TY - JOUR
T1 - A novel enhanced energy function using augmented reality for a bowel
T2 - modified region and weighted factor
AU - Shrestha, Ganesh
AU - Alsadoon, Abeer
AU - Prasad, P. W.C.
AU - Al-Dala’in, Thair
AU - Alrubaie, Ahmad
N1 - Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature.
PY - 2021/5
Y1 - 2021/5
N2 - The popularity of augmented reality in medical application is rising exponentially over time, especially in the medical sector. It possesses a greater possibility in the reduction of surgical risks by raising visual awareness during the operation. Incorrect representation or inadequate detail on target region or delay in processing time may result in serious consequences. Also, the absence of a proper mechanism to handle occlusions like nerves, vessels or medical equipment may affect the performance. Therefore, this research aims to improve the visualization accuracy of bowel region and reduce the processing time. a novel enhanced energy function using augmented reality for bowel is proposed. The proposed system targets the modified region and weighted factor that encompasses the power of the combined region and dense cue with longterm and accurate augmented reality display mechanism. The system is capable of providing detailed visual output by precisely placing the model developed using the CT images of the target object, over the live video. Also, by applying the least square approach, the system is capable of addressing larger deformation and occlusion that appears during the surgical procedure providing the most accurate display. The feature tracking and tracking recovery components help the entire visualization procedure to stay on track by automatically registering and re-registering the surface whenever required. The proposed system capable of running image registration without human involvement and it can even decide when to trigger the reregistration process whenever required. The results from the proposed system has minimized the overlay error by a larger number. We validated the system with different sets of samples from endoscopy. The dataset included the samples from the bowel region from people with three different age groups. The overlay error accuracy was 0.24777px, and the performance was 44fps. The proposed system is concentrated on the overlay accuracy and the processing time. This study has addressed the shortcoming of the previous systems regarding manual registration and rigid assumptions.
AB - The popularity of augmented reality in medical application is rising exponentially over time, especially in the medical sector. It possesses a greater possibility in the reduction of surgical risks by raising visual awareness during the operation. Incorrect representation or inadequate detail on target region or delay in processing time may result in serious consequences. Also, the absence of a proper mechanism to handle occlusions like nerves, vessels or medical equipment may affect the performance. Therefore, this research aims to improve the visualization accuracy of bowel region and reduce the processing time. a novel enhanced energy function using augmented reality for bowel is proposed. The proposed system targets the modified region and weighted factor that encompasses the power of the combined region and dense cue with longterm and accurate augmented reality display mechanism. The system is capable of providing detailed visual output by precisely placing the model developed using the CT images of the target object, over the live video. Also, by applying the least square approach, the system is capable of addressing larger deformation and occlusion that appears during the surgical procedure providing the most accurate display. The feature tracking and tracking recovery components help the entire visualization procedure to stay on track by automatically registering and re-registering the surface whenever required. The proposed system capable of running image registration without human involvement and it can even decide when to trigger the reregistration process whenever required. The results from the proposed system has minimized the overlay error by a larger number. We validated the system with different sets of samples from endoscopy. The dataset included the samples from the bowel region from people with three different age groups. The overlay error accuracy was 0.24777px, and the performance was 44fps. The proposed system is concentrated on the overlay accuracy and the processing time. This study has addressed the shortcoming of the previous systems regarding manual registration and rigid assumptions.
KW - Augmented reality
KW - Colonoscopy
KW - Computer-aided diagnosis
KW - Gastroscopy
KW - Visualization
UR - http://www.scopus.com/inward/record.url?scp=85100853479&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100853479&partnerID=8YFLogxK
U2 - 10.1007/s11042-021-10606-8
DO - 10.1007/s11042-021-10606-8
M3 - Article
AN - SCOPUS:85100853479
VL - 80
SP - 17893
EP - 17922
JO - Multimedia Tools and Applications
JF - Multimedia Tools and Applications
SN - 1380-7501
IS - 12
ER -