Poster Red Neuronal de Enfermedades

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

NEURAL NETWORK AS HELP TO THE MEDICAL

DIAGNOSIS OF 15 DISEASES
Higuera. D, Herrera. H
Flexible computing

Introduction
The low quality of health services and the increase in technology and communication drive people to perform poor health
care exercises such as stop going to the doctor, ask acquaintances what disease they may have and even read on the
Internet, without checking the sources, whether their symptoms should worry them or not. In addition, those who have an
acceptable medical service must be notoriously ill so that doctors can thoroughly examine their problems. This represents a
danger to the patient because if their symptoms are not given importance, a greater degree of illness can develop. That is
Incidencia Social
why the development of an intelligent system of diagnosis of diseases based on 11 symptoms and differentiating 15 diseases
is proposed. The system is developed with Matlab software and its neural network tools. The database was obtained from the
tool: Symptom Verifier on the medical advice page of the American Academy of Family Physicians.

Process and results


Phase I: Data collection Phase II: Creation of the user interface

Phase III: Neural network configuration


Conclusions
Intelligent systems developed with neural networks must
be carefully structured to ensure high performance, speed
of learning and avoid confusion in the network. Another
important factor when implementing this type of systems
net = newff([0 1; 0 1; 0 1;0 1;0 1;0 1;
is the amount of information with which the system learns,
0 1;0 1;0 1;0 1;0 1], as seen in the results the characteristics of the diseases
[11 9 5],{'logsig','logsig','logsig'}) are enough information to generate a reliable system and
nothing confusing.
Phase IV: Results It was possible to structure an intelligent system with
Matlab using a neural network of three layers, one of
entrance, one of exit and one hidden, guaranteeing
veracity in the answers that the user obtains. It is
recommended to increase the description of each disease
and enrich the database to improve the diagnoses and
possibly throw as a response several diseases.

REFERENCES
[1] Ruiz. C, Bualdo. M, «Redes Neuronales: Conceptos Básicos y
Aplicaciones», Universidad Tecnológica Nacional, 2001
[2] Alberto Delgado, «Aplicación de las Redes Neuronales en Medicina,»
Universidad Nacional de Colombia, 1999
[3] M. P. K. F. Maxim Signaevsky, «Artificial intelligence in neuropathology:
deep learning-based assessment of tauopathy.,» Laboratory Investigation,
2019.
[4] C. M. José A. Avellaneda, «IMPLEMENTACIÓN DE REDES NEURONALES
PARA LA DETECCIÓN DE LA PRESENCIA DE ENFERMEDADES EN EL
CORAZÓN,» DOAJ, Colombia, 2012.

Automation Engineering Program

You might also like