The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index; 6 of the 18 sources its claims cite were checked for support.
3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch
Open accessAbstract
Abstract In image-guided spine surgery, robust three-dimensional to two-dimensional (3D–2D) registration of preoperative computed tomography (CT) and intraoperative radiographs can be challenged by the image content mismatch associated with the presence of surgical instrumentation and implants as well as soft-tissue resection or deformation. This work investigates image similarity metrics in 3D–2D registration offering improved robustness against mismatch, thereby improving performance and reducing or eliminating the need for manual masking. The performance of four gradient-based image similarity metrics (gradient information (GI), gradient correlation (GC), gradient information with linear scaling (GS), and gradient orientation (GO)) with a multi-start optimization strategy was evaluated in an institutional review board-approved retrospective clinical study using 51 preoperative CT images and 115 intraoperative mobile radiographs. Registrations were tested with and without polygonal masks as a function of the number of multistarts employed during optimization. Registration accuracy was evaluated in terms of the projection distance error (PDE) and assessment of failure modes (PDE > 30 mm) that could impede reliable vertebral level localization. With manual polygonal masking and 200 multistarts, the GC and GO metrics exhibited robust performance with 0% gross failures and median PDE < 6.4 mm (±4.4 mm interquartile range (IQR)) and a median runtime of 84 s (plus upwards of 1–2 min for manual masking). Excluding manual polygonal masks and decreasing the number of multistarts to 50 caused the GC-based registration to fail at a rate of >14%; however, GO maintained robustness with a 0% gross failure rate. Overall, the GI, GC, and GS metrics were susceptible to registration errors associated with content mismatch, but GO provided robust registration (median PDE = 5.5 mm, 2.6 mm IQR) without manual masking and with an improved runtime (29.3 s). The GO metric improved the registration accuracy and robustness in the presence of strong image content mismatch. This capability could offer valuable assistance and decision support in spine level localization in a manner consistent with clinical workflow.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
This study evaluated four gradient-based similarity metrics for 3D–2D CT-to-radiograph registration in spine surgery under image content mismatch. Gradient orientation (GO) was most robust, maintaining 0% gross failures without manual polygonal masking and with shorter runtime.
Key findings
- With manual polygonal masking and 200 multistarts, GC and GO achieved 0% gross failures and median projection distance error (PDE) < 6.4 mm (±4.4 mm IQR), with median runtime of 84 s plus 1–2 min for manual masking.
- Without manual polygonal masks and with 50 multistarts, GC failed at >14%, while GO maintained 0% gross failure, median PDE = 5.5 mm (2.6 mm IQR), and runtime = 29.3 s.
- GI and GS were susceptible to content mismatch; in the R(p, 200) configuration they had >16% failure, though GS improved gross failure rate over GI by approximately 50%. In R(r, 50), GC gross failure increased to 15%.
- GO had the largest median capture range in the superior–inferior direction: 375 mm ± 122 mm IQR, versus GI 192 mm ± 214 mm, GC 345 mm ± 185 mm, and GS 293 mm ± 180 mm.
Methods
- Institutional review board-approved retrospective clinical study using 51 preoperative CT images and 115 intraoperative mobile radiographs from thoracolumbar spine surgery; radiographs included 105 lateral and 10 anterior–posterior projections.
- Four gradient-based similarity metrics were tested: gradient information (GI), gradient correlation (GC), gradient information with linear scaling (GS), and gradient orientation (GO), using multistart optimization with CMA-ES.
- Registration performance was evaluated under three configurations—R(p, 200), R(p, 50), and R(r, 50)—using PDE, failure defined as PDE > 30 mm, runtime, objective-function search-space quality, and sensitivity to superior–inferior initialization.
Limitations
- The paper states that spine deformation is not specifically addressed by the rigid registration approach, and that ongoing clinical evaluation is needed.
- The paper states that manual polygonal masking is time-consuming, subjective, error-prone, and disruptive to workflow; GS and modified GC also require parameters that are non-trivial to determine automatically.
- Evident from the text: the study is retrospective and single-institution, uses mostly lateral radiographs, and relies on neuroradiologist-annotated vertebral centroids as reference truth.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
34 Own result · 30 Own interpretation · 15 No basis given · 4 Partly supported · 2 Sources not found · 1 Supported · 1 Supported so far · 1 No full text of the sources · 1 Not checked yet · 1 Own method · accuracy not yet measured
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, the challenge due to content mismatch was mitigated by manually defined polygonal masks that excluded extraneous regions during similarity metric evaluation.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
What the sources say
- Not matchedcites Lo et al 2015
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Partly supportedcites Otake et al 2015 sourceExperimental — accuracy not yet measured
The Patient study section (page unavailable) says a weighting mask can down-weight regions with deformation or instrumentation so they contribute less to the similarity metric. It says such masks could potentially be made manually, but does not describe polygonal masks or establish that manually defined masks were used to mitigate content mismatch in this study. (LLM, conf=0.85)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The GI and GC metrics have been shown previously (Penney et al 1998, Pluim et al 2000, Otake et al 2012b) to be useful when handling partial content mismatch.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
What the sources say
- Not checked — no full text of the sourcecites Penney et al 1998 source
- Not checked — no full text of the sourcecites Pluim et al 2000 source
- Partly supportedcites Otake et al 2012b sourceExperimental — accuracy not yet measured
Methods §2.1 says GI is robust to image-content mismatches, including surgical tools visible in fluoroscopy but absent from the preoperative CT. The Introduction identifies GC as a local-intensity metric and describes that category as generally robust in 3D–2D registration, but the paper does not specifically establish GC’s usefulness for partial content mismatch. Page numbers were not provided in our copy. (LLM, conf=0.86)
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“From the similarity metrics that have been previously investigated, gradient-based metrics have shown promise in 3D–2D image registration problems involving intraoperative radiographs and preoperative CT images (Penney et al 1998, Gendrin et al 2011, Otake et al 2012b, 2015).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
What the sources say
- Not checked — no full text of the sourcecites Penney et al 1998 source
- Supportedcites Gendrin et al 2011 source
The Introduction describes registration of intrainterventional 2D images with preinterventional 3D images. The Results and Discussion report accurate gradient-based registration of CT volumes to kV and MV x-ray images, supporting the claim that these metrics show promise for this type of problem. (LLM, conf=0.91)
- Not matchedcites Otake et al 2012b, 2015
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“For example, previous work on vertebral level localization in spine surgery (Lo et al 2015, Otake et al 2015) showed promising results with GC similarity.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
What the sources say
- Not matchedcites Lo et al 2015
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites Otake et al 2015 source
The Methods (§2.2) state that the study used gradient correlation (GC) as its registration similarity metric. The Patient study (§3.3) reports approximately 1.13 mm average landmark projection error with 9-DOF registration in two spine-surgery patients, while the Discussion (§4) identifies vertebral localization in spine surgery as a potential application. These are promising preliminary results, rather than a definitive clinical validation of level localization. (LLM, conf=0.87)
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“GO exhibited the best overall performance with a median capture range of 375 mm ± 122 mm IQR measured at the detector.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.3. Sensitivity to initialization/capture range
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Overall, the GI, GC, and GS metrics were susceptible to registration errors associated with content mismatch,”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > Abstract
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- The authors state this as an assumption: The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These data exhibited a substantial degree of realistic variation in imaging protocols/techniques, image quality, and imaging systems used for preoperative and intraoperative imaging.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.1. Clinical study
plainn = 51- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For all the metrics, the space exhibits numerous false local optima associated with surgical instruments present in the radiograph (but not the CT) or the semiperiodic structure of the spine itself in the S–I direction.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.2. Quality of the objective function search space
plaincorrelationalSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The GC and GO metrics maintained superior performance in comparison to GI and GS.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For the R(p, 200) configuration (i.e. 2D polygonal masks and 200 multistarts), both the GC and GO metrics exhibited an accurate and robust performance.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- It may state two separate claims as one. 34
- Probably not a substantial claim. 13
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“While all of the above metrics rely upon gradient magnitudes, gradient orientation on its own has been widely used for feature matching (Lowe 2004), object detection (Dalal and Triggs 2005), and image registration (De Nigris et al 2012), and shown promise in robust matching under partial occlusion and view point changes.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
What the sources say
- Not checked — no full text of the sourcecites Lowe 2004 source
- Partly supportedcites Dalal and Triggs 2005 sourceExperimental — accuracy not yet measured
The Abstract describes histograms of oriented gradients used for human detection, supporting the object-detection example attributed to Dalal and Triggs. It does not establish that gradient orientation was used on its own, independent of other gradient information. (LLM, conf=0.82)
- Not checked — no full text of the sourcecites De Nigris et al 2012 source
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
Every source the paper cites for this claim was checked and supports it.
In the paper
“Considering an average thoracolumbar vertebral height (superior–inferior direction) of 22 mm (Busscher et al 2010)”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.4. Performance evaluation
What the sources say
- Supportedcites Busscher et al 2010 source
Table 3 (Results, Vertebral bodies; page not provided) reports human vertebral-body heights by thoracic and lumbar level. The values vary by level and measurement position but support 22 mm as an approximate thoracolumbar height; the paper does not give 22 mm as a single pooled mean. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Supported
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“For the acquired radiographs, the magnification factor at the detector typically ranged from 1.4–1.6.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.4. Performance evaluation
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The other metrics demonstrated a fairly broad capture range as well: GI capture range = 192 mm ± 214 mm IQR; GC capture range = 345 mm ± 185 mm IQR; and GS capture range = 293 mm ± 180 mm IQR.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.3. Sensitivity to initialization/capture range
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The PDE following initialization ranged from 11.3 to 111.1 mm (median PDE = 43.1 mm, interquartile range 20.9 mm).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
- The authors state this as an assumption: The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The GO metric, on the other hand, maintained geometric accuracy and robustness.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“On the other hand, GC shows a more distinct optimum at the solution, but it also exhibits sharp local maxima attributed to strong gradient magnitudes far from the solution (surgical instruments).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.2. Quality of the objective function search space
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“While all metrics align corresponding square and circle elements at the desired solution, the mismatching triangle at the degenerate solution causes the GI, GC metrics to yield a larger relative contribution to the overall metric value.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“but GO provided robust registration (median PDE = 5.5 mm, 2.6 mm IQR) without manual masking and with an improved runtime (29.3 s).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- The authors state this as an assumption: The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The performance of the GI, GC, and GS metrics degraded significantly, e.g. the gross failure rate for GC increasing to 15%.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For GI and GS, the peak at the solution is less distinct from surrounding local maxima, although GS appears improved over GI (consistent with improved robustness noted above).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.2. Quality of the objective function search space
hedgedcorrelationalSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the average computation time ranged from 23–29 s for the S = 50 configurations, compared to 65–85 s for the S = 200 configuration.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Among distributions RGO(p, 200), RGO(p, 50), and RGO(r, 50), we found p-value >0.94 failing to reject the null hypothesis that the PDE distributions have the same median.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For the GO metric, the gradient magnitude values do not dominate the overall metric value, and the desired solution is reached wherein corresponding vertebrae gradients align in a manner that is robust against extraneous gradients and occlusions from instrumentation.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The GO metric exhibits the most distinct global optimum at the solution and suppresses false maxima far from the solution.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.2. Quality of the objective function search space
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“With manual polygonal masking and 200 multistarts, the GC and GO metrics exhibited robust performance with 0% gross failures and median PDE < 6.4 mm (±4.4 mm interquartile range (IQR)) and a median runtime of 84 s (plus upwards of 1–2 min for manual masking).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > Abstract
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- The authors state this as an assumption: The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Overall, the RGO(r, 50) configuration outperformed all of the GI, GS, and GC configurations in terms of geometric accuracy, robustness, and run time”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“but compromises the robustness of the registration and results in an increasing failure rate for all of the similarity metrics.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“but such mm-scale variations in PDE in table 2 were not statistically significant (p = 0.52, 0.72, 0.10, and 0.07 for GI, GC, GS, and GO, respectively) and are likely not clinically significant with respect to localizing a particular vertebral body.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Although the GS metric somewhat improved the registration performance compared to GI, it did not exhibit particularly strong robustness to failure (>16%) over all the tested configurations.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“In spine surgery, for example, the following sources of mismatch are common occurrences (as shown in figure 1) and present particular challenges to 3D–2D registration: (a) The presence of surgical instrumentation in the intraoperative radiograph (e.g. retractors, screws, implants, clips, etc); and (b) non-rigid anatomical deformation”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“However, such an approach increases the computational burden and makes the registration algorithm susceptible to segmentation errors.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“GC can accommodate such differences up to a linear factor and therefore offers some degree of robustness against mismatch:”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Such broad capture ranges are partly attributable to the similarity metric and perhaps more so to the robust optimization strategy involving multiple parallel searches from multiple initial seed points.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.3. Sensitivity to initialization/capture range
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Increasing the number of multistarts (denoted S in figure 2) boosts the density of seed points positioned within the multidimensional space and therefore improves the robustness of finding the global optima at the cost of additional computation in performing parallel searches (roughly proportional to the number of multi-starts).”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The min operator in GI assumes that anatomical structures produce a constant gradient magnitude in both DRRs and radiographs.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A claim is checked only against the cited paper’s full text, never its abstract, and none is available for its sources.
In the paper
“Image gradients can be used to filter out low spatial frequency differences such as soft tissue (Penney et al 1998, Pluim et al 2000) and act as a high-pass filter to focus the registration on the bony anatomy clearly visible in x-ray images.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
What the sources say
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The GO similarity metric thereby exhibits robust performance against content mismatch and enables the definition of a simple rectangular mask of the radiograph collimators to reduce or eliminate the need for polygonal masking.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The GO metric improved the registration accuracy and robustness in the presence of strong image content mismatch.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- The authors state this as an assumption: The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the definition of manual polygonal masks is subject to user variability and tends to exclude regions that may be disadvantageous to registration.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Manual/semi-automatic methods to mask out such extraneous (or missing) content are time consuming, user dependent, error prone, and disruptive to workflow.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This capability could offer valuable assistance and decision support in spine level localization in a manner consistent with clinical workflow.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > Abstract
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The definition of polygons to mask extraneous objects is a slow, subjective step that is error prone and limits the practical utility of image-based registration within streamlined workflow.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Content mismatch caused by extraneous objects and tissue deformation can create a challenging search space with multiple local optima.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.3. 3D–2D registration framework
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For the GS metric, this effect is mitigated partly by the scaling parameter α; however, the exclusion of extraneous gradients requires estimating α to accurately match the corresponding gradients in the two images.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 3. Results > 3.1. Registration accuracy and robustness
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The challenging search space illustrated in figure 6 shows multiple local maxima resulting from extraneous objects and stands to benefit from the multistart optimization strategy to improve robustness.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Deformation of the spine is not specifically addressed in the current (rigid) registration approach, but the large clinical dataset did present realistic levels of anatomical deformation.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Moreover, the corresponding increase in robustness has diminishing return due to the ‘curse of dimensionality’ in the 6D search space, whereas the computation time increases approximately linearly with the number of multistarts.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We evaluated this parameter by computing the ratio of maximum gradient magnitudes observed between the two images; however, the parameter should be adjusted according to mismatch caused by strong gradient magnitudes. Automatic determination of the parameter for any given pair of images is non-trivial.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Therefore, it is desirable for the registration framework to rely less upon the number of multistarts, an attribute exhibited with the incorporation of the GO similarity metric.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These improvements better enable clinical translation of the registration algorithm”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Another drawback of the polygonal masking approach is that it requires the user to anticipate and identify problematic gradients in a radiograph with many subtleties.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The small variations in PDE could be attributed to errors in the localization of the vertebrae centroid (expert reader truth definition), the stochasticity of the CMA-ES optimizer, and/or fundamental limitations arising from the use of similarity metrics as surrogate measures of registration accuracy during the optimization process.”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 4. Discussion and conclusions
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The authors state this as an assumption.
The transformation between preoperative CT and intraoperative radiograph can be modeled as a rigid 6D transformation.
“The registration is performed by optimizing the image similarity between the preoperative CT image and the intraoperative radiograph (I2:R2→R) in a rigid 6D transformation space.”
The paper relies on this without support.
Disparity in the superior–inferior (S–I) position is the most susceptible factor for level misidentification in spine registration.
“disparity in the S–I position is believed to be the most susceptible for level misidentification (e.g. registration errors resulting in label positions away from the true level by one or more levels in the S–I direction).”
The paper relies on this without support.
The clinical need is to identify the correct vertebral level within approximately 30 mm PDE.
“Given the clinical need to identify the correct vertebral level within approximately 30 mm PDE (i.e. within the correct vertebral body)”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.2. Image pre-processing and target definition
“minor axis 50 mm”
§ 3D–2D image registration for target localization in spine surgery: investigation of similarity metrics providing robustness to content mismatch > 2. Methods and materials > 2.2. Image pre-processing and target definition
“major axis 25 mm”
Citations
Cites
Distinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589 · cited 1×
“…(GO): While all of the above metrics rely upon gradient magnitudes, gradient orientation on its own has been widely used for feature matching (Lowe 2004), object detection (Dalal and Triggs 2005), and image registration (De Nigris et al 2012), and shown promise in robust matching under…”
BackgroundHistograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2×
“Gradient orientation (GO): While all of the above metrics rely upon gradient magnitudes, gradient orientation on its own has been widely used for feature matching (Lowe 2004), object detection (Dalal and Triggs 2005), and image registration (De Nigris et al 2012), and shown promise in robust matching under partial occ…”
BackgroundMethodThe CMA Evolution Strategy: A Comparing Review
Hansen, N. (2007) · Studies in fuzziness and soft computing · doi:10.1007/3-540-32494-1_4 · Cited by 1,907 · cited 1×
“To achieve robust performance under these conditions, the registration process incorporates a multistart optimization with the covariance matrix adaptation; evolution strategy (CMA-ES) reported by (Hansen 2006).”
MethodA comparison of similarity measures for use in 2-D-3-D medical image registration
Penney, G. P., Weese, J., Little, J. A., Desmedt, P., Hill, D. L. G., & hawkes, D. J. (1998) · IEEE Transactions on Medical Imaging · doi:10.1109/42.730403 · Cited by 782
Image registration by maximization of combined mutual information and gradient information
Pluim, J. P. W., Maintz, J. B. A., & Viergever, M. A. (2000) · IEEE Transactions on Medical Imaging · doi:10.1109/42.876307 · Cited by 670
Towards a New Evolutionary Computation
Lozano, J. A., Larrañaga, P., Inza, I., & Bengoetxea, E. (2006) · Springer eBooks · doi:10.1007/3-540-32494-1 · Cited by 519
Automated model-based vertebra detection, identification, and segmentation in CT images
Klinder, T., Östermann, J., Ehm, M., et al. (2009) · Medical Image Analysis · doi:10.1016/j.media.2009.02.004 · Cited by 331 · cited 1×
“Labeling in CT can potentially be done automatically (Klinder et al 2009, Ma and Lu 2013), but in this study, a boardcertified neuroradiologist annotated the vertebral labels at the approximate centroid of each vertebral body.”
Comparative anatomical dimensions of the complete human and porcine spine
Busscher, I., Ploegmakers, J. J. W., Verkerke, G. J., & Veldhuizen, A. G. (2010) · European Spine Journal · doi:10.1007/s00586-010-1326-9 · Cited by 295
The Prevalence of Wrong Level Surgery Among Spine Surgeons
Mody, M. G., Nourbakhsh, A., Stahl, D. L., Gibbs, M., Alfawareh, M. D., & Garges, K. J. (2008) · Spine · doi:10.1097/brs.0b013e31816043d1 · Cited by 221 · cited 1×
“This could potentially decrease wrong-level errors, reported to occur in approximately 1 in 3110 spine surgery procedures (Mody et al 2008), and reduce the time and stress associated with resolving anatomical uncertainty.”
BackgroundIntensity-based 2-D-3-D registration of cerebral angiograms
Hipwell, J. H., Penney, G., McLaughlin, R. A., Rhode, K., Summers, P. E., Cox, T. C. S., Byrne, J. V., Noble, J. A., & Hawkes, D. J. (2003) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2003.819283 · Cited by 146 · cited 2×
“…Metrics Validation Penney et al (1998) CT to projection x-ray in spine surgery NCC, entropy, MI, GC, PI, GD Phantom + simulated clinical images Hipwell et al (2003) Magnetic resonance angiography (MRA) to digital subtraction angiography (DSA) in neurointerventions NCC, GC, entropy, MI, PI, GD…”
BackgroundKhamene, A., Bloch, P., Wein, W., Svatos, M. M., & Sauer, F. (2005) · Medical Image Analysis · doi:10.1016/j.media.2005.06.002 · Cited by 130 · cited 2×
“…for clinical applications ranging from image guidance in the spine (Penney et al 1998) and intracranial interventions (Hipwell et al 2003, McLaughlin et al 2005) to patient positioning/target alignment in radiotherapy (Khamene et al 2006, Kim et al 2007, Wu et al 2009, Gendrin et al 2011).”
BackgroundA protocol for evaluation of similarity measures for rigid registration
Škerl, D., Likar, B., & Pernuš, F. (2006) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2006.874963 · Cited by 93 · cited 1×
“The profiles were normalized such that a unit change in all the transform directions (i.e. both rotations and translations) cause the same mean shift in physical dimensions in the image (Škerl et al 2006).”
BackgroundDemons deformable registration of CT and cone‐beam CT using an iterative intensity matching approach
Nithiananthan, S., Schäfer, S., Uneri, A., et al. (2011) · Medical Physics · doi:10.1118/1.3555037 · Cited by 87
Self-citationOtake, Y., Wang, A. S., Webster Stayman, J., Uneri, A., Kleinszig, G., Vogt, S., Khanna, A. J., Gokaslan, Z. L., & Siewerdsen, J. H. (2013) · Physics in Medicine and Biology · doi:10.1088/0031-9155/58/23/8535 · Cited by 85 · cited 4×
“For the multistart strategy, the six-dimensional (6D) search space is partitioned to multiple, equal-size subspaces following a k-d tree partitioning algorithm, and a separate CMA-ES search was performed for each subspace initialized at their respective centers (Otake et al 2013).”
BackgroundMethodInfluentialSelf-citationJun, M., Lü, L., Zhan, Y., et al. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15705-9_3 · Cited by 84 · cited 1×
“Labeling in CT can potentially be done automatically (Klinder et al 2009, Ma and Lu 2013), but in this study, a boardcertified neuroradiologist annotated the vertebral labels at the approximate centroid of each vertebral body.”
Otake, Y., Schafer, S., Stayman, J. W., Zbijewski, W., Kleinszig, G., Graumann, R., Khanna, A. J., & Siewerdsen, J. H. (2012) · Physics in Medicine and Biology · doi:10.1088/0031-9155/57/17/5485 · Cited by 79 · cited 10×
“From the similarity metrics that have been previously investigated, gradient-based metrics have shown promise in 3D–2D image registration problems involving intraoperative radiographs and preoperative CT images (Penney et al 1998, Gendrin et al 2011, Otake et al 2012b, 2015).”
BackgroundMethodInfluentialSelf-citationMulti-modal image registration based on gradient orientations of minimal uncertainty
De Nigris, D., Collins, D. L., & Arbel, T. (2012) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2012.2218116 · Cited by 73 · cited 1×
“…metrics rely upon gradient magnitudes, gradient orientation on its own has been widely used for feature matching (Lowe 2004), object detection (Dalal and Triggs 2005), and image registration (De Nigris et al 2012), and shown promise in robust matching under partial occlusion and view point changes.”
MethodWu, J., Kim, M., Peters, J., Chung, H., & Samant, S. S. (2009) · Medical Physics · doi:10.1118/1.3250843 · Cited by 66 · cited 3×
“…for clinical applications ranging from image guidance in the spine (Penney et al 1998) and intracranial interventions (Hipwell et al 2003, McLaughlin et al 2005) to patient positioning/target alignment in radiotherapy (Khamene et al 2006, Kim et al 2007, Wu et al 2009, Gendrin et al 2011).”
BackgroundMethodJun, M., & Lü, L. (2013) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2012.11.016 · Cited by 66 · cited 1×
“Labeling in CT can potentially be done automatically (Klinder et al 2009, Ma and Lu 2013), but in this study, a boardcertified neuroradiologist annotated the vertebral labels at the approximate centroid of each vertebral body.”
McLaughlin, R. A., Hipwell, J., Hawkes, D. J., Noble, J. A., Byrne, J. V., & Cox, T. C. (2005) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2005.852067 · Cited by 63 · cited 3×
“While the amount of clinical data used for validation in the studies listed in table 1 is limited, clinical images often pose additional challenges for similarity metrics during registration (Penney et al 1998, McLaughlin et al 2005, Wu et al 2009).”
BackgroundMethodUneri, A., Nithiananthan, S., Schäfer, S., et al. (2012) · Medical Physics · doi:10.1118/1.4767757 · Cited by 43 · cited 1×
“The development and validation of image registration methods (Nithiananthan et al 2011, Uneri et al 2013) that are robust in the presence of content mismatch are important to achieving reliable registration in a manner consistent with the needs of intraoperative workflow and clinical integration.”
MethodSelf-citation3D–2D registration in mobile radiographs: algorithm development and preliminary clinical evaluation
Otake, Y., Wang, A. S., Uneri, A., Kleinszig, G., Vogt, S., Aygun, N., Lo, S.-F. L., Wolinsky, J.-P., Gokaslan, Z. L., & Siewerdsen, J. H. (2015) · Physics in Medicine & Biology · doi:10.1088/0031-9155/60/5/2075 · Cited by 41 · cited 7×
“From the similarity metrics that have been previously investigated, gradient-based metrics have shown promise in 3D–2D image registration problems involving intraoperative radiographs and preoperative CT images (Penney et al 1998, Gendrin et al 2011, Otake et al 2012b, 2015).”
BackgroundResultInfluentialSelf-citationGendrin, C., Markelj, P., Pawiro, S. A., Spoerk, J., Bloch, C., Weber, C., Figl, M., Bergmann, H., Birkfellner, W., Likar, B., et al. (2011) · Medical Physics · doi:10.1118/1.3553403 · Cited by 38
Automatic Localization of Target Vertebrae in Spine Surgery
Lo, S.-F. L., Otake, Y., Puvanesarajah, V., et al. (2015) · Spine · doi:10.1097/brs.0000000000000814 · Cited by 32
Self-citationComparison of Similarity Measures for Rigid-body CT/Dual X-ray Image Registrations
Kim, J., Kim, J., Li, S., et al. (2007) · Technology in Cancer Research & Treatment · doi:10.1177/153303460700600411 · Cited by 24 · cited 3×
“…for clinical applications ranging from image guidance in the spine (Penney et al 1998) and intracranial interventions (Hipwell et al 2003, McLaughlin et al 2005) to patient positioning/target alignment in radiotherapy (Khamene et al 2006, Kim et al 2007, Wu et al 2009, Gendrin et al 2011).”
BackgroundStatistical Segmentation of Surgical Instruments in 3-D Ultrasound Images
Linguraru, M. G., Vasilyev, N. V., del Nido, P. J., et al. (2007) · Ultrasound in Medicine & Biology · doi:10.1016/j.ultrasmedbio.2007.03.003 · Cited by 22 · cited 1×
“Alternatively, automatic/semi-automatic segmentation methods (Linguraru et al 2007) could be implemented as a pre-processing step to mask extraneous objects within the field-of-view of the radiograph.”
MethodOtake, Y., Schafer, S., Stayman, J. W., Zbijewski, W. B., Kleinszig, G., Graumann, R., Khanna, A. J., & Siewerdsen, J. H. (2012) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.911308 · Cited by 11 · cited 10×
“From the similarity metrics that have been previously investigated, gradient-based metrics have shown promise in 3D–2D image registration problems involving intraoperative radiographs and preoperative CT images (Penney et al 1998, Gendrin et al 2011, Otake et al 2012b, 2015).”
BackgroundMethodInfluentialSelf-citation
Cited by
Sorted by citations
Deep learning in medical image registration: a survey
Haskins, G., Kruger, U., & Yan, P. (2020) · Machine Vision and Applications · doi:10.1007/s00138-020-01060-x · Cited by 704
“They were able to significantly outperform registration approaches that used a state-of-the-art similarity metric given by [24].”
Deep learning in medical image registration: a review
Fu, Y., Lei, Y., Wang, T., Curran, W. J., Liu, T., & Yang, X. (2020) · Physics in Medicine & Biology · doi:10.1088/1361-6560/ab843e · Cited by 661
Dilated FCN for Multi-Agent 2D/3D Medical Image Registration
Miao, S., Piat, S., Fischer, P., Tuysuzoglu, A., Mewes, P., Mansi, T., & Liao, R. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11576 · Cited by 113
Colonoscopy 3D video dataset with paired depth from 2D-3D registration
Bobrow, T. L., Golhar, M., Vijayan, R. C., et al. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102956 · Cited by 113
“Depth-based registrations were optimized using L 1 , L 2 , Normalized Cross Correlation (NCC), and Gradient Correlation (GC) loss functions (De Silva et al., 2016).”
MethodMultiview 2D/3D Rigid Registration via a Point-Of-Interest Network for Tracking and Triangulation
Liao, H., Lin, W.-A., Zhang, J., Zhang, J., Luo, J., & Zhou, S. K. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.01292 · Cited by 66
Metadata
- Authors
- De Silva, T., Uneri, A., Ketcha, M. D., Reaungamornrat, S., Kleinszig, G., Vogt, S., Aygun, N., Lo, S.-F., Wolinsky, J.-P., & Siewerdsen, J. H.
- Affiliations
- Johns Hopkins University
- Johns Hopkins University (US)
- Siemens (Germany) (DE)
- Johns Hopkins Medical Institute
- Corresponding author
- Siewerdsen, J H
- Venue
- Physics in Medicine & Biology
- Publisher
- IOP Publishing
- Published
- 2016-03-18
- Volume
- 61
- Issue
- 8
- Pages
- 3009-3025
- ISSN
- 0031-9155 (ISSN-L)
- 1361-6560
- PMC
- PMC4915376
- OpenAlex
- W2329673985
- Semantic Scholar
- b79816585c14219cea6401f1f6b70af222d8b96a
- Cited by
- 107
- References
- 27
- Citations by year
- 2026: 3 · 2025: 4 · 2024: 9 · 2023: 16 · 2022: 12 · 2021: 13 · 2020: 14 · 2019: 13 · 2018: 13 · 2017: 8 · 2016: 2
- FWCI
- 5.49
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Licence
- Locations
- Physics in Medicine and Biology (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (submittedVersion)
- Funders
- National Institutes of Health (R01-EB-017226)
- Siemens Healthcare
- Topics
- Physical Sciences › Engineering › Biomedical Engineering › Medical Imaging and Analysis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Health Sciences › Medicine › Surgery › Spinal Fractures and Fixation Techniques
- Keywords
- image-guided spine surgery, image similarity measures, gradient direction, gradient correlation, gradient information, preoperative CT, multi-start optimization, surgical instrumentation, soft tissue deformation
- MeSH terms
- Humans, Retrospective Studies, Spinal Diseases, Tomography, X-Ray Computed, Imaging, Three-Dimensional, Surgery, Computer-Assisted
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-14, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-14 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
