Matrices de confusión y otros valores estadísticos - Historial de revisiones

Alexsavio: /* '''Valores estadísticos''' */

2010-07-27T11:28:43Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 13:28 27 jul 2010
Línea 50:		Línea 50:
	For medical images, values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].		For medical images, values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].

	'''Precision:''' TP / N
			'''Precision''': is the proportion of the predicted positive cases that were correct, as calculated using the equation: '''precision''' = TP / (TP+FP)

Alexsavio: /* '''Definiciones''' */

2010-07-27T11:20:35Z

'''Definiciones'''

← Revisión anterior		Revisión del 13:20 27 jul 2010
Línea 17:		Línea 17:


	La matriz de ~~contingencia~~ donde las filas corresponden a la verdad del terreno y las columnas a los resultados:		La matriz de confusión donde las filas corresponden a la verdad del terreno y las columnas a los resultados:

	'''TP FP'''		'''TP FP'''
Línea 23:		Línea 23:
	'''FN TN'''		'''FN TN'''

	En muchas referencias las columnas corresponden a la verdad del terreno y las filas a los resultados, entonces la matriz de ~~contingencia~~ es la traspuesta de la descrita arriba.		En muchas referencias las columnas corresponden a la verdad del terreno y las filas a los resultados, entonces la matriz de confusión es la traspuesta de la descrita arriba.


Línea 29:		Línea 29:

	'''N = número total de vóxeles de la imagen'''		'''N = número total de vóxeles de la imagen'''


	== '''Valores estadísticos''' ==		== '''Valores estadísticos''' ==

Alexsavio: /* '''Valores estadísticos''' */

2009-12-20T17:48:17Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 19:48 20 dic 2009
Línea 51:		Línea 51:
	For medical images, values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].		For medical images, values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].

			'''Precision:''' TP / N


	'''True Positive ~~Fraction~~ or Sensitivity:''' TPF = TP / (TP + FN) = sensitivity		'''True Positive Rate or Sensitivity or Recall:''' TPF = TP / (TP + FN) = sensitivity

	Puede ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.


			'''F-score:''' Precision x Recall

	'''True Negative ~~Fraction~~ or Specificity:''' TNF = TN / (TN + FP) = specificity
			'''True Negative Rate or Specificity:''' TNF = TN / (TN + FP) = specificity

	Puede ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.
Línea 65:		Línea 68:


	'''False Positive ~~Fraction~~:''' FPF = FP / (FP + TN) = 1 - specificity		'''False Positive Rate:''' FPF = FP / (FP + TN) = 1 - specificity

	Puede ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.

Alexsavio: /* '''Valores estadísticos''' */

2009-05-13T15:52:40Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 17:52 13 may 2009
Línea 55:		Línea 55:
	'''True Positive Fraction or Sensitivity:''' TPF = TP / (TP + FN) = sensitivity		'''True Positive Fraction or Sensitivity:''' TPF = TP / (TP + FN) = sensitivity

	~~Esta puede~~ ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.


Línea 61:		Línea 61:
	'''True Negative Fraction or Specificity:''' TNF = TN / (TN + FP) = specificity		'''True Negative Fraction or Specificity:''' TNF = TN / (TN + FP) = specificity

	~~Esta puede~~ ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.


Línea 67:		Línea 67:
	'''False Positive Fraction:''' FPF = FP / (FP + TN) = 1 - specificity		'''False Positive Fraction:''' FPF = FP / (FP + TN) = 1 - specificity

	~~Esta puede~~ ser multiplicada por 100 para expresarlo como un porcentaje.		Puede ser multiplicada por 100 para expresarlo como un porcentaje.

Alexsavio: /* '''Valores estadísticos''' */

2009-05-13T15:52:11Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 17:52 13 may 2009
Línea 59:		Línea 59:


	'''True Negative Fraction or Specificity:''' TNF = TN / (FP ~~+ TN~~) = specificity		'''True Negative Fraction or Specificity:''' TNF = TN / (TN + FP) = specificity

	Esta puede ser multiplicada por 100 para expresarlo como un porcentaje.		Esta puede ser multiplicada por 100 para expresarlo como un porcentaje.

Alexsavio: /* '''Valores estadísticos''' */

2009-02-17T16:55:54Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 18:55 17 feb 2009
Línea 49:		Línea 49:
	It is also known as "overlap ratio" when used for segmentation validation. This sensitive measure varies in the range [0,1]...		It is also known as "overlap ratio" when used for segmentation validation. This sensitive measure varies in the range [0,1]...

	~~(for~~ medical images)		For medical images, values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].

	~~where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].~~

	~~Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.~~

Alexsavio: /* '''Valores estadísticos''' */

2009-02-17T16:47:17Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 18:47 17 feb 2009
Línea 51:		Línea 51:
	(for medical images)		(for medical images)

	where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors (Avants et al., 2007).		where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors [http://dx.doi.org/10.1016/j.media.2007.06.004 (Avants et al., 2007)].

	Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.		Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.

Alexsavio: /* '''Valores estadísticos''' */

2009-02-17T16:43:33Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 18:43 17 feb 2009
Línea 47:		Línea 47:
	Dice is similar to Jaccard but gives twice the weight to agreements.		Dice is similar to Jaccard but gives twice the weight to agreements.

	It is also known as "overlap ratio" when used for segmentation validation ~~and computed as: S(R1,R2) = (2*(R1 - R2)) / (R1 + R2)~~.		It is also known as "overlap ratio" when used for segmentation validation. This sensitive measure varies in the range [0,1]...
	~~where R1 and R2 are the number of pixels within the segmented regions to be compared.~~
	This sensitive measure varies in the range [0,1]...

	(for medical images)		(for medical images)

Alexsavio: /* '''Valores estadísticos''' */

2009-02-17T16:39:13Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 18:39 17 feb 2009
Línea 50:		Línea 50:
	where R1 and R2 are the number of pixels within the segmented regions to be compared.		where R1 and R2 are the number of pixels within the segmented regions to be compared.
	This sensitive measure varies in the range [0,1]...		This sensitive measure varies in the range [0,1]...

	(for medical images)		(for medical images)

	where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors (Avants et al., 2007).		where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors (Avants et al., 2007).

	Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.		Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.

Alexsavio: /* '''Valores estadísticos''' */

2009-02-17T16:38:42Z

'''Valores estadísticos'''

← Revisión anterior		Revisión del 18:38 17 feb 2009
Línea 46:		Línea 46:

	Dice is similar to Jaccard but gives twice the weight to agreements.		Dice is similar to Jaccard but gives twice the weight to agreements.

			It is also known as "overlap ratio" when used for segmentation validation and computed as: S(R1,R2) = (2*(R1 - R2)) / (R1 + R2).
			where R1 and R2 are the number of pixels within the segmented regions to be compared.
			This sensitive measure varies in the range [0,1]...
			(for medical images)
			where values greater than 0.6 for smaller structures and 0.8 for larger structures are considered good by some authors (Avants et al., 2007).
			Apart from being highly dependent upon the application, isn't it also dependent upon image resolution? As the higher my resolution is, then better should the segmentation algorithm work.

Matrices de confusión y otros valores estadí­sticos - Historial de revisiones

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Definiciones''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Alexsavio: /* '''Valores estadísticos''' */

Matrices de confusión y otros valores estadísticos - Historial de revisiones