Diagnosis of pathology on the vertebral column with backpropagation and Naive Bayes classifier

Abstract

Minimizing human errors in diagnosis is one of the most important aims of Computer-Aided Diagnosis (CAD) systems. CAD software is the software helping radiologists , specialist doctor for detect abnormality on medical views by using advanced pattern diagnosis and view processing methods. In this study, a CAD system that can help detection and diagnosis of Pathology on the Vertebral Column is introduced. The vertebral database used in the application is obtained from UCI (Machine Learning Repository). Various artificial intelligence methods were applied for the classification of this database obtained. The results obtained are comparatively explained in detail in the related part.

Genetic Algorithm to Optimize k-Nearest Neighbor Parameter for Benchmarked Medical Datasets Classification

“…Mostly, diagnosis methods in medical field are based on data classification approaches and systematized [1]. Use of Computer-Aided Diagnosis (CAD) systems can assist doctors to diagnose patient illnesses [2], Classification is the most commonly performed CAD system among the various tasks that can be performed [3]. In an effort to ensure accurate diagnostic assistance, the main problems with the classification of medical datasets can be categorized as complex optimization problems [1].…”

Effective Prediction of Vertebral Column Pathologies Using Random Forest

“…The lumbar region is the major part where vertebral column pathologies are present. For detecting those pathologies MRI or radiology images are the best sources [10].…”

“…The experimental work dataset is taken from the UCI machine learning repository and accrued the classification accuracy of 88.56% compared to others. Yavuzunal and ErdincKocer et al [17] proposed the identification of vertebral column disorders using the combined approach of backpropagation neural network and naive Bayesian classifiers. The author conducted the observation over the input dataset taken from the UCI machine learning repository and achieves the average classification accuracy of about 84.35% on both classifiers.…”

Clasificación de patologías presentes en la columna vertebral mediante técnicas de máquinas de aprendizaje

“…De forma paralela con lo anterior, investigadores de las universidades de Amasya y Selcuk (Turquía) realizaron una publicación sobre el diagnóstico de patologías en columna vertebral a través de técnicas como perceptrón multicapa (mlp) y teorema de Bayes. El trabajo realizado consistió en presentar un sistema de diagnóstico asistido por computadora (cad) que pudiera ayudar minimizar los errores humanos en el diagnóstico [11]…”