Information Flow Summary

1. A new user is added.  The user is assigned a password, and given permissions to access specified regions of the TMAJ program.
   
2. Specimens are put in the database.  This can happen in two ways.
   
   A) Specimens can be entered by going into the input application and manually entering a Specimen.  This is done by opening the Specimens Application, going to the Specimens tab, and clicking the Add button.  The patients table is ignored, and the foreign key PatientID in the Specimens table is given a default value.  This is the most common way data input starts off in TMAJ. 
   
   Admin Steps: The user must be given permission to access the Specimens Application before this can happen.
   
   B) Alternatively, the user can import a tab-delimited file in the Import Application.  If the user were at hopkins, the tab-delimited file would probably come from PDS (Pathology Data Systems), a centralized Johns Hopkins data system for storing specimens that come out of Surgical Pathology.  However, the Import application can be configured for different types of sources.  In a tab-delimted PDS file, each row represents a case.  A case is a combination of a patient and a patient's specimen.  In the specimen-import, both the Patients table and Specimens table are populated.  This is the only way the patients table is modified in TMAJ.  Also, a user can import a tab-delimited file of specimens.  This tab-delimited file is obtained from PDS, a centralized Hopkins data system for storing specimen information.  Note: This is a weaker area of TMAJ, and we intend to upgrade to a more robust system using XML to do imports.
   
3. Blocks and Tissue Diagnosis's (Sub-Blocks) are put in the database.  Blocks and TissueDiagnosis's must be manually entered through the Specimens application.
4. A new ArrayBlock is entered. This is done in the ArrayBlock Tab of the ArrayManager application.
   
5. The ArrayBlock is Designed.  The arrayblock is designed in the ArrayBuilder application.  First the user chooses an ArrayBlock.  Next, for each coordinate in the arrayBlock, the user assigns a tissueDiagnosis. 
   
   Note: Steps 1-3 are often combined into one with the "Express Entry" feature.  Basically the user just selects a coordinate, enters minimum information about the specimen, block, and TissueDiagnosis, and all 3 are created on the fly, with the generated TissueDiagnosisID being assigned to the coordinate.
   
   Admin Steps: The user must be given permission to the ArrayBlock.  Open the Admin application, and then click on the UsersArrayBlocks tab.
   
6. An ArraySlide from the ArrayBlock is cut and scanned.  This step is physical in nature, and is performed by the tma technicians.  An ArraySlide is cut along the z-axis on an ArrayBlock.  The ArraySlide is scanned using a scanning machine and the images are dumped into a directory on the bui1 machine.  A scan of an ArraySlide will output 1 directory, with about 400  jpegs inside it (the arrays are often 20 by 20).  Each JPEG file represents a coordinate on the ArraySlide, and you will see an X and Y position number in each of the image filenames. 
   
7. The jpeg images in the directory just created are imported into TMAJ.  Go to the "ArrayImages" tab on the Import application.  Choose the directory that contains the ArraySlide's scanned images, as mentioned in the previous step.  Make sure that:
   * The directory is in the appropriate place (in the [[main images directory]])
   * The directory is named using the scan standard format. 
   This program populates the ArrayImages table. 
   
   Optional Step: Sometimes you will be asked to import machine-scored Data.  Machine-Scored Data are scores that are automatically assigned to images by the scanning machines themselves.  For example, this data might be how much "blue" or "brown" there is in an image.  To do this, click on the "MachineScoredData" button in the import application, and choose the file that contains the tab-delimited machne-scored data.  Make sure the filename is in scan standard format. Machine-Scored data will show up in the DataPanel and InfoPanel in the Images application.
   
8. The ArraySlide is assigned to a Project. If necessary, a new Project is created.  Note: You may skip this step if you added the ArraySlide to a project while doing the import. 
   A project is a collection of ArraySlides.  In addition, there is a UsersProjects table that links Users to Projects.  Thus, Users are linked to Projects, and Projects are linked to ArraySlides--in effect, an administrator can control what Users can see what ArraySlides.   A new project can be created in the Projects tab of the Admin application.  Users are assigned to that project in the UsersProjects tab.  ArraySlides are assigned to that project is the Projects-ArraySlides tab.  The step is always performed by a TMAJ administrator. 
   
9. Create a Session.  A user logs on to the image manager application.  If the end-user hasn't done so already, they must first create a new session.  A session is a combination of a Scan and a User, and represents a user's scoring of an scanned ArraySlide.  Thus, a scan can be scored multiple times, by different users, at different times.   A user can even decide to rescore an scanned ArraySlide with a different session, thus keeping the scores for both sessions.
   
   Admin Steps: You must have created the user and given permission to access the Images application. 
   
   
   
10. End-Users assign their own scores to the images.    A user then double-clicks on the appropriate session to open the ArraySlide and the current ArrayImage, on position x=1, y=1. The user presses the next button until all the images are scored.
    
11. Session is published.  After a user has finished scoring a session, the session may be published so that the data can no longer be changed, and that all users may see.  Alternatively, a user may finalize the data, which ensures that the data will not be modified.
    
    
    

© Copyright 2009 | All Rights Reserved | The Johns Hopkins University