Counting trees in Random Forests: Predicting symptom severity in psychiatric intake reports

Abstract

The CEGS N-GRID 2016 Shared Task (Filannino, Stubbs, Uzuner (2017)) in Clinical Natural Language Processing introduces the assignment of a severity score to a psychiatric symptom, based on a psychiatric intake report. We present a method that employs the inherent interview-like structure of the report to extract relevant information from the report and generate a representation. The representation consists of a restricted set of psychiatric concepts (and the context they occur in), identified using medical concepts defined in UMLS that are directly related to the psychiatric diagnoses present in the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) ontology. Random Forests provides a generalization of the extracted, case-specific features in our representation. The best variant presented here scored an inverse mean absolute error (MAE) of 80.64%. A concise concept-based representation, paired with identification of concept certainty and scope (family, patient), shows a robust performance on the task.

A natural language processing challenge for clinical records: Research Domains Criteria (RDoC) for psychiatry

“…They experimented with different classifiers and approaches. The classifiers most prominently used were Support Vector Regressors [22], Decision Trees [23], Random Forests [24] and Gradient Tree Boosting [25]. The approaches included an ensemble of Convolutional Neural Networks (CNN) with word embeddings [26] and a mixture of Regularized Multinomial Logistic Regression classifiers and Neural Networks [27].…”

“…Due to the availability of unannotated data, in addition to their supervised solutions, four teams experimented with semi-supervised approaches, e.g., [22, 24]. Only three teams involved medical experts.…”

Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients

“…The goal is to decrease the correlation between individual trees, which results in diminished variance when the trees are aggregated. Random forests accommodate sparsity [12], which is favorable in this case, due to a low percentage of patients who reached the primary outcome. The individual trees are designed to overfit on features (making very specific decisions that only account for part of the data set), whereas the voting strategy mitigates these effects by generalizing over the decisions of multiple trees.…”

Symptom severity prediction from neuropsychiatric clinical records: Overview of 2016 CEGS N-GRID shared tasks Track 2

“…They predicted the labels by using a SVM classifier with re-weighted responses to reflect the unbalanced nature of the data. The Antwerp Universitys research team corrected misspellings and the erroneously concatenated words by using hand-written regular expressions and mapped the words to UMLS concepts by using fuzzy matching rules [24]. Instead of using the entire set of mapped concepts, they restricted it to the ones related to psychiatric diagnoses by means of the Diagnostic & Statistical Manual of Mental Disorders (DSM).…”