Vanessa - Revision history

imported>Psych204B at 22:20, 10 March 2012

2012-03-10T22:20:22Z

← Older revision		Revision as of 22:20, 10 March 2012
Line 115:		Line 115:
	* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.ppt PPT Presentation]		* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.ppt PPT Presentation]
	* [http://www.vbmis.com/bmi/class/psych204b/finalica.pdf PPT Presentation PDF]		* [http://www.vbmis.com/bmi/class/psych204b/finalica.pdf PPT Presentation PDF]
	* [http://www.vbmis.com/bmi/class/psych204b/finalica html Presentation]		* [http://www.vbmis.com/bmi/class/psych204b/finalica.zip html (.zip) Presentation]

	====Script====		====Script====

imported>Psych204B at 22:18, 10 March 2012

2012-03-10T22:18:04Z

← Older revision		Revision as of 22:18, 10 March 2012
Line 77:		Line 77:
	==CLASSIFICATION==		==CLASSIFICATION==
	To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This mask was used to extract Z score values from individual DMN maps to result in an n x p matrix, with n rows of subjects and p columns of features. The data was split into a test (n = 100) and training (n=89) set with an equal distribution of ADHD and Control. The resulting SVM performed with only 44% accuracy, worse than chance.		To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This mask was used to extract Z score values from individual DMN maps to result in an n x p matrix, with n rows of subjects and p columns of features. The data was split into a test (n = 100) and training (n=89) set with an equal distribution of ADHD and Control. The resulting SVM performed with only 44% accuracy, worse than chance.


	[[Image:6.png]]		[[Image:6.png]]
			Structure of features data matrix


	The poor performance of the classifier called for further exploration as to if it is possible to separate ADHD and Control. Data was transformed with quadratic, absolute value, logarithmic, sigmoid, and exponential functions to explore differences in distributions of ADHD and Control. Further, each transformation was decomposed with singular value decomposition. Results of the transformations were used to train and test the classifier, and results were not better than chance. An example exploration of a logarithmic transformation is shown below. The plots on the far left show the different distributions between ADHD (Group 1, top) and Control (Group 2, bottom), the middle plots show the features (voxel values) for each subject (on the x axis), and the final plots are the total and first six components produced by singular value decomposition. For each visualization, the goal of the exploration would be to find a transformation that makes the data (ADHD vs Control) very different so it might be separable with an SVM.		The poor performance of the classifier called for further exploration as to if it is possible to separate ADHD and Control. Data was transformed with quadratic, absolute value, logarithmic, sigmoid, and exponential functions to explore differences in distributions of ADHD and Control. Further, each transformation was decomposed with singular value decomposition. Results of the transformations were used to train and test the classifier, and results were not better than chance. An example exploration of a logarithmic transformation is shown below. The plots on the far left show the different distributions between ADHD (Group 1, top) and Control (Group 2, bottom), the middle plots show the features (voxel values) for each subject (on the x axis), and the final plots are the total and first six components produced by singular value decomposition. For each visualization, the goal of the exploration would be to find a transformation that makes the data (ADHD vs Control) very different so it might be separable with an SVM.


	[[Image:7.png]]		[[Image:7.png]]
	Comparing ADHD (top) vs (control) for 1) distribution of values 2) features, and 3) svd components		Comparing ADHD (top) vs (control) for 1) distribution of values 2) features, and 3) svd components


	While there were slight differences in the plots for the quadratic and logarithmic transformation, these new distributions used to train and test the classifier only improved performance moderately (44% accuracy for original Z scores to 46% accuracy for transformations). To further explore the possibility of the data being different enough to be separable with a SVM, correlation coefficients were calculated for each comparison between ADHD and ADHD, ADHD and Control, and Control and Control, for each of the distributions. True difference in ADHD vs. Control would be represented by a dark region (low values indicate a low correlation coefficient, which means that the vector of features between the two subjects is more different) in the square in the top right (comparing ADHD and control) as compared to the triangles showing within-group differences (top left and bottom right). As an example, the logarithmic correlation matrix is shown below, demonstrating no visual difference between ADHD and control. This pattern was true for all transformations, leading to the conclusion that these particular features are not different enough for an SVM to use to classify.		While there were slight differences in the plots for the quadratic and logarithmic transformation, these new distributions used to train and test the classifier only improved performance moderately (44% accuracy for original Z scores to 46% accuracy for transformations). To further explore the possibility of the data being different enough to be separable with a SVM, correlation coefficients were calculated for each comparison between ADHD and ADHD, ADHD and Control, and Control and Control, for each of the distributions. True difference in ADHD vs. Control would be represented by a dark region (low values indicate a low correlation coefficient, which means that the vector of features between the two subjects is more different) in the square in the top right (comparing ADHD and control) as compared to the triangles showing within-group differences (top left and bottom right). As an example, the logarithmic correlation matrix is shown below, demonstrating no visual difference between ADHD and control. This pattern was true for all transformations, leading to the conclusion that these particular features are not different enough for an SVM to use to classify.


	[[Image:8.png]]		[[Image:8.png]]

	Matrix of correlation coefficients demonstrates no visual difference between ADHD vs Control (box in top right) as compared to ADHD vs ADHD (triangle top left) and Control vs Control (triangle bottom right)		Matrix of correlation coefficients demonstrates no visual difference between ADHD vs Control (box in top right) as compared to ADHD vs ADHD (triangle top left) and Control vs Control (triangle bottom right)


	===Discussion and Future Work===		===Discussion and Future Work===

imported>Psych204B at 22:13, 10 March 2012

2012-03-10T22:13:41Z

← Older revision		Revision as of 22:13, 10 March 2012
Line 106:		Line 106:
	====Additional Formats====		====Additional Formats====
	* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.ppt PPT Presentation]		* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.ppt PPT Presentation]
	* [http://www.vbmis.com/bmi/class/psych204b/~~FINALICAppt~~.pdf PPT Presentation PDF]		* [http://www.vbmis.com/bmi/class/psych204b/finalica.pdf PPT Presentation PDF]
	* [http://www.vbmis.com/bmi/class/psych204b/finalica.pdf Final Writeup]
	* [http://www.vbmis.com/bmi/class/psych204b/finalica html Presentation]		* [http://www.vbmis.com/bmi/class/psych204b/finalica html Presentation]

imported>Psych204B at 22:11, 10 March 2012

2012-03-10T22:11:29Z

← Older revision		Revision as of 22:11, 10 March 2012
Line 76:		Line 76:

	==CLASSIFICATION==		==CLASSIFICATION==
	To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This ~~resulted~~ in an n x p matrix, with n rows of subjects and p ~~rows~~ of ~~voxels~~. ~~A support vector machine~~ was used with a ~~linear kernel~~. **(~~More results~~ to ~~be posted~~, ~~this part~~ of ~~analysis still~~ in ~~progress~~)		To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This mask was used to extract Z score values from individual DMN maps to result in an n x p matrix, with n rows of subjects and p columns of features. The data was split into a test (n = 100) and training (n=89) set with an equal distribution of ADHD and Control. The resulting SVM performed with only 44% accuracy, worse than chance.

			[[Image:6.png]]

			The poor performance of the classifier called for further exploration as to if it is possible to separate ADHD and Control. Data was transformed with quadratic, absolute value, logarithmic, sigmoid, and exponential functions to explore differences in distributions of ADHD and Control. Further, each transformation was decomposed with singular value decomposition. Results of the transformations were used to train and test the classifier, and results were not better than chance. An example exploration of a logarithmic transformation is shown below. The plots on the far left show the different distributions between ADHD (Group 1, top) and Control (Group 2, bottom), the middle plots show the features (voxel values) for each subject (on the x axis), and the final plots are the total and first six components produced by singular value decomposition. For each visualization, the goal of the exploration would be to find a transformation that makes the data (ADHD vs Control) very different so it might be separable with an SVM.

			[[Image:7.png]]
			Comparing ADHD (top) vs (control) for 1) distribution of values 2) features, and 3) svd components

			While there were slight differences in the plots for the quadratic and logarithmic transformation, these new distributions used to train and test the classifier only improved performance moderately (44% accuracy for original Z scores to 46% accuracy for transformations). To further explore the possibility of the data being different enough to be separable with a SVM, correlation coefficients were calculated for each comparison between ADHD and ADHD, ADHD and Control, and Control and Control, for each of the distributions. True difference in ADHD vs. Control would be represented by a dark region (low values indicate a low correlation coefficient, which means that the vector of features between the two subjects is more different) in the square in the top right (comparing ADHD and control) as compared to the triangles showing within-group differences (top left and bottom right). As an example, the logarithmic correlation matrix is shown below, demonstrating no visual difference between ADHD and control. This pattern was true for all transformations, leading to the conclusion that these particular features are not different enough for an SVM to use to classify.

			[[Image:8.png]]
			Matrix of correlation coefficients demonstrates no visual difference between ADHD vs Control (box in top right) as compared to ADHD vs ADHD (triangle top left) and Control vs Control (triangle bottom right)

	===Discussion and Future Work===		===Discussion and Future Work===
			This analysis has shown results approaching significance for the strength of connectivity of the default mode network between a set of individuals with ADHD and healthy controls. Attempts to build an elementary classifier with support vector machines based on these DMN maps were not successful, however, given the importance in the choice of features, it cannot be concluded that classification is impossible.

			Statistical Z values from the thresholded component maps were used as a representation of the strength of contribution of any particular voxel to a resting state network, and it would be suggested to instead use filtered data without performing independent component analysis that is masked by the binarized DMN map (produced from the ICA analysis in the same manner). It would be interesting to attempt to incorporate spatial data, perhaps using K-Means clustering of voxel coordinates or a k-nearest neighbor approach to identify spatially similar voxels, and then using the Z score as an additional parameter to the distance metric to account for related voxels (adding 1/Z score, for example, would make voxels with a higher Z score rated as “more similar” as represented by smaller values).

	The DMN was originally selected because it has been shown to be aberrant across many disorders, and because it is a relatively easy network to identify out of the many components. It is salient, however, that this same analysis done with an attention network might be more relevant to ADHD. I did not have the time to perform all of the analysis over again with a different network, however I want to note that it is something that I am going to pursue.		The DMN was originally selected because it has been shown to be aberrant across many disorders, and because it is a relatively easy network to identify out of the many components. It is salient, however, that this same analysis done with an attention network might be more relevant to ADHD. I did not have the time to perform all of the analysis over again with a different network, however I want to note that it is something that I am going to pursue.

	~~My long term goals in this field pertain~~ to ~~classification~~ of ~~large, publicly available datasets~~. ~~With the~~ goal ~~of identifying biomarkers of disease to aid with diagnosis, what~~ would be ~~useful is~~ to develop ~~pipelines~~ that ~~take raw resting BOLD data~~ for different groups~~, filters~~ and ~~derives individual networks~~, ~~and then matches them to templates, derives differences~~. ~~A classifier might then~~ be ~~built~~ to ~~distinguish between normal and~~ disease~~, and used to diagnose a novel dataset~~. While one lone classifier probably is not good enough to be used to classify ~~this~~ new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.		Lastly, singular value decomposition of different transformations of the data might be pursued further to create specific subsets of features for the classifier. The long term goal would be to develop a data-driven analysis pipeline that produces output to create many different classifiers (for different groups of disorder versus healthy control) to classify a new dataset. This long term goal fits well with the growing trend of labs and medical institutions releasing large, publicly available datasets. Such a pipeline would be used to build classifiers to identify biomarkers of disease. While one lone classifier probably is not good enough to be used to classify a new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.

	~~[test]~~

	===Appendix===		===Appendix===
			====Presentation====
			* [http://www.vbmis.com/bmi/class/psych204b/finalica.swf Flash Presentation]
			* [http://www.vbmis.com/bmi/class/psych204b/FinalProjectOverview.pdf Final Writeup]

			====Additional Formats====
			* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.ppt PPT Presentation]
			* [http://www.vbmis.com/bmi/class/psych204b/FINALICAppt.pdf PPT Presentation PDF]
			* [http://www.vbmis.com/bmi/class/psych204b/finalica.pdf Final Writeup]
			* [http://www.vbmis.com/bmi/class/psych204b/finalica html Presentation]

			====Script====
	* [https://github.com/vsoch/ica-/blob/master/melodic_ss.sh Single Subject Processing]		* [https://github.com/vsoch/ica-/blob/master/melodic_ss.sh Single Subject Processing]
	* [https://github.com/vsoch/ica-/blob/master/melodic_gp.sh Group Processing]		* [https://github.com/vsoch/ica-/blob/master/melodic_gp.sh Group Processing]
Line 92:		Line 117:
	* [https://github.com/vsoch/ica-/blob/master/melodic_qa.py Quality Analysis]		* [https://github.com/vsoch/ica-/blob/master/melodic_qa.py Quality Analysis]
	* [https://github.com/vsoch/ica-/blob/master/MRTools.py MR Tools (used by Template matching)]		* [https://github.com/vsoch/ica-/blob/master/MRTools.py MR Tools (used by Template matching)]
			* [https://github.com/vsoch/fvplay Data Exploration Tools**]
	* [http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m neuroGLMnet - extract data --> SVM]		* [http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m neuroGLMnet - extract data --> SVM]
	* [http://www.vbmis.com/bmi/class/psych204b/plotty.m plotty - also for SVM exploration]		* [http://www.vbmis.com/bmi/class/psych204b/plotty.m plotty - also for SVM exploration]
	* [https://github.com/vsoch/fvplay Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!		** I am working on this last set of scripts with Samir!
	** Also note that these versions online are not the exact scripts that I used... each was modified pretty extensively on my local machine to account for different data and processing (for example, bandpass and not highpass, different parameters, etc)		** Also note that these versions online are not the exact scripts that I used... each was modified pretty extensively on my local machine to account for different data and processing (for example, bandpass and not highpass, different parameters, etc)

imported>Psych204B at 05:24, 7 March 2012

2012-03-07T05:24:08Z

← Older revision		Revision as of 05:24, 7 March 2012
Line 95:		Line 95:
	* [http://www.vbmis.com/bmi/class/psych204b/plotty.m plotty - also for SVM exploration]		* [http://www.vbmis.com/bmi/class/psych204b/plotty.m plotty - also for SVM exploration]
	* [https://github.com/vsoch/fvplay Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!		* [https://github.com/vsoch/fvplay Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!
	** Also note that these versions online are not the exact scripts that I used... each was modified on my local machine!		** Also note that these versions online are not the exact scripts that I used... each was modified pretty extensively on my local machine to account for different data and processing (for example, bandpass and not highpass, different parameters, etc)

	===REFERENCES===		===REFERENCES===

imported>Psych204B at 05:22, 7 March 2012

2012-03-07T05:22:06Z

← Older revision		Revision as of 05:22, 7 March 2012
Line 86:		Line 86:

	===Appendix===		===Appendix===
	* [https://github.com/vsoch/ica-/blob/master/melodic_ss.sh\|Single Subject Processing]		* [https://github.com/vsoch/ica-/blob/master/melodic_ss.sh Single Subject Processing]
	* [https://github.com/vsoch/ica-/blob/master/melodic_gp.sh\|Group Processing]		* [https://github.com/vsoch/ica-/blob/master/melodic_gp.sh Group Processing]
	* [https://github.com/vsoch/ica-/blob/master/ica+.py\|Submission Script]		* [https://github.com/vsoch/ica-/blob/master/ica+.py Submission Script]
	* [https://github.com/vsoch/ica-/blob/master/pyMatch.py\|Template Matching Script]		* [https://github.com/vsoch/ica-/blob/master/pyMatch.py Template Matching Script]
	* [https://github.com/vsoch/ica-/blob/master/melodic_qa.py\|Quality Analysis]		* [https://github.com/vsoch/ica-/blob/master/melodic_qa.py Quality Analysis]
	* [https://github.com/vsoch/ica-/blob/master/MRTools.py\|MR Tools (used by Template matching)]		* [https://github.com/vsoch/ica-/blob/master/MRTools.py MR Tools (used by Template matching)]
	* [http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m\|neuroGLMnet - extract data --> SVM]		* [http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m neuroGLMnet - extract data --> SVM]
	* [http://www.vbmis.com/bmi/class/psych204b/plotty.m\|plotty - also for SVM exploration]		* [http://www.vbmis.com/bmi/class/psych204b/plotty.m plotty - also for SVM exploration]
	* [https://github.com/vsoch/fvplay\|Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!		* [https://github.com/vsoch/fvplay Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!
	** Also note that these versions online are not the exact scripts that I used... each was modified on my local machine!		** Also note that these versions online are not the exact scripts that I used... each was modified on my local machine!
	===REFERENCES===		===REFERENCES===

imported>Psych204B at 05:20, 7 March 2012

2012-03-07T05:20:27Z

imported>Psych204B at 05:18, 7 March 2012

2012-03-07T05:18:02Z

← Older revision		Revision as of 05:18, 7 March 2012
Line 84:		Line 84:

	===Appendix===		===Appendix===
	[[~~Vanessa~~ - ~~ICA and Preprocessing Scripts~~]]		[[https://github.com/vsoch/ica-/blob/master/melodic_ss.sh\|Single Subject Processing]] \\
	[[~~Vanessa~~ - Template Matching]]		[[https://github.com/vsoch/ica-/blob/master/melodic_gp.sh\|Group Processing]] \\
	[[~~Vanessa~~ - ~~MATLAB Scripts~~, ~~Data Extraction and Classification~~]]		[[https://github.com/vsoch/ica-/blob/master/ica+.py\|Submission Script]] \\
			[[https://github.com/vsoch/ica-/blob/master/pyMatch.py\|Template Matching Script]] \\
			[[https://github.com/vsoch/ica-/blob/master/melodic_qa.py\|Quality Analysis]] \\
			[[https://github.com/vsoch/ica-/blob/master/MRTools.py\|MR Tools (used by Template matching)]] \\
			[[http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m\|neuroGLMnet - extract data --> SVM]] \\
			[[http://www.vbmis.com/bmi/class/psych204b/plotty.m\|plotty - also for SVM exploration]] \\
			[[https://github.com/vsoch/fvplay\|Classification Tool (so far, just play!)]] \\
			** I am working on this last set of scripts with Samir!

	===REFERENCES===		===REFERENCES===

imported>Psych204B at 05:02, 7 March 2012

2012-03-07T05:02:03Z

← Older revision		Revision as of 05:02, 7 March 2012
Line 78:		Line 78:
	To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This resulted in an n x p matrix, with n rows of subjects and p rows of voxels. A support vector machine was used with a linear kernel. **(More results to be posted, this part of analysis still in progress)		To perform classification of disease based on the default mode network, a 1 sample T-test was done for all participants, and the resulting image saved, masked to gray matter, and binarized to represent a DMN template for the entire group. This resulted in an n x p matrix, with n rows of subjects and p rows of voxels. A support vector machine was used with a linear kernel. **(More results to be posted, this part of analysis still in progress)

	===Future Work===		===Discussion and Future Work===
	The DMN was originally selected because it has been shown to be aberrant across many disorders, and because it is a relatively easy network to identify out of the many components. It is salient, however, that this same analysis done with an attention network might be more relevant to ADHD. I did not have the time to perform all of the analysis over again with a different network, however I want to note that it is something that I am going to pursue.		The DMN was originally selected because it has been shown to be aberrant across many disorders, and because it is a relatively easy network to identify out of the many components. It is salient, however, that this same analysis done with an attention network might be more relevant to ADHD. I did not have the time to perform all of the analysis over again with a different network, however I want to note that it is something that I am going to pursue.

	My long term goals in this field pertain to classification of large, publicly available datasets. With the goal of identifying biomarkers of disease to aid with diagnosis, what would be useful is to develop pipelines that take raw resting BOLD data for different groups, filters and derives individual networks, and then matches them to templates, derives differences. A classifier might then be built to distinguish between normal and disease, and used to diagnose a novel dataset. While one lone classifier probably is not good enough to be used to classify this new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.		My long term goals in this field pertain to classification of large, publicly available datasets. With the goal of identifying biomarkers of disease to aid with diagnosis, what would be useful is to develop pipelines that take raw resting BOLD data for different groups, filters and derives individual networks, and then matches them to templates, derives differences. A classifier might then be built to distinguish between normal and disease, and used to diagnose a novel dataset. While one lone classifier probably is not good enough to be used to classify this new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.

			===Appendix===
			[[Vanessa - ICA and Preprocessing Scripts]]
			[[Vanessa - Template Matching]]
			[[Vanessa - MATLAB Scripts, Data Extraction and Classification]]

	===REFERENCES===		===REFERENCES===

imported>Psych204B at 04:57, 7 March 2012

2012-03-07T04:57:44Z

← Older revision		Revision as of 04:57, 7 March 2012
Line 47:		Line 47:
	The default mode network (DMN) was manually identified from the group networks (below, left) based on visual matching to verified ventral and dorsal DMN maps produced by the Grecius lab at Stanford (below, right). The network was also verified by a postdoc in the Stanford Cognitive and Systems Neuroscience Laboratory.		The default mode network (DMN) was manually identified from the group networks (below, left) based on visual matching to verified ventral and dorsal DMN maps produced by the Grecius lab at Stanford (below, right). The network was also verified by a postdoc in the Stanford Cognitive and Systems Neuroscience Laboratory.

	[[Image:~~3DMN~~.png]]		[[Image:3_DMN.png]]

	Group derived DMN component (left), as compared to templates from Grecius Lab (right)		Group derived DMN component (left), as compared to templates from Grecius Lab (right)

← Older revision		Revision as of 05:20, 7 March 2012
Line 83:		Line 83:
	My long term goals in this field pertain to classification of large, publicly available datasets. With the goal of identifying biomarkers of disease to aid with diagnosis, what would be useful is to develop pipelines that take raw resting BOLD data for different groups, filters and derives individual networks, and then matches them to templates, derives differences. A classifier might then be built to distinguish between normal and disease, and used to diagnose a novel dataset. While one lone classifier probably is not good enough to be used to classify this new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.		My long term goals in this field pertain to classification of large, publicly available datasets. With the goal of identifying biomarkers of disease to aid with diagnosis, what would be useful is to develop pipelines that take raw resting BOLD data for different groups, filters and derives individual networks, and then matches them to templates, derives differences. A classifier might then be built to distinguish between normal and disease, and used to diagnose a novel dataset. While one lone classifier probably is not good enough to be used to classify this new ICA run, the technique of “boosting” could be used with multiple classifiers to come to an informed diagnosis. This technique can be thought of as taking a “majority vote.” Further extension of this method would be to extract relevant features from these maps to build a database with supporting infrastructure to query semantically accessible results, however this is more of a problem for bioinformatics than neuroscience.

			[test]

	===Appendix===		===Appendix===
	[[https://github.com/vsoch/ica-/blob/master/melodic_ss.sh\|Single Subject Processing]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/melodic_ss.sh\|Single Subject Processing]
	[[https://github.com/vsoch/ica-/blob/master/melodic_gp.sh\|Group Processing]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/melodic_gp.sh\|Group Processing]
	[[https://github.com/vsoch/ica-/blob/master/ica+.py\|Submission Script]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/ica+.py\|Submission Script]
	[[https://github.com/vsoch/ica-/blob/master/pyMatch.py\|Template Matching Script]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/pyMatch.py\|Template Matching Script]
	[[https://github.com/vsoch/ica-/blob/master/melodic_qa.py\|Quality Analysis]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/melodic_qa.py\|Quality Analysis]
	[[https://github.com/vsoch/ica-/blob/master/MRTools.py\|MR Tools (used by Template matching)]~~] \\~~		* [https://github.com/vsoch/ica-/blob/master/MRTools.py\|MR Tools (used by Template matching)]
	[[http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m\|neuroGLMnet - extract data --> SVM]~~] \\~~		* [http://www.vbmis.com/bmi/class/psych204b/neuroGLMNET.m\|neuroGLMnet - extract data --> SVM]
	[[http://www.vbmis.com/bmi/class/psych204b/plotty.m\|plotty - also for SVM exploration]~~] \\~~		* [http://www.vbmis.com/bmi/class/psych204b/plotty.m\|plotty - also for SVM exploration]
	[[https://github.com/vsoch/fvplay\|Classification Tool (so far, just play!)]~~] \\~~		* [https://github.com/vsoch/fvplay\|Classification Tool (so far, just play!)] ** I am working on this last set of scripts with Samir!
	** I am working on this last set of scripts with Samir!		** Also note that these versions online are not the exact scripts that I used... each was modified on my local machine!

	===REFERENCES===		===REFERENCES===