Corpus ID: 51940203

Learning to Write Notes in Electronic Health Records

@article{Liu2018LearningTW,
  title={Learning to Write Notes in Electronic Health Records},
  author={Peter J. Liu},
  journal={ArXiv},
  year={2018},
  volume={abs/1808.02622}
}
Clinicians spend a significant amount of time inputting free-form textual notes into Electronic Health Records (EHR) systems. Much of this documentation work is seen as a burden, reducing time spent with patients and contributing to clinician burnout. With the aspiration of AI-assisted note-writing, we propose a new language modeling task predicting the content of notes conditioned on past data from a patient's medical record, including patient demographics, labs, medications, and past notes… Expand
Fast, Structured Clinical Documentation via Contextual Autocomplete
TLDR
To the knowledge, this system is the only machine learning-based documentation utility for clinical notes deployed in a live hospital setting, and it reduces keystroke burden of clinical concepts by 67% in real environments. Expand
Clinical Note Generation to Address Physician Burnout
Physician burnout is one of the main contributors to the decreasing quality and personalized touch of clinical visits. Burnout in large part can be attributed to tedious and inefficient documentationExpand
Generation and evaluation of artificial mental health records for Natural Language Processing
TLDR
It is found that using this artificial data as training data can lead to classification results that are comparable to the original results, which is an important finding for the long-term goal of being able to generate artificial clinical data that can be released to the wider research community and accelerate advances in developing computational methods that use healthcare data. Expand
Smart Vet : Autocompleting Sentences in Veterinary Medical Records
Every day, veterinarians write tens of thousands of medical records, mostly in standard formats following the SOAP structure: “Subjective”, “Objective”, “Assessment”, and “Plan”. These notes recordExpand
Generating Synthetic Training Data for Supervised De-Identification of Electronic Health Records
TLDR
The experiments show that artificial documents can be used to train a supervised named-entity recognition model for de-identification, which outperforms a state-of-theart rule-based baseline, and it is shown that combining real data with synthetic data improves the recall of the method, without manual annotation effort. Expand
The digital scribe
TLDR
Digital scribes promisingly offer a gateway into the clinical workflow for more advanced support for diagnostic, prognostic and therapeutic tasks, and are likely to evolve through three broad stages. Expand
A CNN-LSTM for predicting mortality in the ICU
TLDR
This thesis investigates the potential of a combination of a deep learning model and notes for predicting mortality with a higher accuracy and finds that the CNN-LSTM outperforms the baseline on all experiments which serves as a proof-of-concept of how notes and deep learning can better outcome prediction. Expand
Unsupervised Pseudo-Labeling for Extractive Summarization on Electronic Health Records
TLDR
This work studied how to utilize the intrinsic correlation between multiple EHRs to generate pseudo-labels and train a supervised model with no external annotation that is effective in summarizing crucial disease-specific information for patients. Expand
Fast Prototyping a Dialogue Comprehension System for Nurse-Patient Conversations on Symptom Monitoring
TLDR
This work proposes a framework inspired by nurse-initiated clinical symptom monitoring conversations to construct a simulated human-human dialogue dataset, embodying linguistic characteristics of spoken interactions like thinking aloud, self-contradiction, and topic drift, and demonstrates the feasibility for efficient and effective extraction, retrieval and comprehension of symptom checking information discussed in multi-turn human- human spoken conversations. Expand
Exploring Transformer Text Generation for Medical Dataset Augmentation
TLDR
A methodology guiding the generation with structured patient information in a sequence-to-sequence manner is proposed, and it is demonstrated that the augmented dataset is capable of beating the authors' baselines on a downstream classification task. Expand
...
1
2
...

References

SHOWING 1-10 OF 25 REFERENCES
Scalable and accurate deep learning with electronic health records
TLDR
A representation of patients’ entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format is proposed, and it is demonstrated that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. Expand
Doctor AI: Predicting Clinical Events via Recurrent Neural Networks
TLDR
Great generalizability of Doctor AI is demonstrated by adapting the resulting models from one institution to another without losing substantial accuracy, significantly higher than several baselines. Expand
Automatic generation of textual summaries from neonatal intensive care data
TLDR
A prototype, called BT-45, is presented, which generates textual summaries of about 45 minutes of continuous physiological signals and discrete events and brings together techniques from the different areas of signal processing, medical reasoning, knowledge engineering, and natural language generation. Expand
On the Automatic Generation of Medical Imaging Reports
TLDR
This work builds a multi-task learning framework which jointly performs the prediction of tags and the generation of paragraphs, proposes a co-attention mechanism to localize regions containing abnormalities and generate narrations for them, and develops a hierarchical LSTM model to generate long paragraphs. Expand
The rise of the medical scribe industry: implications for the advancement of electronic health records.
TLDR
With federal meaningful-use incentives driving adoption of electronic health records (EHRs), physicians are increasingly concerned about the time spent documenting patient information and managing orders via computerized patient order entry (CPOE), and a new industry has emerged nationally to provide physicians with medical scribes. Expand
Reproducibility in critical care: a mortality prediction case study
TLDR
This study reviews publications that have reported performance of mortality prediction models based on the Medical Information Mart for Intensive Care (MIMIC) database and attempts to reproduce the cohorts used in their studies, and compares the performance reported in the studies against gradient boosting and logistic regression models. Expand
Research Paper: Automated Encoding of Clinical Documents Based on Natural Language Processing
TLDR
Extraction of relevant clinical information and UMLS coding were accomplished using a method based on NLP that appeared to be comparable to or better than six experts, rendering the coded output suitable for effective retrieval. Expand
MIMIC-III, a freely accessible critical care database
MIMIC-III (‘Medical Information Mart for Intensive Care’) is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary careExpand
Deep learning for healthcare: review, opportunities and challenges
TLDR
It is suggested that deep learning approaches could be the vehicle for translating big biomedical data into improved human health and develop holistic and meaningful interpretable architectures to bridge deep learning models and human interpretability. Expand
Allocation of Physician Time in Ambulatory Practice: A Time and Motion Study in 4 Specialties
TLDR
The goal was to describe time allocation and practice characteristics for physicians in the era of EHRs and federal incentive and penalty programs and to ensure a participant base that was representative of a large and inclusive number of physicians. Expand
...
1
2
3
...