Deformable image registration is a fundamental task in medical image processing. Among its most important applications, one may cite: 1) multi-modality fusion, where information acquired by different imaging devices or protocols is fused to facilitate diagnosis and treatment planning; 2) longitudinal studies, where temporal structural or anatomical changes are investigated; and 3) population modeling and statistical atlases used to study normal anatomical variability. In this paper, we attempt to give an overview of deformable registration methods, putting emphasis on the most recent advances in the domain. Additional emphasis has been given to techniques applied to medical images. In order to study image registration methods in depth, their main components are identified and studied independently. The most recent techniques are presented in a systematic fashion. The contribution of this paper is to provide an extensive account of registration techniques in a systematic manner.

/pdf/deformable-medical-image-registration-a-survey-2uee59wp1z.pdf

Deformable Medical Image Registration: A Survey

Imaging is central to radiation oncology practice, with advances in radiation oncology occurring in parallel to advances in imaging. Targets to be irradiated and normal tissues to be spared are delineated on computed tomography (CT) scans in the planning process. Computer-assisted design of the radiation dose distribution ensures that the objectives for target coverage and avoidance of healthy tissue are achieved. The radiation treatment units are now recognized as state-of-the-art robotics capable of three-dimensional soft tissue imaging immediately before, during, or after radiation delivery, improving the localization of the target at the time of radiation delivery, to ensure that radiation therapy is delivered as planned. Frequent imaging in the treatment room during a course of radiation therapy, with decisions made on the basis of imaging, is referred to as image-guided radiation therapy (IGRT). IGRT allows changes in tumor position, size, and shape to be measured during the course of therapy, with adjustments made to maximize the geometric accuracy and precision of radiation delivery, reducing the volume of healthy tissue irradiated and permitting dose escalation to the tumor. These geometric advantages increase the chance of tumor control, reduce the risk of toxicity after radiotherapy, and facilitate the development of shorter radiotherapy schedules. By reducing the variability in delivered doses across a population of patients, IGRT should also improve interpretation of future clinical trials.

Advances in Image-Guided Radiation Therapy

Radiotherapy is a highly effective, targeted therapy for the management of cancer. Technological innovations have enabled the direct integration of imaging technology into the radiation treatment device to increase the precision and accuracy of radiation delivery. As well as addressing a clinical need to better control the placement of the dose within the body, image-guided radiotherapy has enabled innovators in the field to accelerate their exploration of a number of different paradigms of radiation delivery, including toxicity reduction, dose escalation, hypofractionation, voxelization, and adaptation. Although these approaches are already innovative trends in radiation oncology, it is anticipated that they will work synergistically with other innovations in cancer management (including biomarker strategies, novel systemic and local therapies) as part of the broader goal of personalized cancer medicine. This Review discusses the rationale for adopting image-guidance approaches in radiotherapy, and the technology for achieving precision and accuracy in the context of different paradigms within the evolving radiation oncology practice. It also examines exciting advances in radiotherapy technology that suggest a convergence of radiotherapy practice in which patient-specific radiotherapy treatment courses are one of the most personalized forms of intervention in cancer medicine.

Image-guided radiotherapy: from current concept to future perspectives

Technological advances have greatly enhanced the specialty of radiation oncology by allowing more healthy tissue to be spared for the same or better tumour coverage. Developments in medical imaging are integral to radiation oncology, both for design of treatment plans and to localise the target for precise administration of radiation. At planning, definition of the tumour and healthy tissue is based on CT, augmented frequently with MRI and PET. At treatment, three-dimensional soft-tissue imaging can also be used to localise the target and tumour motion can be tracked with fluoroscopic imaging of radio-opaque markers implanted in or near the tumour. These developments allow changes in tumour position, size, and shape that take place during radiotherapy to be measured and accounted for to boost geometric accuracy and precision of radiation delivery. Image-guided treatment also enhances uniformity in doses administered in a population of patients, thus improving our ability to measure the effect of dosimetric and non-dosimetric factors on tumour and healthy tissue outcomes in clinical trials. Increased precision and accuracy of radiotherapy are expected to augment tumour control, reduce incidence and severity of toxic effects after radiotherapy, and facilitate development of more efficient shorter schedules than currently available.

https://www.thelancet.com/pdfs/journals/lanonc/PIIS1470-2045(06)70904-4.pdf

Image-guided radiotherapy: rationale, benefits, and limitations

Purpose: Computed tomography (CT) streak artifacts caused by metallic implants remain a challenge for the automatic processing of image data. The impact of metal artifacts in the soft-tissue region is magnified in cone-beam CT (CBCT), because the soft-tissue contrast is usually lower in CBCT images. The goal of this study was to develop an effective offline processing technique to minimize the effect. Methods and Materials: The geometry calibration cue of the CBCT system was used to track the position of the metal object in projection views. The three-dimensional (3D) representation of the object can be established from only two user-selected viewing angles. The position of the shadowed region in other views can be tracked by projecting the 3D coordinates of the object. Automatic image segmentation was used followed by a Laplacian diffusion method to replace the pixels inside the metal object with the boundary pixels. The modified projection data were then used to reconstruct a new CBCT image. The procedure was tested in phantoms, prostate cancer patients with implanted gold markers and metal prosthesis, and a head-and-neck patient with dental amalgam in the teeth. Results: Both phantom and patient studies demonstrated that the procedure was able to minimize the metal artifacts. Soft-tissue visibility was improved near or away from the metal object. The processing time was 1–2 s per projection. Conclusion: We have implemented an effective metal artifact-suppressing algorithm to improve the quality of CBCT images.

Reducing metal artifacts in cone-beam CT images by preprocessing projection data

This paper presents a comprehensive extension of the Scale Invariant Feature Transform (SIFT), originally introduced in 2D, to volumetric images. While tackling the significant computational efforts required by such multiscale processing of large data volumes, our implementation addresses two important mathematical issues related to the 2D-to-3D extension. It includes efficient steps to filter out extracted point candidates that have low contrast or are poorly localized along edges or ridges. In addition, it achieves, for the first time, full 3D orientation invariance of the descriptors, which is essential for 3D feature matching. An application of this technique is demonstrated to the feature-based automated registration and segmentation of clinical datasets in the context of radiation therapy.

Full orientation invariance and improved feature selectivity of 3D SIFT with application to medical image analysis

When modeling anatomical structures in a patient for diagnosis or therapeutic planning, an atlas (26) of predesigned anatomical structure models can be accessed, and model of one or more such structures can be selected and overlaid on an a 3D image of corresponding structure(s) in a clinic image of a patient. A user can click and drag a cursor on the model to deform the model to align with the clinical image. Additionally, a processor (16) can generate a volumetric deformation function using splines, parametric techniques, or the like, and can deform the model to fit the image in real time, in response to user manipulation of the model.

Interactive atlas to image registration

A respiratory marker (40, 140, 240, 340, 440) includes an elongated detectable portion (42, 342, 442) that is operatively coupled with respiration of an imaging subject such that the elongated detectable portion moves with the respiration. The elongated detectable portion is arranged to intersect images acquired by an imaging scanner (10) at different times and at different positions along a scanner axis (20), and is detectable as a marker feature in the images. A marker position finder (52, 54) is configured to determine positions of the marker features in the images.

Retrospective sorting of 4d ct into breathing phases based on geometric analysis of imaging fiducials

Adaptive Planning and Delivery to Account for Anatomical Changes Induced by Radiation Therapy of Head and Neck Cancer

Gy/2-2.5 weeks), and mid-RT (at 45-50 Gy/4-5 weeks). The time-signal intensity (SI) curve and the plateau SI were derived for each tumor pixel of each DCE MRI. The poorly perfused tumor subregions were quantified as lower 10th percentiles of SI among the tumor pixels (SI10). Outcome endpoints, local tumor control and disease-free survival, were correlated with perfusion parameters of the pre-RT, early RT, and mid-RT MRI. Median follow was 4.6 (range: 0.2-9.0) years. Results: Patterns of DCE of the tumor pixels changed during treatment, and the change influenced therapy outcome: In the pre-RT MRI, 16 patients with initial high perfusion (SI10 $ 2.1) had excellent outcome (100% local control and 81% disease-free survival).In28(34%)ofthe82patientswithinitiallylow perfusion(SI10\2.1) onpre-RTMR, thepatternchangedto highperfusion in early-RT (at 20-25 Gy). These patients had an improved local control rate of 93%, compared to those with persistently low perfusion in pre-RT and early-RT MRIs (72%, p = 0.014). If low perfusion persisted to the mid-RT phase (at 45-50 Gy), local control rate decreased further to 66% (p = 0.008). Patients with persistently low perfusion had a disease-free survival rate of 44%, compared to 81% in those with high pre-RT perfusion, and 72% in those with low pre-RT and subsequently high perfusion (p = 0.01). In patients with initially high perfusion pre-RT, a subsequent decrease in SI did not negatively impact on outcome. Conclusions: Tumor perfusion pattern varies during treatment, and such changes critically influence treatment outcome. Persistently low perfusion from the pre-RT through mid-RT phase indicates high risk of treatment failure, while better outcome is associatedwithhighinitial,orhighsubsequentperfusionpatterns,evenwithpriorstatusofpoorperfusion.Ourlimiteddatasuggest thatDCEMRIisaneffectiveandnon-invasivemeanstoassessthetemporalchangesoftumorperfusionstatusduringthetreatment, which may play an essential role in tumor reoxygenation and radioresponsiveness.

D.A. Jaffray

Papers

Full orientation invariance and improved feature selectivity of 3D SIFT with application to medical image analysis

Interactive atlas to image registration

Retrospective sorting of 4d ct into breathing phases based on geometric analysis of imaging fiducials

Adaptive Planning and Delivery to Account for Anatomical Changes Induced by Radiation Therapy of Head and Neck Cancer

Automated Weekly Online Replanning for IMRT of Cervix Cancer