Semantic Localization-driven Partial Image Retrieval in CT Series

2012 ◽  
Vol 51 (06) ◽  
pp. 557-565 ◽  
Author(s):  
A. Cavallaro ◽  
H.-P. Kriegel ◽  
M. Schubert ◽  
M. Petri

SummaryBackground: Picture archiving and communication systems (PACS) contain very large amounts of computed tomography (CT) data. When querying a PACS for a particular series, the user is often not interested in the complete series but in a certain region of interest (ROI), described e.g. by an example view in another series or an anatomical concept.Objectives: Restricting a retrieval query to such an ROI saves both loading time and navigational effort. In this paper, we propose an efficient method for defining and retrieving ROIs.Methods: We employ interpolation and regression techniques for mapping the slices of a series to a newly generated standardized height atlas of the human body.Results: Examinations of the accuracy and the saved input/output (I/O) costs of our new method on a repository of 1,360 CT series demonstrate the advantages of our system. Depending on the scope of the retrieval query, we can economize up to 99% of the total loading time.Conclusion: Our proposed method for flexible, context-based, partial image retrieval enables the user to directly focus on the relevant portion of the image material and it targets the high potential of I/O cost reduction of a common PACS.

Author(s):  
Chee-Chiang Chen ◽  
Pai-Jung Huang ◽  
Chih-Ying Gwo ◽  
Yue Li ◽  
Chia-Hung Wei

Content-based image retrieval (CBIR) has been proposed by the medical community for inclusion in picture archiving and communication systems (PACS). In CBIR, relevance feedback is developed for bridging the semantic gap and improving the effectiveness of image retrieval systems. With relevance feedback, CBIR systems can return refined search results using a learning algorithm and selection strategy. In this study, as the retrieving process proceeds further, the proposed learning algorithm can reduce the influence of the original query point and increase the significance of the centroid of the clusters comprising the features of those relevant images identified in the most recent round of search. The proposed selection strategy is used to find a good starting point and select a set of images at each round to show that search result and ask for the user’s feedback. In addition, a benchmark is proposed to measure the learning ability to explain the retrieval performance as relevance feedback is incorporated in CBIR systems. The performance evaluation shows that the average precision rate of the proposed scheme was 0.98 and the learning ability reach to 7.17 through the five rounds of relevance feedback.


2020 ◽  
Vol 5 (4) ◽  
pp. 2473011420S0049
Author(s):  
Caroline Williams ◽  
John Y. Kwon ◽  
Max Michalski ◽  
Christopher P. Miller

Category: Other; Ankle; Midfoot/Forefoot Introduction/Purpose: Computed tomography (CT) advanced imaging techniques are a valuable tool for orthopedic surgeons when evaluating complex foot and ankle pathology. With advances in CT data processing, surgeons can create custom reformation of the imaging data in real time using postprocessing imaging tools. This article presents a technique describing how to manipulate CT data using two of these tools, multiplanar reformation (MPR) and maximal intensity projection (MIP), to better visualize pathology and allow a more definitive diagnosis preoperatively. Despite their availability on most modern picture archiving and communication systems, most surgeons across orthopaedics disciplines fail to utilize these powerful tools. Methods: Patients with complex midfoot deformities underwent CT scans for diagnosis and preoperative planning. Case 1 underwent percutaneous pinning of multiple metatarsal fractures. Post-operatively she had severe pain localized to the second and third metatarsals. A CT scan obtained in standard axes failed to interpret the multiplanar deformities when viewed initially on PACS. Using MPR/MIP, the axes were rotated to generate a detailed view of the deformities and subsequently template osteotomies. Case two underwent arthrodesis of the second and third tarsometatarsal (TMT) joints for an unstable Lisfranc injury. Post-operatively she presented with severe midfoot pain prohibiting her return to athletics. Radiographs demonstrated normal alignment with hardware obscuring the joints. Standard CT axes were difficult to interpret due to metal artifact and oblique planes of the TMT joints. MPR/MIP reformatting allowed metal artifact reduction through axes adjustment, improving visualization and facilitating diagnosis of nonunion of the second and third TMT fusions. Results: Case One:MIP/MPR allowed manipulation of the CT scan in the axial and coronal planes provided a sagittal reformat of the entire second and third metatarsals. The second metatarsal had a gradual plantarflexion malunion of the diaphysis. The third metatarsal had an acute plantarflexed deformity of the metatarsal neck. Reformatting provided an accurate preoperative template for planned dorsiflexion osteotomies. Case Two: Metal artifact and the oblique orientation of the TMT joints prevented evaluation of previous fusions. MIP/MPR reformatting provided axes which were perpendicular to the TMT joints decreased metal artifact and demonstrated nonunion of the previous fusions. Use of MIP/MPR in this case provided valuable diagnostic information regarding the source of pain and a plan for removal of hardware and revision TMT fusions. Conclusion: The MPR/MIP function is available in most PACS systems and allows customizable CT reformats. The technique is easy to learn and can be done quickly in the office or pre-operative setting. The technology has facilitated streamlined pre- operative planning and improvements in injury visualization numerous times, particularly in complex deformities and trauma. Using MIP/MPR reformats, the authors have been able to identify critical fracture lines and relationships between anatomic structures which may otherwise have been missed or less precisely understood. The authors hope that this article will enhance awareness and encourage others to utilize this powerful technology.


2018 ◽  
Vol 4 (1) ◽  
pp. 331-335
Author(s):  
David Schote ◽  
Tim Pfeiffer ◽  
Georg Rose

AbstractComputed tomography (CT) scans are frequently used intraoperatively, for example to control the positioning of implants during intervention. Often, to provide the required information, a full field of view is unnecessary. I nstead, the region-of-interest (ROI) imaging can be performed, allowing for substantial reduction in the applied X-ray dose. However, ROI imaging leads to data inconsistencies, caused by the truncation of the projections. This lack of information severely impairs the quality of the reconstructed images. This study presents a proof-of-concept for a new approach that combines the incomplete CT data with ultrasound data and time of flight measurements in order to restore some of the lacking information. The routine is evaluated in a simulation study using the original Shepp-Logan phantom in ROI cases with different degrees of truncation. Image quality is assessed by means of normalized root mean square error. The proposed method significantly reduces truncation artifacts in the reconstructions and achieves considerable radiation exposure reductions.


Sign in / Sign up

Export Citation Format

Share Document