Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Local Anesthetics: Differential Sensitivity of Nerve Fibers

Local Anesthetics: Differential Sensitivity of Nerve Fibers

Local anesthetics (LAs) block the sodium channels of nerve trunks, sensory nerve endings, and neuromuscular junctions. Although LAs can block all kinds of nerves, the sensitivity of nerve fibers differs according to nerve types and structures. LAs are known to block myelinated fibers faster than unmyelinated ones. Also, they block pain or sensory neurons at low concentrations without affecting the motor neurons involved in muscle contractions. This helps relieve labor pain without affecting the...

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...

Sensitivity, Specificity, and Predicted Value

Sensitivity, Specificity, and Predicted Value

In healthcare diagnostics, laboratory tests play a crucial role in identifying and diagnosing a wide range of medical conditions. However, interpreting test results is not always straightforward. An abnormal test result does not always confirm the presence of a disease, just as a normal result does not guarantee its absence. To assess the reliability of these diagnostic tools, healthcare practitioners rely on two key statistical indicators: sensitivity and specificity.
Sensitivity is the...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Molecular assembly and disassembly: novel photolabile molecular hosts

Organic letters·2000

Same author

Co-Infection by HTLV-I/II is Associated With Increased Viral Load in PBMC of HIV-1 Infected Patients in Bahia, Brazil.

The Brazilian journal of infectious diseases : an official publication of the Brazilian Society of Infectious Diseases·2000

Same author

Dynamic electrorheological effects and interparticle force between a pair of rotating spheres

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2000

Same author

Mechanism of edge recurrence following brachytherapy treatment of in-stent restenosis: a serial intravascular ultrasound study.

Cardiovascular radiation medicine·2000

Same author

Analytical systematic approximate method of a two-state dissipative system

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2000

Same author

Force between two spherical inclusions in a nonlinear host medium

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2000

Search research articles

Related Experiment Video

Updated: Jul 11, 2026

Functional Evaluation of Biological Neurotoxins in Networked Cultures of Stem Cell-derived Central Nervous System Neurons

Functional Evaluation of Biological Neurotoxins in Networked Cultures of Stem Cell-derived Central Nervous System Neurons

Published on: February 5, 2015

Evaluation of neural network models with generalized sensitivity analysis

Harrington¹, Urbas, Wan

¹Ohio University Center for Intelligent Chemical Instrumentation, Department of Chemistry and Biochemistry, Ohio University, Athens 45701-2979, USA. Peter.Harrington@ohio.edu

Analytical Chemistry

|October 31, 2000

Summary

A new sensitivity analysis method enhances neural network models by analyzing input data features. This approach improves model predictability and identifies linearity in data, aiding in error diagnosis.

More Related Videos

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Related Experiment Videos

Last Updated: Jul 11, 2026

Functional Evaluation of Biological Neurotoxins in Networked Cultures of Stem Cell-derived Central Nervous System Neurons

Functional Evaluation of Biological Neurotoxins in Networked Cultures of Stem Cell-derived Central Nervous System Neurons

Published on: February 5, 2015

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Area of Science:

Machine Learning
Data Analysis

Background:

Neural network models are complex and nonlinear, making feature interpretation challenging.
Existing methods may not fully capture the nuances of input data characteristics.

Purpose of the Study:

To develop a novel sensitivity analysis method for neural network classification models.
To enhance the interpretability of characteristic features within input data.
To improve the predictability and diagnose errors in neural network models.

Main Methods:

Devised a sensitivity analysis method based on the gradient of the neural network response function.
Utilized two criteria for measuring sensitivity: gradient with respect to class average and average sensitivity of class objects.
Applied the method to temperature-constrained cascade correlation network (TCCCN) models with weight constraints and conjugate gradient training.
Evaluated the method using synthetic data and experimental mobility spectra.

Main Results:

Temperature-constrained hidden units and weight constraints yielded more sensitive neural network models.
The sensitivity analysis successfully identified linearity in input variables by comparing class mean input sensitivity and mean sensitivity.
The method diagnosed errors in training data and indicated a TCCCN architecture with better predictability.

Conclusions:

The developed sensitivity analysis method effectively reveals characteristic input data features for neural networks.
TCCCNs with specific constraints offer enhanced sensitivity and predictability.
The method provides a robust tool for model interpretability, error detection, and architecture optimization.