Anoto Medical Image Annotator: Interactive Prototype and Progress Report1. Member Names and RolesGroup Name: The AnnototatorsAnirudh Vemprala: User interface design and programmingEdward Karuna: User interface design, programming, and testing. Gene Zhang: User interface design, programming, and testing.Robert Held: Paper interface design and programming. Report and presentation.1. Problem and Solution OverviewRadiologists and medical imaging researchers work in a field that demands constant, careful evaluation of complex images. The ability to accurately evaluate a medical image and detect important features, whether they are broken bones or metastasized tumors, is tantamount to success in radiology. Therefore, it is important for radiologists and researchers to have tools that complement their abilities by making it easy to record observations and sharing them with others. Currently, radiologists have limited on-screen tools to select regions of interest and share their observations with colleagues. Most people review medical images by opening the files on their respective computers, making personal notes, and then presenting their observations in a group setting. The solution to the current, inefficient system is an integrated approach that allows one to attach comments to a medical image and permit others to add their own observations. The spirit of collaboration could also be extended to include a student-teacher relationship. That is, students could indicate what they believe to be important features within an image, and then the teacher could review the students’ work, indicate missed features, and provide tips for future analyses. The solution should also include the ability to create the observations directly on a printout, allowing the user to use a pen to circle features and jot notes and avoid the relatively clumsy use of a mouse for the same actions. The Anoto digital pen and its range of abilities are well-suited for such a solution. We have named our solution the “Anoto Medical Imaging Annotators (AMIA)”.2. TasksThe following tasks outline the key functionality provided by our system.Task 1 (Easy): Login and load previous annotations for viewing.In order to facilitate the sharing features of the program, users will need to be uniquely identified. The user login will also be crucial for the teacher-student relationship, since students will not be able to view their peers’ reviews of a given image if the program is being used to evaluate their performance. The teacher, on the other hand, would be given access to everyone’s files in order to deliver his/her comments.When the program loads up, a sign-in dialogue box will appear and ask the user for his/her screen name and password. The task then demands that the user load a previous set ofannotations, an action which will be necessary in order to add comments to other users’ notes for a given set of medical images. The task should be a straightforward process.Task 2 (Medium): Add text comments to another person’s annotationsTo collaborate on the review of patient’s medical images, multiple users must be able to comment on the same images. The program includes a text box for providing additional comments on top of those already recorded using the anoto pen. The task involves selecting an image with pre-existing comments, selecting the text box, adding a few lines of text, and committing the changes to the file using an “add comments” button.Task 3 (Hard): Print and annotate a set of image.The core functionality of AMIA lies in its use of the anoto digital pen. Therefore, it was crucial to include a task that required the user to print out a set of medical images, add comments, and register those comments with the program. The task is accomplished by reviewing a medical image set, clicking the “Print” button, selecting the images to be annotated, retrieving the printed images, using the pen to indicate comments and various shapes on the images, and then verifying that the strokes were loaded into the computer and associated with the correct images.3. Revised Interface DesignResults of Contextual InquiryWe learned from the contextual inquiry assignment that our target users (radiologists and medical imaging researchers) would like a quick, intuitive way to choose regions of interest (ROI’s). Our target users work with many medical images on a regular basis, so any device or system that can improve efficiency and reduce the time spent on each image would be welcomed. Additionally, a system for the simultaneous review of resident sample analyses of images would enhance an instructing radiologist’s ability to assess their findings and return helpful comments. One interviewee emphasized her desire for a quicker, more direct way to annotate medical images in general. We also learned that radiologists typically assess images in a dedicating reading room that includes a handful of computer stations. They also share their findings with their patients in their offices, and may verbally share findings in group meetings. Researchers process medical images at the point of acquisition, at lab workstations, at their desktop computers, and even on their laptops. Sharing of written observations was therefore lacking. Based on the preceding information, the interface we had in mind was one that resembled a multi-document image viewer. We thought it would be wise to include a preview pane, would allow quick selection of image sets from within a given directory. The main document space would present its user with the medical images under consideration, along with representations of layers that store the annotations made to the images. A slidebar at the bottom of the screen would allow the user to rapidly scan through the slices within the imaging set. For a given image, a set of layer thumbnails would be displayed near the bottom of the screen. Each layer would represent comments from a different user. By selecting one of the thumbnails, one could place that layer over the original image in the main window. A "Comment" button would then create a dialog box where the user can add his/her comments about the currently viewed layer. These comments would then be attached to that layer's metadata, and
View Full Document
Unlocking...