A similarity measure for case based reasoning modeling with temporal abstraction based on cross-correlation

Florian Hartge¹, Thomas Wetter, Walter E Haefeli

¹Institute for Medical Biometry and Informatics, Department Medical Informatics, Im Neuenheimer Feld 400, D-69120 Heidelberg, Germany. florian.hartge@med.uni-heidelberg.de

Computer Methods and Programs in Biomedicine

|December 20, 2005

Summary

This study introduces a novel cross-correlation algorithm to detect similarities in patient drug treatment patterns. The method effectively identifies similar dosing and timing, potentially reducing medication errors and adverse drug events (ADEs).

Area of Science:

Pharmacovigilance
Computational Biology
Signal Processing

Background:

Adverse drug events (ADEs) significantly impact drug safety, often stemming from incorrect dosing or drug interactions.
Current risk assessment strategies need to integrate co-administered drugs, individual doses, and their timing for comprehensive safety evaluation.

Purpose of the Study:

To evaluate the performance of cross-correlation, a signal processing technique, for identifying similarities in patient treatment regimens.
To develop a computational approach for analyzing temporal patterns in drug administration and dosage.

Main Methods:

Patient treatments were modeled as vectors representing discrete time intervals.
Cross-correlation analysis was applied to these vectors to identify clusters indicating treatment similarities.
A test case dataset was created to evaluate the algorithm's specificity and performance.

Main Results:

The algorithm achieved high similarity scores for identical (0.699) and similar (0.528) treatment patterns.
It demonstrated low similarity scores for dissimilar treatment patterns.
The approach showed a high specificity of 27/30 in distinguishing between similar and dissimilar treatments.

Conclusions:

Cross-correlation analysis is a viable method for detecting similarities in patient treatment time courses and dosages.
This approach holds potential for reducing false alarms in medication error and ADE alerting systems.
The findings suggest a promising new direction for enhancing drug safety through computational analysis of treatment data.

Related Experiment Videos

A similarity measure for case based reasoning modeling with temporal abstraction based on cross-correlation

Related Concept Videos

Prospective MRI Assessment of Early Myelination Over the First Month of Life in Neonates With Neonatal Hypoxic-Ischemic Encephalopathy Treated With Therapeutic Hypothermia With or Without Sildenafil.

Evaluating the quality, feasibility and patient satisfaction of medication history taking by telephone for patients with scheduled admissions: a pilot study.

Quantification of Arginine-Rich Cyclic Cell-Penetrating Peptide-Lipid Conjugates Using a Surrogate Peptide Extracted by Phospholipase D Digestion and Trifluoroacetic Acid-Based UPLC-MS/MS Analysis.

Transporting trial results to synthetic real-world populations in order to estimate real-world effectiveness of newly marketed medicines.

Towards a prescribing monitoring system for medication safety evaluation within electronic health records: a scoping review.

Testing Higher Doses of Sildenafil to Repair Brain Injury Secondary to Birth Asphyxia: An Open-Label Dose-Finding Phase 1b Clinical Trial-Sildenafil Administration to Treat Neonatal Encephalopathy-Study 02.

Reinforcement learning-based dynamic ensemble for missense variant effect prediction and tiered prioritization of VUS.

Information bottleneck-driven Gaussian PID enables modality-level interpretability in multimodal survival prediction.

Interpretable Machine Learning Model Using Oxygenation and Respiratory Variability to Predict Hemorrhagic Stroke Mortality: A Multicenter Validation Study.

Adaptive graph convolutional neural network incorporating ECG for individualized motor imagery EEG classification.

Sigmatism detection from child speech spectrograms using convolutional autoencoders and support vector machines.

Alzheimer's disease-related β-amyloid deposition prediction based on plasma biomarkers, ApoE4, and CDR with a machine learning approach.

Discovery of spirocyclic amide scaffolds as novel neddylation E1 inhibitors to suppress the growth and survival of lung cancer cells.

In silico discovery of CETP inhibitors from ZINC15 compounds using docking, ADMET filtering, and molecular dynamics simulations.

Machine learning-based prediction of SARS-CoV-2 bioactivity: integrating IC50 regression and activity classification using multi-task neural networks.

Novel salicylamido sulfonamides as CD73 immune checkpoint inhibitors: Synthesis and in vitro biological evaluation.

High-throughput screening and quantification of eight anti-dandruff agents and antibiotics in hair care products by direct analysis in real-time high-resolution mass spectrometry and HPLC.

Genome-wide analysis of lipoxygenase genes in Coffea arabica and its diploid progenitors.

Related Experiment Videos

A similarity measure for case based reasoning modeling with temporal abstraction based on cross-correlation

Area of Science:

Background:

Purpose of the Study:

Main Methods:

Main Results:

Conclusions:

Related Concept Videos

Related Articles