Demonstrating the electronic cocktail napkin: a paper-like interface for early design
Conference Companion on Human Factors in Computing Systems
Sketching for knowledge capture: a progress report
Proceedings of the 7th international conference on Intelligent user interfaces
Sketching for military courses of action diagrams
Proceedings of the 8th international conference on Intelligent user interfaces
Structure in On-line Documents
ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Discerning Structure from Freeform Handwritten Notes
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
A collaborative intelligent tutoring system for medical problem-based learning
Proceedings of the 9th international conference on Intelligent user interfaces
Sketch-based retrieval of ClipArt drawings
Proceedings of the working conference on Advanced visual interfaces
Distinguishing Text from Graphics in On-Line Handwritten Ink
IWFHR '04 Proceedings of the Ninth International Workshop on Frontiers in Handwriting Recognition
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
COMET: A Collaborative Tutoring System for Medical Problem-Based Learning
IEEE Intelligent Systems
Anatomical sketch understanding: recognizing explicit and implicit structure
AIME'05 Proceedings of the 10th conference on Artificial Intelligence in Medicine
Hi-index | 0.00 |
Note taking is a common way for physicians to collect information from their patients in medical inquiries and diagnoses. Many times, when describing the pathology in medical records, a physician also draws diagrams and/or anatomical sketches along with the free-text narratives. The ability to understand unstructured handwritten texts and drawings in patient record could lead to implementation of automated patient record systems with more natural interfaces than current highly structured systems. The first and crucial step in automated processing of free-hand medical records is to segment the record into handwritten text and drawings, so that appropriate recognizers can be applied to different regions. This paper presents novel algorithms that separate text from non-text strokes in an on-line handwritten patient record. The algorithm is based on analyses of spatio-temporal graphs extracted from an on-line patient record and support vector machine (SVM) classification. Experiments demonstrate that the proposed approach is effective and robust.