed class distributions. Empirical experiments demonstrate significant improvement in predicting preterm birth compared to past work.' volume: 56 URL: https://proceedings.mlr.press/v56/Vovsha16.html PDF: http://proceedings.mlr.press/v56/Vovsha16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Vovsha16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Ilia family: Vovsha - given: Ansaf family: Salleb-Aouissi - given: Anita family: Raja - given: Thomas family: Koch - given: Alex family: Rybchuk - given: Axinia family: Radeva - given: Ashwath family: Rajan - given: Yiwen family: Huang - given: Hatim family: Diab - given: Ashish family: Tomar - given: Ronald family: Wapner editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 55-72 id: Vovsha16 issued: date-parts: - 2016 - 12 - 10 firstpage: 55 lastpage: 72 published: 2016-12-10 00:00:00 +0000 - title: 'Multi-task Prediction of Disease Onsets from Longitudinal Laboratory Tests' abstract: 'Disparate areas of machine learning have benefited from models that can take raw data with little preprocessing as input and learn rich representations of that raw data in order to perform well on a given prediction task. We evaluate this approach in healthcare by using longitudinal measurements of lab tests, one of the more raw signals of a patient’s health state widely available in clinical data, to predict disease onsets. In particular, we train a Long Short-Term Memory (LSTM) recurrent neural network and two novel convolutional neural networks for multi-task prediction of disease onset for 133 conditions based on 18 common lab tests measured over time in a cohort of 298K patients derived from 8 years of administrative claims data. We compare the neural networks to a logistic regression with several hand-engineered, clinically relevant features. We find that the representation-based learning approaches significantly outperform this baseline. We believe that our work suggests a new avenue for patient risk stratification based solely on lab results.' volume: 56 URL: https://proceedings.mlr.press/v56/Razavian16.html PDF: http://proceedings.mlr.press/v56/Razavian16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Razavian16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Narges family: Razavian - given: Jake family: Marcus - given: David family: Sontag editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 73-100 id: Razavian16 issued: date-parts: - 2016 - 12 - 10 firstpage: 73 lastpage: 100 published: 2016-12-10 00:00:00 +0000 - title: 'Deep Survival Analysis' abstract: 'The electronic health record (EHR) provides an unprecedented opportunity to build actionable tools to support physicians at the point of care. In this paper, we introduce deep survival analysis, a hierarchical generative approach to survival analysis in the context of the EHR. It departs from previous approaches in two main ways: (1) all observations, including covariates, are modeled jointly conditioned on a rich latent structure; and (2) the observations are aligned by their failure time, rather than by an arbitrary time zero as in traditional survival analysis. Further, it handles heterogeneous data types that occur in the EHR. We validate deep survival analysis by stratifying patients according to risk of developing coronary heart disease (CHD) on 313,000 patients corresponding to 5.5 million months of observations. When compared to the clinically validated Framingham CHD risk score, deep survival analysis is superior in stratifying patients according to their risk.' volume: 56 URL: https://proceedings.mlr.press/v56/Ranganath16.html PDF: http://proceedings.mlr.press/v56/Ranganath16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Ranganath16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Rajesh family: Ranganath - given: Adler family: Perotte - given: Noémie family: Elhadad - given: David family: Blei editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 101-114 id: Ranganath16 issued: date-parts: - 2016 - 12 - 10 firstpage: 101 lastpage: 114 published: 2016-12-10 00:00:00 +0000 - title: 'Multi-task Learning with Weak Class Labels: Leveraging iEEG to Detect Cortical Lesions in Cryptogenic Epilepsy' abstract: 'Multi-task learning (MTL) is useful for domains in which data originates from multiple sources that are individually under-sampled. MTL methods are able to learn classification models that have higher performance as compared to learning a single model by aggregating all the data together or learning a separate model for each data source. The performance of these methods relies on label accuracy. We develop two models that address the problem of multitask learning when the training data has imprecise labels. We apply these methods to the task of detecting abnormal cortical regions in the MRIs of patients suffering from epilepsy whose MRI were deemed normal by neuroradiologists. We use the results of intracranial-EEG exam as an auxiliary source of supervision. The proposed methods successfully detect abnormal regions for all patients in our sample and achieve higher performance as compared to other methods.' volume: 56 URL: https://proceedings.mlr.press/v56/Ahmed16.html PDF: http://proceedings.mlr.press/v56/Ahmed16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Ahmed16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Bilal family: Ahmed - given: Thomas family: Thesen - given: Karen family: Blackmon - given: Ruben family: Kuzniecky - given: Orrin family: Devinsky - given: Jennifer family: Dy - given: Carla family: Brodley editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 115-133 id: Ahmed16 issued: date-parts: - 2016 - 12 - 10 firstpage: 115 lastpage: 133 published: 2016-12-10 00:00:00 +0000 - title: 'gLOP: the global and Local Penalty for Capturing Predictive Heterogeneity' abstract: 'When faced with a supervised learning problem, we hope to have rich enough data to build a model that predicts future instances well. However, in practice, problems can exhibit predictive heterogeneity: most instances might be relatively easy to predict, while others might be predictive outliers for which a model trained on the entire dataset does not perform well. Identifying these can help focus future data collection. We present gLOP, the global and Local Penalty, a framework for capturing predictive heterogeneity and identifying predictive outliers. gLOP is based on penalized regression for multitask learning, which improves learning by leveraging training signal information from related tasks. We give two optimization algorithms for gLOP, one space-efficient, and another giving the full regularization path. We also characterize uniqueness in terms of the data and tuning parameters, and present empirical results on synthetic data and on two health research problems.' volume: 56 URL: https://proceedings.mlr.press/v56/Rose16.html PDF: http://proceedings.mlr.press/v56/Rose16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Rose16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Rhiannon family: Rose - given: Daniel family: Lizotte editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 134-149 id: Rose16 issued: date-parts: - 2016 - 12 - 10 firstpage: 134 lastpage: 149 published: 2016-12-10 00:00:00 +0000 - title: 'Transferring Knowledge from Text to Predict Disease Onset' abstract: 'In many domains such as medicine, training data is in short supply. In such cases, external knowledge is often helpful in building predictive models. We propose a novel method to incorporate publicly available domain expertise to build accurate models. Specifically, we use word2vec models trained on a domain-specific corpus to estimate the relevance of each feature’s text description to the prediction problem. We use these relevance estimates to rescale the features, causing more important features to experience weaker regularization. We apply our method to predict the onset of five chronic diseases in the next five years in two genders and two age groups. Our rescaling approach improves the accuracy of the model, particularly when there are few positive examples. Furthermore, our method selects 60% fewer features, easing interpretation by physicians. Our method is applicable to other domains where feature and outcome descriptions are available.' volume: 56 URL: https://proceedings.mlr.press/v56/Liu16.html PDF: http://proceedings.mlr.press/v56/Liu16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Liu16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Yun family: Liu - given: Collin family: Stultz - given: John family: Guttag - given: Kun-Ta family: Chuang - given: Kun-Ta family: Chuang - given: Fu-Wen family: Liang - given: Huey-Jen family: Su editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 150-163 id: Liu16 issued: date-parts: - 2016 - 12 - 10 firstpage: 150 lastpage: 163 published: 2016-12-10 00:00:00 +0000 - title: 'Preterm Birth Prediction: Stable Selection of Interpretable Rules from High Dimensional Data' abstract: 'Preterm births occur at an alarming rate of 10-15%. Preemies have a higher risk of infant mortality, developmental retardation and long-term disabilities. Predicting preterm birth is difficult, even for the most experienced clinicians. The most well-designed clinical study thus far reaches a modest sensitivity of 18.2–24.2% at specificity of 28.6–33.3%. We take a different approach by exploiting databases of normal hospital operations. We aims are twofold: (i) to derive an easy-to-use, interpretable prediction rule with quantified uncertainties, and (ii) to construct accurate classifiers for preterm birth prediction. Our approach is to automatically generate and select from hundreds (if not thousands) of possible predictors using stability-aware techniques. Derived from a large database of 15,814 women, our simplified prediction rule with only 10 items has sensitivity of 62.3% at specificity of 81.5%.' volume: 56 URL: https://proceedings.mlr.press/v56/Tran16.html PDF: http://proceedings.mlr.press/v56/Tran16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Tran16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Truyen family: Tran - given: Wei family: Luo - given: Dinh family: Phung - given: Jonathan family: Morris - given: Kristen family: Rickard - given: Svetha family: Venkatesh editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 164-177 id: Tran16 issued: date-parts: - 2016 - 12 - 10 firstpage: 164 lastpage: 177 published: 2016-12-10 00:00:00 +0000 - title: 'Learning Robust Features using Deep Learning for Automatic Seizure Detection' abstract: 'We present and evaluate the capacity of a deep neural network to learn robust features from EEG to automatically detect seizures. This is a challenging problem because seizure manifestations on EEG are extremely variable both inter- and intra-patient. By simultaneously capturing spectral, temporal and spatial information our recurrent convolutional neural network learns a general spatially invariant representation of a seizure. The proposed approach exceeds significantly previous results obtained on cross-patient classifiers both in terms of sensitivity and false positive rate. Furthermore, our model proves to be robust to missing channel and variable electrode montage.' volume: 56 URL: https://proceedings.mlr.press/v56/Thodoroff16.html PDF: http://proceedings.mlr.press/v56/Thodoroff16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Thodoroff16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Pierre family: Thodoroff - given: Joelle family: Pineau - given: Andrew family: Lim editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 178-190 id: Thodoroff16 issued: date-parts: - 2016 - 12 - 10 firstpage: 178 lastpage: 190 published: 2016-12-10 00:00:00 +0000 - title: 'Mitochondria-based Renal Cell Carcinoma Subtyping: Learning from Deep vs. Flat Feature Representations' abstract: 'Accurate subtyping of renal cell carcinoma (RCC) is of crucial importance for understanding disease progression and for making informed treatment decisions. New discoveries of significant alterations to mitochondria between subtypes make immunohistochemical (IHC) staining based image classification an imperative. Until now, accurate quantification and subtyping was made impossible by huge IHC variations, the absence of cell membrane staining for cytoplasm segmentation as well as the complete lack of systems for robust and reproducible image based classification. In this paper we present a comprehensive classification framework to overcome these challenges for tissue microarrays of RCCs. We compare and evaluate models based on domain specific hand-crafted "flat"-features versus "deep" feature representations from various layers of a pre-trained convolutional neural network (CNN). The best model reaches a cross-validation accuracy of 89%, which demonstrates for the first time, that robust mitochondria-based subtyping of renal cancer is feasible.' volume: 56 URL: https://proceedings.mlr.press/v56/Schuffler16.html PDF: http://proceedings.mlr.press/v56/Schuffler16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Schuffler16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Peter J. family: Schüffler - given: Judy family: Sarungbam - given: Hassan family: Muhammad - given: Ed family: Reznik - given: Satish family: Tickoo - given: Thomas family: Fuchs editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 191-208 id: Schuffler16 issued: date-parts: - 2016 - 12 - 10 firstpage: 191 lastpage: 208 published: 2016-12-10 00:00:00 +0000 - title: 'Clinical Tagging with Joint Probabilistic Models' abstract: '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 clinical condition to patients based on retrospective analysis of the records, and outperforms baselines which do not account for the noisiness in the labels or do not model the conditions jointly.' volume: 56 URL: https://proceedings.mlr.press/v56/Halpern16.html PDF: http://proceedings.mlr.press/v56/Halpern16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Halpern16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Yoni family: Halpern - given: Steven family: Horng - given: David family: Sontag editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 209-225 id: Halpern16 issued: date-parts: - 2016 - 12 - 10 firstpage: 209 lastpage: 225 published: 2016-12-10 00:00:00 +0000 - title: 'Diagnostic Prediction Using Discomfort Drawings with IBTM' abstract: 'In this paper, we explore the possibility to apply machine learning to make diagnostic predictions using discomfort drawings. Discomfort drawings have proven to be an effective method to collect patient data and make diagnostic decisions in real-life practice. A dataset from relevant patient cases is collected for which medical experts provide diagnostic labels. Next, we use a factorized multimodal topic model, Inter-Battery Topic Model (IBTM), to train a system that can make diagnostic predictions given an unseen discomfort drawing. Experimental results show reasonable predictions of diagnostic labels given an unseen discomfort drawing. Additionally, we generate synthetic discomfort drawings with IBTM given a diagnostic label, which results in typical cases of symptoms. The positive result indicates a significant potential of machine learning to be used for parts of the pain diagnostic process and to be a decision support system for physicians and other health care personnel.' volume: 56 URL: https://proceedings.mlr.press/v56/Zhang16.html PDF: http://proceedings.mlr.press/v56/Zhang16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Zhang16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Cheng family: Zhang - given: Hedvig family: Kjellström - given: Carl Henrik family: Ek - given: Bo family: Bertilson editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 226-238 id: Zhang16 issued: date-parts: - 2016 - 12 - 10 firstpage: 226 lastpage: 238 published: 2016-12-10 00:00:00 +0000 - title: 'Uncovering Voice Misuse Using Symbolic Mismatch' abstract: 'Voice disorders affect an estimated 14 million working-aged Americans, and many more worldwide. We present the first large scale study of vocal misuse based on long-term ambulatory data collected by an accelerometer placed on the neck. We investigate an unsupervised data mining approach to uncovering latent information about voice misuse. We segment signals from over 253 days of data from 22 subjects into over a hundred million single glottal pulses (closures of the vocal folds), cluster segments into symbols, and use symbolic mismatch to uncover differences between patients and matched controls, and between patients pre- and post-treatment. Our results show significant behavioral differences between patients and controls, as well as between some pre- and post-treatment patients. Our proposed approach provides an objective basis for helping diagnose behavioral voice disorders, and is a first step towards a more data-driven understanding of the impact of voice therapy.' volume: 56 URL: https://proceedings.mlr.press/v56/Ghassemi16.html PDF: http://proceedings.mlr.press/v56/Ghassemi16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Ghassemi16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Marzyeh family: Ghassemi - given: Zeeshan family: Syed - given: Daryush family: Mehta - given: Jarrad Van family: Stan - given: Robert family: Hillman - given: John family: Guttag editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 239-252 id: Ghassemi16 issued: date-parts: - 2016 - 12 - 10 firstpage: 239 lastpage: 252 published: 2016-12-10 00:00:00 +0000 - title: 'Directly Modeling Missing Data in Sequences with RNNs: Improved Classification of Clinical Time Series' abstract: 'We demonstrate a simple strategy to cope with missing data in sequential inputs, addressing the task of multilabel classification of diagnoses given clinical time series. Collected from the intensive care unit (ICU) of a major urban medical center, our data consists of multivariate time series of observations. The data is irregularly sampled, leading to missingness patterns in re-sampled sequences. In this work, we show the remarkable ability of RNNs to make effective use of binary indicators to directly model missing data, improving AUC and F1significantly. However, while RNNs can learn arbitrary functions of the missing data and observations, linear models can only learn substitution values. For linear models and MLPs, we show an alternative strategy to capture this signal. Additionally, we evaluate LSTMs, MLPs, and linear models trained on missingness patterns only, showing that for several diseases, what tests are run can be more predictive than the results themselves.' volume: 56 URL: https://proceedings.mlr.press/v56/Lipton16.html PDF: http://proceedings.mlr.press/v56/Lipton16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Lipton16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Zachary C family: Lipton - given: David family: Kale - given: Randall family: Wetzel editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 253-270 id: Lipton16 issued: date-parts: - 2016 - 12 - 10 firstpage: 253 lastpage: 270 published: 2016-12-10 00:00:00 +0000 - title: 'Deep Convolutional Neural Networks for Microscopy-Based Point of Care Diagnostics' abstract: 'Point of care diagnostics using microscopy and computer vision methods have been applied to a number of practical problems, and are particularly relevant to low-income, high disease burden areas. However, this is subject to the limitations in sensitivity and specificity of the computer vision methods used. In general, deep learning has recently revolutionised the field of computer vision, in some cases surpassing human performance for other object recognition tasks. In this paper, we evaluate the performance of deep convolutional neural networks on three different microscopy tasks: diagnosis of malaria in thick blood smears, tuberculosis in sputum samples, and intestinal parasite eggs in stool samples. In all cases accuracy is very high and substantially better than an alternative approach more representative of traditional medical imaging techniques.' volume: 56 URL: https://proceedings.mlr.press/v56/Quinn16.html PDF: http://proceedings.mlr.press/v56/Quinn16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Quinn16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: John A family: Quinn - given: Rose family: Nakasi - given: Pius K. B. family: Mugagga - given: Patrick family: Byanyima - given: William family: Lubega - given: Alfred family: Andama editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 271-281 id: Quinn16 issued: date-parts: - 2016 - 12 - 10 firstpage: 271 lastpage: 281 published: 2016-12-10 00:00:00 +0000 - title: 'A Bayesian Nonparametric Approach for Estimating Individualized Treatment-Response Curves' abstract: 'We study the problem of estimating the continuous response over time of actions from observational time series—a retrospective dataset where the policy by which the data are generated are unknown to the learner. We develop a novel method based on Bayesian nonparametrics (BNP) that can flexibly model functional data and provide posterior inference over the treatment response curves both at the individual and population level. On a challenging dataset containing time series from patients admitted to a hospital, we estimate treatment responses for 8 different treatments used in managing blood pressure and kidney function and show that the resulting fits are more accurate than alternative approaches. Accurate methods for obtaining ITRs from observational data can dramatically accelerate the pace at which personalized treatment plans become possible.' volume: 56 URL: https://proceedings.mlr.press/v56/Xu16.html PDF: http://proceedings.mlr.press/v56/Xu16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Xu16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Yanbo family: Xu - given: Yanxun family: Xu - given: Suchi family: Saria editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 282-300 id: Xu16 issued: date-parts: - 2016 - 12 - 10 firstpage: 282 lastpage: 300 published: 2016-12-10 00:00:00 +0000 - title: 'Doctor AI: Predicting Clinical Events via Recurrent Neural Networks' abstract: 'Leveraging large historical data in electronic health record (EHR), we developed Doctor AI, a generic predictive model that covers observed medical conditions and medication uses. Doctor AI is a temporal model using recurrent neural networks (RNN) and was developed and applied to longitudinal time stamped EHR data from 260K patients and 2,128 physicians over 8 years. Encounter records (e.g. diagnosis codes, medication codes or procedure codes) were input to RNN to predict (all) the diagnosis and medication categories for a subsequent visit. Doctor AI assesses the history of patients to make multilabel predictions (one label for each diagnosis or medication category). Based on separate blind test set evaluation, Doctor AI can perform differential diagnosis with up to 79% recall@30, significantly higher than several baselines. Moreover, we demonstrate great generalizability of Doctor AI by adapting the resulting models from one institution to another without losing substantial accuracy.' volume: 56 URL: https://proceedings.mlr.press/v56/Choi16.html PDF: http://proceedings.mlr.press/v56/Choi16.pdf edit: https://github.com/mlresearch//v56/edit/gh-pages/_posts/2016-12-10-Choi16.md series: 'Proceedings of Machine Learning Research' container-title: 'Proceedings of the 1st Machine Learning for Healthcare Conference' publisher: 'PMLR' author: - given: Edward family: Choi - given: Mohammad Taha family: Bahadori - given: Andy family: Schuetz - given: Walter F. family: Stewart - given: Jimeng family: Sun editor: - given: Finale family: Doshi-Velez - given: Jim family: Fackler - given: David family: Kale - given: Byron family: Wallace - given: Jenna family: Wiens address: Northeastern University, Boston, MA, USA page: 301-318 id: Choi16 issued: date-parts: - 2016 - 12 - 10 firstpage: 301 lastpage: 318 published: 2016-12-10 00:00:00 +0000