Electronic medical record phenotyping using the anchor and learn framework

By Yoni Halpern, Steven Horng, Youngduck Choi, David A. Sontag. 2016

153 citations recorded by the source corpus.

Background Electronic medical records (EMRs) hold a tremendous amount of information about patients that is relevant to determining the optimal approach to patient care. As medicine becomes increasingly precise, a patient’s electronic medical record phenotype will play an important role in triggering clinical decision support systems that can deliver personalized recommendations in real time. Learning with anchors presents a method of efficiently learning statistically driven phenotypes with minimal manual intervention. Materials and Methods We developed a phenotype library that uses both structured and unstructured data from the EMR to represent patients for real-time clinical decision support. Eight of the phenotypes were evaluated using retrospective EMR data on emergency department patients using a set of prospectively gathered gold standard labels. Results We built a phenotype library with 42 publicly available phenotype definitions. Using information from triage time, the phenotype classifiers have an area under the ROC curve (AUC) of infection 0.89, cancer 0.88, immunosuppressed 0.85, septic shock 0.93, nursing home 0.87, anticoagulated 0.83, cardiac etiology 0.89, and pneumonia 0.90. Using information available at the time of disposition from the emergency department, the AUC values are infection 0.91, cancer 0.95, immunosuppressed 0.90, septic shock 0.97, nursing home 0.91, anticoagulated 0.94, cardiac etiology 0.92, and pneumonia 0.97. Discussion The resulting phenotypes are interpretable and fast to build, and perform comparably to statistically learned phenotypes developed with 5000 manually labeled patients. Conclusion Learning with anchors is an attractive option for building a large public repository of phenotype definitions that can be used for a range of health IT applications, including real-time decision support.

Subfields: Artificial Intelligence, Health Informatics, Health Information Management

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Informatics

In Health Information Management

Citations in the corpus

Interpretable Representation Learning for Healthcare via Capturing Disease Progression through Time (2018, methodology): “For example, EHR data have been widely used in various healthcare research tasks such as risk prediction [15, 16, 33], retrospective epidemiologic studies [17, 26, 29] and phenotyping [12, 31].”
Opportunities and obstacles for deep learning in biology and medicine (2017, background): “The anchor and learn framework [92] uses expert knowledge to identify high-confidence observations from which labels can be inferred.”
Comparing deep learning and concept extraction based methods for patient phenotyping from clinical narratives (2018, background): “[16], the TF-IDF of the features reflects the importance of a feature to a note.”
Artificial intelligence in retina. (2018, background): “…the physician will be able to extract the relevant features with a mouse click freeing-up more time for human-to-patient interactions, which will improve care and allow physicians to record and accurately register more individual phenotypes with added individual nuance (Halpern et al., 2016).”
Confident Learning: Estimating Uncertainty in Dataset Labels (2019, background): “, 2014) to human-centric datasets like electronic health records (Halpern et al., 2016) and educational data (Northcutt et al.”
Ethical Machine Learning in Healthcare (2020, background): “Methods to address the positive-unlabeled setting use estimated noise rates (87) or hand-curated labels that are strongly correlated with positive labels, known also as “silver-standard” labels, from clinicians (88).”
Multitask learning and benchmarking with clinical time series data (2017, methodology): “s use simple deﬁnitions based primarily on billing, e.g., ICD-9, codes. The adoption of EHRs has led to increased interest in machine learning approaches to phenotyping that treat it as classiﬁcation [70,71] or clustering [55,72] In this task we classify 25 conditions that are common in adult ICUs, ”
Comparing Rule-Based and Deep Learning Models for Patient Phenotyping (2017, methodology): “Halpern et al.15 address the heavy workload for clinicians and describe a semi-supervised approach to this problem that uses the Anchor and Learn Framework.53 In this scheme, the clinicians only need to define a few anchors, which are phrases that identify concepts with a very high positive predicti”
Opportunities in Machine Learning for Healthcare (2018, unknown): “Machine learning methods to pool different data types and obtain a more reliable label is known as phenotyping (Richesson et al., 2013), and is an important subﬁeld of machine learning in healthcare (Halpern et al., 2016; Yu et al., 2017). Recent work has emphasized the need to integrate the rich in”
A Review of Challenges and Opportunities in Machine Learning for Health. (2018, methodology): “Machine learning methods to pool different data types and obtain a more reliable label is known as phenotyping 30 , and is an important subfield of machine learning in healthcare 31,32 .”
Secure and Robust Machine Learning for Healthcare: A Survey (2020, background): “cation of ML in healthcare [206], which is known as phenotyping [207].”