2011 IEEE International Symposium on IT in Medicine and Education 2011
DOI: 10.1109/itime.2011.6132126
|View full text |Cite
|
Sign up to set email alerts
|

Intelligent brain hemorrhage diagnosis system

Abstract: Diagnosing brain hemorrhage, which is a condition caused by a brain artery busting and causing bleeding is currently done by medical experts using a CT scan. Periodic examination of scans enable the accurate prediction of disease and the extraction of reliable and robust measurement for patients in order to describe the morphological changes in the brain as the recovery progresses. Prior attempts to use medical image processing techniques to extract relevant information from a CT scan has shortcoming due to th… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
2
0
2

Year Published

2012
2012
2023
2023

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(5 citation statements)
references
References 11 publications
(1 reference statement)
0
2
0
2
Order By: Relevance
“…In order to diagnose the brain hemorrhage along with the geometrical and textural feature, Shelke et al uses image processing techniques and medical filters, as input into the neural network machine [12]. AH Ali et al suggested that the bloody strokes can be detected and segmented using texture analyzes from brain CT images, the threshold segmentation process is utilized in the study to obtain the area of the stroke from the CT brain image and a first-order histogram was calculated the statistical feature, as a result, the white color of the image was the mean value, the relatively high mean shows an abnormal region of the brain [13].…”
Section: Related Workmentioning
confidence: 99%
“…In order to diagnose the brain hemorrhage along with the geometrical and textural feature, Shelke et al uses image processing techniques and medical filters, as input into the neural network machine [12]. AH Ali et al suggested that the bloody strokes can be detected and segmented using texture analyzes from brain CT images, the threshold segmentation process is utilized in the study to obtain the area of the stroke from the CT brain image and a first-order histogram was calculated the statistical feature, as a result, the white color of the image was the mean value, the relatively high mean shows an abnormal region of the brain [13].…”
Section: Related Workmentioning
confidence: 99%
“…Tyan et al [36], Balasooriya y Perera [37] usan los métodos de Otsu, filtro anisotrópico, máscara gaussiana, métodos morfológicos (erosión y dilatación) para determinar el área cerebral izquierda o derecha, en la que sucedió la ECV y eliminar ruido de fondo; Haiyan Zhang [70] usa un modelo de contorno geométrico activo mejorado. En imágenes de electroencefalogramas, se captura la duración y el tamaño de la amplitud de las ondas mediante la técnica de ventana deslizante [28,32], la descomposición wavelet [44], el dominio de frecuencias características, transformación de wavelet discreta, la longitud de la transformada de Fourier rápida [21,24,59,71,72,73], y la transformada de ondula discreta de superposición máxima (MODWT) [72]; Ceballos [28] con una RNA multi-layer perceptron crea una matriz de características y patrones del electroencefalograma. Otras técnicas mencionadas en general, para realzar el contraste de la imagen en búsqueda de tumores son: técnicas de media, valores de histograma [44], operaciones morfológicas (dilatación y erosión) [11,25,38,42,62]; curtosis [28,33], matriz de co-ocurrencia de niveles de gris [74,75], desviación estándar, asimetría [57], entropía de Shanon [32], algoritmo Gabor [44], el algoritmo k-means [33]; máscara binaria basada en regiones(max-min) [38], así como diversas técnicas de minería de datos [44].…”
Section: Representación Y Descripción (Extracción Y Selección De Caraunclassified
“…Chawla et al [81] usa una metodología de tres pasos: primero mejora la imagen, después hace una detección de la simetría cerebral y al último hay una clasificación de cortes anormales, similar a la realizada por Bhaiya y Verma [49] donde se utilizan técnicas como la transformación wavelet, el análisis de componentes principales y métodos de aprendizaje supervisado como el algoritmo de retropropagación, red neuronal de funciones radiales y un vector de aprendizaje de cuantización. Toprak [73], Menaka y Kanchana [13] usan al final de la clasificación RNAs con algoritmos basados en en Levenberg-Marquardt, BFGS Quasi-Newton (Broyden-Fletcher-Goldfarb-Shanno with Quasi-Newton, en inglés) y máquinas de soporte vectorial. Otras técnicas descritas en esta etapa son: la técnica de características binarizadas, el algoritmo de código de cadena y la técnica de transformada de características invariantes, usando una distancia euclidiana [13].…”
Section: Interpretación Y Reconocimientounclassified
“…Although they found that higher intensity peak correspond to the soft tissue region, it cannot be generalizable to all brain CT scan. Balasooriya and Perera [2] designed a system to diagnose brain hemorrhage by applying Fuzzy c-means and Watershed algorithm with neural network to CT scan images. To determine the usefulness of the proposed system, they performed an evaluation by conducting domain experts (including technical experts and professors) and intended users.…”
Section: IImentioning
confidence: 99%