TY - GEN
T1 - Eliciting tacit expertise in 3D volume segmentation
AU - West, Ruth
AU - Kajihara, Meghan
AU - Parola, Max
AU - Holloway, Michelle
AU - Hays, Kathryn
AU - Hillard, Luke
AU - Carlew, Anne
AU - Deutsch, Jeremey
AU - Lane, Brandon
AU - John, Brendan
AU - Sanandaji, Anahita
AU - Grimm, Cindy
N1 - Publisher Copyright:
© 2016 Copyright.
PY - 2016/9/24
Y1 - 2016/9/24
N2 - The output of 3D volume segmentation is crucial to a wide range of endeavors. Producing accurate segmentations often proves to be both ineficient and challenging, in part due to lack of imaging data quality (contrast and resolution), and because of ambiguity in the data that can only be resolved with higher-level knowledge of the structure and the context wherein it resides. Automatic and semi-Automatic approaches are improving, but in many cases still fail or require substantial manual clean-up or intervention. Expert manual segmentation and review is therefore still the gold standard for many applications. Unfortunately, existing tools (both custom-made and commercial) are often designed based on the underlying algorithm, not the best method for expressing higher-level intention. Our goal is to analyze manual (or semi-Automatic) segmentation to gain a better understanding of both low-level (perceptual tasks and actions) and high-level decision making. This can be used to produce segmentation tools that are more accurate, effcient, and easier to use. Questioning or observation alone is insu ffcient to capture this information, so we utilize a hybrid capture protocol that blends observation, surveys, and eye tracking. We then developed, and validated, data coding schemes capable of discerning low-level actions and overall task structures.
AB - The output of 3D volume segmentation is crucial to a wide range of endeavors. Producing accurate segmentations often proves to be both ineficient and challenging, in part due to lack of imaging data quality (contrast and resolution), and because of ambiguity in the data that can only be resolved with higher-level knowledge of the structure and the context wherein it resides. Automatic and semi-Automatic approaches are improving, but in many cases still fail or require substantial manual clean-up or intervention. Expert manual segmentation and review is therefore still the gold standard for many applications. Unfortunately, existing tools (both custom-made and commercial) are often designed based on the underlying algorithm, not the best method for expressing higher-level intention. Our goal is to analyze manual (or semi-Automatic) segmentation to gain a better understanding of both low-level (perceptual tasks and actions) and high-level decision making. This can be used to produce segmentation tools that are more accurate, effcient, and easier to use. Questioning or observation alone is insu ffcient to capture this information, so we utilize a hybrid capture protocol that blends observation, surveys, and eye tracking. We then developed, and validated, data coding schemes capable of discerning low-level actions and overall task structures.
KW - 3D volume segmentation
KW - Conceptual framework
UR - http://www.scopus.com/inward/record.url?scp=85010950710&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85010950710&partnerID=8YFLogxK
U2 - 10.1145/2968220.2968235
DO - 10.1145/2968220.2968235
M3 - Conference contribution
AN - SCOPUS:85010950710
T3 - ACM International Conference Proceeding Series
SP - 59
EP - 66
BT - VINCI 2016 - 9th International Symposium on Visual Information Communication and Interaction
A2 - Zhang, Kang
A2 - Kerren, Andreas
PB - Association for Computing Machinery
T2 - 9th International Symposium on Visual Information Communication and Interaction, VINCI 2016
Y2 - 24 September 2016 through 26 September 2016
ER -