The Use of Autoencoders for Discovering Patient Phenotypes
نویسندگان
چکیده
We use autoencoders to create low-dimensional embeddings of underlying patient phenotypes that we hypothesize are a governing factor in determining how different patients will react to different interventions. We compare the performance of autoencoders that take fixed length sequences of concatenated timesteps as input with a recurrent sequence-to-sequence autoencoder. We evaluate our methods on around 35,500 patients from the latest MIMIC III dataset from Beth Israel Deaconess Hospital.
منابع مشابه
Sparse Autoencoders in Sentiment Analysis
This paper examines the utilization of sparse autoencoders in the task of sentiment analysis. The autoencoders can be used for pre-training a deep neural network, discovering new features or for dimensionality reduction. In this paper, sparse autoencoders were used for parameters initialization in deep neural network. Experiments showed that the accuracy of text classification to a particular s...
متن کاملSemi-Supervised Learning of the Electronic Health Record with Denoising Autoencoders for Phenotype Stratification
Patient interactions with health care providers result in entries to electronic health records (EHRs). EHRs were built for clinical and billing purposes but contain many data points about an individual. Mining these records provides opportunities to extract electronic phenotypes that can be paired with genetic data to identify genes underlying common human diseases. This task remains challengin...
متن کاملDetermination of Alpha 1-Antitrypsin Phenotypes and Genotypes in Iranian Patients
Alpha 1-antitrypsin (AAT) or alpha 1-protease inhibitor (PI) is the principal inhibitor of proteolytic enzyme in serum. Its phenotypic variability has been reported to be associated with liver, lung diseases and rheumatoid arthritis in humans. There is much documentation about high risk phenotypes of PI in some regions of the world, however, there are no reliable reports on these phenotypes and...
متن کاملModeling a semantic recommender system for medical prescriptions and drug interaction detection
Introduction: The administration of appropriate drugs to patients is one of the most important processes of treatment and requires careful decision-making based-on the current conditions of the patient and its history and symptoms. In many cases, patients may require more than one drug, or in addition to having a previous illness and receiving the drug, they need new drugs for the new illness, ...
متن کاملSemi-supervised learning of the electronic health record for phenotype stratification
Patient interactions with health care providers result in entries to electronic health records (EHRs). EHRs were built for clinical and billing purposes but contain many data points about an individual. Mining these records provides opportunities to extract electronic phenotypes, which can be paired with genetic data to identify genes underlying common human diseases. This task remains challeng...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1703.07004 شماره
صفحات -
تاریخ انتشار 2017