Automatic quantification of multi-modal rigid registration accuracy using feature detectors

F. Hauler, H. Furtado, M. Jurisic, S. H. Polanec, C. Spick, A. Laprie, U. Nestle, U. Sabatini, W. Birkfellner

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In radiotherapy, the use of multi-modal images can improve tumor and target volume delineation. Images acquired at different times by different modalities need to be aligned into a single coordinate system by 3D/3D registration. State of the art methods for validation of registration are visual inspection by experts and fiducial-based evaluation. Visual inspection is a qualitative, subjective measure, while fiducial markers sometimes suffer from limited clinical acceptance. In this paper we present an automatic, non-invasive method for assessing the quality of intensity-based multi-modal rigid registration using feature detectors. After registration, interest points are identified on both image data sets using either speeded-up robust features or Harris feature detectors. The quality of the registration is defined by the mean Euclidean distance between matching interest point pairs. The method was evaluated on three multi-modal datasets: an ex vivo porcine skull (CT, CBCT, MR), seven in vivo brain cases (CT, MR) and 25 in vivo lung cases (CT, CBCT). Both a qualitative (visual inspection by radiation oncologist) and a quantitative (mean target registration error - mTRE - based on selected markers) method were employed. In the porcine skull dataset, the manual and Harris detectors give comparable results but both overestimated the gold standard mTRE based on fiducial markers. For instance, for CT-MR-T1 registration, the mTREman (based on manually annotated landmarks) was 2.2 mm whereas mTREHarris (based on landmarks found by the Harris detector) was 4.1 mm, and mTRESURF (based on landmarks found by the SURF detector) was 8 mm. In lung cases, the difference between mTREman and mTREHarris was less than 1 mm, while the difference between mTREman and mTRESURF was up to 3 mm. The Harris detector performed better than the SURF detector with a resulting estimated registration error close to the gold standard. Therefore the Harris detector was shown to be the more suitable method to automatically quantify the geometric accuracy of multimodal rigid registration.

Original languageEnglish
Article number5198
Pages (from-to)5198-5214
Number of pages17
JournalPhysics in Medicine and Biology
Volume61
Issue number14
DOIs
Publication statusPublished - Jun 28 2016

Fingerprint

Fiducial Markers
Skull
Swine
Lung
Tumor Burden
Radiotherapy
Brain
Datasets

Keywords

  • automatic landmark based validation
  • feature based evaluation of rigid registration
  • multi-modal image registration
  • quantitative evaluation

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Medicine(all)
  • Radiology Nuclear Medicine and imaging

Cite this

Hauler, F., Furtado, H., Jurisic, M., Polanec, S. H., Spick, C., Laprie, A., ... Birkfellner, W. (2016). Automatic quantification of multi-modal rigid registration accuracy using feature detectors. Physics in Medicine and Biology, 61(14), 5198-5214. [5198]. https://doi.org/10.1088/0031-9155/61/14/5198

Automatic quantification of multi-modal rigid registration accuracy using feature detectors. / Hauler, F.; Furtado, H.; Jurisic, M.; Polanec, S. H.; Spick, C.; Laprie, A.; Nestle, U.; Sabatini, U.; Birkfellner, W.

In: Physics in Medicine and Biology, Vol. 61, No. 14, 5198, 28.06.2016, p. 5198-5214.

Research output: Contribution to journalArticle

Hauler, F, Furtado, H, Jurisic, M, Polanec, SH, Spick, C, Laprie, A, Nestle, U, Sabatini, U & Birkfellner, W 2016, 'Automatic quantification of multi-modal rigid registration accuracy using feature detectors', Physics in Medicine and Biology, vol. 61, no. 14, 5198, pp. 5198-5214. https://doi.org/10.1088/0031-9155/61/14/5198
Hauler, F. ; Furtado, H. ; Jurisic, M. ; Polanec, S. H. ; Spick, C. ; Laprie, A. ; Nestle, U. ; Sabatini, U. ; Birkfellner, W. / Automatic quantification of multi-modal rigid registration accuracy using feature detectors. In: Physics in Medicine and Biology. 2016 ; Vol. 61, No. 14. pp. 5198-5214.
@article{4151430065654a0bad5ebb1e6b99ee6c,
title = "Automatic quantification of multi-modal rigid registration accuracy using feature detectors",
abstract = "In radiotherapy, the use of multi-modal images can improve tumor and target volume delineation. Images acquired at different times by different modalities need to be aligned into a single coordinate system by 3D/3D registration. State of the art methods for validation of registration are visual inspection by experts and fiducial-based evaluation. Visual inspection is a qualitative, subjective measure, while fiducial markers sometimes suffer from limited clinical acceptance. In this paper we present an automatic, non-invasive method for assessing the quality of intensity-based multi-modal rigid registration using feature detectors. After registration, interest points are identified on both image data sets using either speeded-up robust features or Harris feature detectors. The quality of the registration is defined by the mean Euclidean distance between matching interest point pairs. The method was evaluated on three multi-modal datasets: an ex vivo porcine skull (CT, CBCT, MR), seven in vivo brain cases (CT, MR) and 25 in vivo lung cases (CT, CBCT). Both a qualitative (visual inspection by radiation oncologist) and a quantitative (mean target registration error - mTRE - based on selected markers) method were employed. In the porcine skull dataset, the manual and Harris detectors give comparable results but both overestimated the gold standard mTRE based on fiducial markers. For instance, for CT-MR-T1 registration, the mTREman (based on manually annotated landmarks) was 2.2 mm whereas mTREHarris (based on landmarks found by the Harris detector) was 4.1 mm, and mTRESURF (based on landmarks found by the SURF detector) was 8 mm. In lung cases, the difference between mTREman and mTREHarris was less than 1 mm, while the difference between mTREman and mTRESURF was up to 3 mm. The Harris detector performed better than the SURF detector with a resulting estimated registration error close to the gold standard. Therefore the Harris detector was shown to be the more suitable method to automatically quantify the geometric accuracy of multimodal rigid registration.",
keywords = "automatic landmark based validation, feature based evaluation of rigid registration, multi-modal image registration, quantitative evaluation",
author = "F. Hauler and H. Furtado and M. Jurisic and Polanec, {S. H.} and C. Spick and A. Laprie and U. Nestle and U. Sabatini and W. Birkfellner",
year = "2016",
month = "6",
day = "28",
doi = "10.1088/0031-9155/61/14/5198",
language = "English",
volume = "61",
pages = "5198--5214",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "14",

}

TY - JOUR

T1 - Automatic quantification of multi-modal rigid registration accuracy using feature detectors

AU - Hauler, F.

AU - Furtado, H.

AU - Jurisic, M.

AU - Polanec, S. H.

AU - Spick, C.

AU - Laprie, A.

AU - Nestle, U.

AU - Sabatini, U.

AU - Birkfellner, W.

PY - 2016/6/28

Y1 - 2016/6/28

N2 - In radiotherapy, the use of multi-modal images can improve tumor and target volume delineation. Images acquired at different times by different modalities need to be aligned into a single coordinate system by 3D/3D registration. State of the art methods for validation of registration are visual inspection by experts and fiducial-based evaluation. Visual inspection is a qualitative, subjective measure, while fiducial markers sometimes suffer from limited clinical acceptance. In this paper we present an automatic, non-invasive method for assessing the quality of intensity-based multi-modal rigid registration using feature detectors. After registration, interest points are identified on both image data sets using either speeded-up robust features or Harris feature detectors. The quality of the registration is defined by the mean Euclidean distance between matching interest point pairs. The method was evaluated on three multi-modal datasets: an ex vivo porcine skull (CT, CBCT, MR), seven in vivo brain cases (CT, MR) and 25 in vivo lung cases (CT, CBCT). Both a qualitative (visual inspection by radiation oncologist) and a quantitative (mean target registration error - mTRE - based on selected markers) method were employed. In the porcine skull dataset, the manual and Harris detectors give comparable results but both overestimated the gold standard mTRE based on fiducial markers. For instance, for CT-MR-T1 registration, the mTREman (based on manually annotated landmarks) was 2.2 mm whereas mTREHarris (based on landmarks found by the Harris detector) was 4.1 mm, and mTRESURF (based on landmarks found by the SURF detector) was 8 mm. In lung cases, the difference between mTREman and mTREHarris was less than 1 mm, while the difference between mTREman and mTRESURF was up to 3 mm. The Harris detector performed better than the SURF detector with a resulting estimated registration error close to the gold standard. Therefore the Harris detector was shown to be the more suitable method to automatically quantify the geometric accuracy of multimodal rigid registration.

AB - In radiotherapy, the use of multi-modal images can improve tumor and target volume delineation. Images acquired at different times by different modalities need to be aligned into a single coordinate system by 3D/3D registration. State of the art methods for validation of registration are visual inspection by experts and fiducial-based evaluation. Visual inspection is a qualitative, subjective measure, while fiducial markers sometimes suffer from limited clinical acceptance. In this paper we present an automatic, non-invasive method for assessing the quality of intensity-based multi-modal rigid registration using feature detectors. After registration, interest points are identified on both image data sets using either speeded-up robust features or Harris feature detectors. The quality of the registration is defined by the mean Euclidean distance between matching interest point pairs. The method was evaluated on three multi-modal datasets: an ex vivo porcine skull (CT, CBCT, MR), seven in vivo brain cases (CT, MR) and 25 in vivo lung cases (CT, CBCT). Both a qualitative (visual inspection by radiation oncologist) and a quantitative (mean target registration error - mTRE - based on selected markers) method were employed. In the porcine skull dataset, the manual and Harris detectors give comparable results but both overestimated the gold standard mTRE based on fiducial markers. For instance, for CT-MR-T1 registration, the mTREman (based on manually annotated landmarks) was 2.2 mm whereas mTREHarris (based on landmarks found by the Harris detector) was 4.1 mm, and mTRESURF (based on landmarks found by the SURF detector) was 8 mm. In lung cases, the difference between mTREman and mTREHarris was less than 1 mm, while the difference between mTREman and mTRESURF was up to 3 mm. The Harris detector performed better than the SURF detector with a resulting estimated registration error close to the gold standard. Therefore the Harris detector was shown to be the more suitable method to automatically quantify the geometric accuracy of multimodal rigid registration.

KW - automatic landmark based validation

KW - feature based evaluation of rigid registration

KW - multi-modal image registration

KW - quantitative evaluation

UR - http://www.scopus.com/inward/record.url?scp=84978288636&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978288636&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/61/14/5198

DO - 10.1088/0031-9155/61/14/5198

M3 - Article

VL - 61

SP - 5198

EP - 5214

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 14

M1 - 5198

ER -