Dataset Information

A roadmap for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling.

ABSTRACT: Predictive modeling based on machine learning with medical data has great potential to improve healthcare and reduce costs. However, two hurdles, among others, impede its widespread adoption in hdealthcare. First, medical data are by nature longitudinal. Pre-processing them, particularly for feature engineering, is labor intensive and often takes 50-80% of the model building effort. Predictive temporal features are the basis of building accurate models, but are difficult to identify. This is problematic. Healthcare systems have limited resources for model building, while inaccurate models produce sub-optimal outcomes and are often useless. Second, most machine learning models provide no explanation of their prediction results. However, offering such explanations is essential for a model to be used in usual clinical practice. To address these two hurdles, this paper outlines: 1) a data-driven method for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling; and 2) a method of using these features to automatically explain machine learning prediction results and suggest tailored interventions. This provides a roadmap for future research.

SUBMITTER: Luo G

PROVIDER: S-EPMC6482973 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A roadmap for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling.

Luo Gang G

Global transitions 20190327

Predictive modeling based on machine learning with medical data has great potential to improve healthcare and reduce costs. However, two hurdles, among others, impede its widespread adoption in hdealthcare. First, medical data are by nature longitudinal. Pre-processing them, particularly for feature engineering, is labor intensive and often takes 50-80% of the model building effort. Predictive temporal features are the basis of building accurate models, but are difficult to identify. This is pro ...[more]

PMID: 31032483

Similar Datasets

Project description:BackgroundThere is a frequent need to obtain sets of functionally equivalent homologous proteins (FEPs) from different species. While it is usually the case that orthology implies functional equivalence, this is not always true; therefore datasets of orthologous proteins are not appropriate. The information relevant to extracting FEPs is contained in databanks such as UniProtKB/Swiss-Prot and a manual analysis of these data allow FEPs to be extracted on a one-off basis. However there has been no resource allowing the easy, automatic extraction of groups of FEPs - for example, all instances of protein C.We have developed FOSTA, an automatically generated database of FEPs annotated as having the same function in UniProtKB/Swiss-Prot which can be used for large-scale analysis. The method builds a candidate list of homologues and filters out functionally diverged proteins on the basis of functional annotations using a simple text mining approach.ResultsLarge scale evaluation of our FEP extraction method is difficult as there is no gold-standard dataset against which the method can be benchmarked. However, a manual analysis of five protein families confirmed a high level of performance. A more extensive comparison with two manually verified functional equivalence datasets also demonstrated very good performance.ConclusionIn summary, FOSTA provides an automated analysis of annotations in UniProtKB/Swiss-Prot to enable groups of proteins already annotated as functionally equivalent, to be extracted. Our results demonstrate that the vast majority of UniProtKB/Swiss-Prot functional annotations are of high quality, and that FOSTA can interpret annotations successfully. Where FOSTA is not successful, we are able to highlight inconsistencies in UniProtKB/Swiss-Prot annotation. Most of these would have presented equal difficulties for manual interpretation of annotations. We discuss limitations and possible future extensions to FOSTA, and recommend changes to the UniProtKB/Swiss-Prot format, which would facilitate text-mining of UniProtKB/Swiss-Prot.

Project description:A dataset range of isocenter congruency verification tests have been examined from a statistical perspective for the purpose of establishing tolerance levels that are meaningful, based on the fundamental limitation of linear accelerator isocentricity and the demands of a high-precision stereotactic radiosurgery program. Using a laser-defined isocenter, a total of 149 individual isocenter congruency tests were examined with recorded values for ideal spatial corrections to the isocenter test tool. These spatial corrections were determined from radiation exposures recorded on an electronic portal imaging device (EPID) at various gantry, collimator, and treatment couch combinations. The limitations of establishing an ideal isocenter were quantified from each variable which contributed to uncertainty in isocenter definition. Individual contributors to uncertainty, specifically, daily positioning setup errors, gantry sag, multileaf collimator (MLC) offset, and couch walkout, were isolated from isocenter congruency measurements to determine a clinically meaningful isocenter measurement. Variations in positioning of the test tool constituted, on average, 0.38 mm magnitude of correction. Gantry sag and MLC offset contributed 0.4 and 0.16 mm, respectively. Couch walkout had an average degrading effect to isocenter of 0.72 mm. Considering the magnitude of uncertainty contributed by each uncertainty variable and the nature of their combination, an appropriate schedule action and immediate action level were determined for use in analyzing daily isocenter congruency test results in a stereotactic radiosurgery (SRS) program. The recommendations of this study for this linear accelerator include a schedule action level of 1.25 mm and an immediate action level of 1.50mm, requiring prompt correction response from clinical medical physicists before SRS or stereotactic body radiosurgery (SBRT) is administered. These absolute values were derived from considering relative data from a specific linear accelerator and, therefore, represent a means by which a numerical quantity can be used as a test threshold with relative specificity to a particular linear accelerator.

Dataset Information

A roadmap for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling.

Publications

A roadmap for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets