MR imaging-guided prostate interventional imaging: Ready for a clinical use?

Prostate interventional magnetic resonance imaging (MRI) is now routinely performed in many centers. Its more widespread acceptance is limited by the cost of the use of MRI largely related to the long duration time of the procedures. However, the benefit of a robotic assistance has generated a new interest, because it substantially shortens the procedure time, while improving the accuracy. MRI-guided biopsy is considered as an appealing alternative to transrectal ultrasound (TRUS)-guided fusion biopsy, given the limitations of TRUS-MRI image registration systems. MRI-guided focal treatment also benefits from robotic assistance and from the unique property of MRI, which allows the measurement of the temperature in real-time during tumor ablation. The transrectal and transperineal approaches can be used and the respective indications of each pathway will depend on several factors, including the location of the tumor and the examination time, which will condition the occupation time of the MR room, a major factor influencing the overall cost of MRI-guided procedures. This review addresses the current practice of prostate MRI-guided interventional procedures and potential future applications.

Analysis of the correlation between endorectal MRI response to neoadjuvant chemotherapy and biochemical recurrence in patients with high-risk localized prostate cancer.

BACKGROUND: Intermediate end points are desirable to expedite the integration of neoadjuvant systemic therapy into the treatment strategy for high-risk localized prostate cancer. Endorectal magnetic resonance imaging at 1.5 Tesla (1.5T erMRI) response has been utilized as an end point in neoadjuvant trials but has not been correlated with clinical outcomes. METHODS: Data were pooled from two trials exploring neoadjuvant chemotherapy in high-risk localized prostate cancer. Trial 1 explored docetaxel for 6 months and Trial 2 explored docetaxel plus bevacizumab for 4.5 months, both before radical prostatectomy. erMRI was done at baseline and end of chemotherapy. 1.5T erMRI response, based upon T2W sequences, was recorded. Multivariable Cox regression was undertaken to evaluate the association between clinical parameters and biochemical recurrence. RESULTS: There were 53 evaluable patients in the combined analysis: 20 (33%) achieved a PSA response, 16 (27%) achieved an erMRI partial response and 24 (40%) achieved an erMRI minor response. Median follow-up was 4.2 years, and 33 of 53 evaluable (62%) patients developed biochemical recurrence. On multivariable analysis, PSA response did not correlate with biochemical recurrence (hazard ratio=0.58, 95% confidence interval (CI) 0.25-1.33) and paradoxically erMRI response was associated with a significantly shorter time to biochemical recurrence (hazard ratio=2.47, 95% CI 1.00-6.13). CONCLUSIONS: Response by 1.5T erMRI does not correlate with a decreased likelihood of biochemical recurrence in patients with high-risk localized prostate cancer treated with neoadjuvant docetaxel and may be associated with inferior outcomes. These data do not support the use of 1.5T erMRI response as a primary end point in neoadjuvant chemotherapy trials.

Magnetic resonance-guided focused ultrasound (MRgFUS) compared with abdominal hysterectomy for treatment of uterine leiomyomas.

OBJECTIVES: To compare women undergoing magnetic resonance-guided focused ultrasound (MRgFUS) to a group of contemporaneously recruited women undergoing total abdominal hysterectomy. Patient demographics, safety parameters, quality of life outcomes and disability measures are reported. METHODS: One hundred and nine women were recruited in seven centers for MRgFUS treatment and 83 women who underwent abdominal hysterectomy were recruited in seven separate centers to provide contemporaneous assessment of safety. The adverse-event profile and disability parameters were prospectively assessed. Patients were also screened at baseline and at 1, 3 and 6 months using the SF-36 health survey questionnaire. RESULTS: There were no life-threatening adverse events in either group. Overall, the number of significant clinical complications and adverse events was lower in women in the MRgFUS group compared to women undergoing hysterectomy. MRgFUS was associated with significantly faster recovery, including resumption of usual activities. At 6 months of follow-up, there were four (4%) treatment failures in the MRgFUS arm. Regarding SF-36 subscale scores, at 6 months there was improvement in all SF-36 subscales for both treatment groups. However, most of the SF-36 subscale scores were significantly better at this stage in the hysterectomy group than in the MRgFUS group. Women undergoing MRgFUS had steady improvement in all parameters throughout the 6-month follow-up period, despite the fact that they continued to have myomatous uteri and menstruation, which at baseline had given them significant symptomatology. CONCLUSIONS: The results of this study show that MRgFUS treatment of uterine leiomyomas leads to clinical improvement with fewer significant clinical complications and adverse events compared to hysterectomy at 6 months' follow-up.

Dynamic MRI scan plane control for passive tracking of instruments and devices.

This paper describes a novel image-based method for tracking robotic mechanisms and interventional devices during Magnetic Resonance Image (MRI)-guided procedures. It takes advantage of the multi-planar imaging capabilities of MRI to optimally image a set of localizing fiducials for passive motion tracking in the image coordinate frame. The imaging system is servoed to adaptively position the scan plane based on automatic detection and localization of fiducial artifacts directly from the acquired image stream. This closed-loop control system has been implemented using an open-source software framework and currently operates with GE MRI scanners. Accuracy and performance were evaluated in experiments, the results of which are presented here.

Robot-assisted needle placement in open MRI: system architecture, integration and validation.

In prostate cancer treatment, there is a move toward targeted interventions for biopsy and therapy, which has precipitated the need for precise image-guided methods for needle placement. This paper describes an integrated system for planning and performing percutaneous procedures with robotic assistance under MRI guidance. A graphical planning interface allows the physician to specify the set of desired needle trajectories, based on anatomical structures and lesions observed in the patient's registered pre-operative and pre-procedural MR images, immediately prior to the intervention in an open-bore MRI scanner. All image-space coordinates are automatically computed, and are used to position a needle guide by means of an MRI-compatible robotic manipulator, thus avoiding the limitations of the traditional fixed needle template. Automatic alignment of real-time intra-operative images aids visualization of the needle as it is manually inserted through the guide. Results from in-scanner phantom experiments are provided.

Robot-assisted needle placement in open-MRI: system architecture, integration and validation.

This work describes an integrated system for planning and performing percutaneous procedures-such as prostate biopsy-with robotic assistance under MRI-guidance. The physician interacts with a planning interface in order to specify the set of desired needle trajectories, based on anatomical structures and lesions observed in the patient's MR images. All image-space coordinates are automatically computed, and used to position a needle guide by means of an MRI-compatible robotic manipulator, thus avoiding the limitations of the traditional fixed needle template. Direct control of real-time imaging aids visualization of the needle as it is manually inserted through the guide. Results from in-scanner phantom experiments are provided.

Mutual information in coupled multi-shape model for medical image segmentation.

This paper presents extensions which improve the performance of the shape-based deformable active contour model presented earlier in [IEEE Conf. Comput. Vision Pattern Recog. 1 (2001) 463] for medical image segmentation. In contrast to that previous work, the segmentation framework that we present in this paper allows multiple shapes to be segmented simultaneously in a seamless fashion. To achieve this, multiple signed distance functions are employed as the implicit representations of the multiple shape classes within the image. A parametric model for this new representation is derived by applying principal component analysis to the collection of these multiple signed distance functions. By deriving a parametric model in this manner, we obtain a coupling between the multiple shapes within the image and hence effectively capture the co-variations among the different shapes. The parameters of the multi-shape model are then calculated to minimize a single mutual information-based cost criterion for image segmentation. The use of a single cost criterion further enhances the coupling between the multiple shapes as the deformation of any given shape depends, at all times, upon every other shape, regardless of their proximity. We found that this resulting algorithm is able to effectively utilize the co-dependencies among the different shapes to aid in the segmentation process. It is able to capture a wide range of shape variability despite being a parametric shape-model. And finally, the algorithm is robust to large amounts of additive noise. We demonstrate the utility of this segmentation framework by applying it to a medical application: the segmentation of the prostate gland, the rectum, and the internal obturator muscles for MR-guided prostate brachytherapy.

Coupled multi-shape model and mutual information for medical image segmentation.

This paper presents extensions which improve the performance of the shape-based deformable active contour model presented earlier in [9]. In contrast to that work, the segmentation framework that we present in this paper allows multiple shapes to be segmented simultaneously in a seamless fashion. To achieve this, multiple signed distance functions are employed as the implicit representations of the multiple shape classes within the image. A parametric model for this new representation is derived by applying principal component analysis to the collection of these multiple signed distance functions. By deriving a parametric model in this manner, we obtain a coupling between the multiple shapes within the image and hence effectively capture the co-variations among the different shapes. The parameters of the multi-shape model are then calculated to minimize a single mutual information-based cost functional for image segmentation. The use of a single cost criterion further enhances the coupling between the multiple shapes as the deformation of any given shape depends, at all times, upon every other shape, regardless of their proximity. We demonstrate the utility of this algorithm to the segmentation of the prostate gland, the rectum, and the internal obturator muscles for MR-guided prostate brachytherapy.

A practical method to achieve prostate gland immobilization and target verification for daily treatment.

PURPOSE: A practical method to achieve prostate immobilization and daily target localization for external beam radiation treatment is described. METHODS AND MATERIALS: Ten patients who underwent prostate brachytherapy using permanent radioactive source placement were selected for study. To quantify prostate motion both with and without the presence of a specially designed inflatable intrarectal balloon, the computerized tomography-based coordinates of all intraprostatic radioactive sources were compared over 3 consecutive measurements at 1-min intervals. RESULTS: The placement and inflation of the intrarectal balloon were well tolerated by all patients. The mean (range) displacement of the prostate gland when the intrarectal balloon was present vs. absent was 1.3 (0-2.2) mm vs. 1.8 (0-9.1) mm (p = 0.03) at 2 min respectively. The maximum displacement in any direction (anterior-posterior, superior-inferior, or right-left) when the intrarectal balloon was inflated vs. absent was reduced to < or =1 mm from 4 mm. CONCLUSIONS: Both prostate gland immobilization and target verification are possible using a specially designed inflatable intrarectal balloon. Using this device, the posterior margin necessary on the lateral fields to ensure dosimetric coverage of the entire prostate gland could be safely reduced to 5 mm and treatment could be set up and verified using a lateral portal image.

Prenatal diagnosis of pyruvate dehydrogenase deficiency using magnetic resonance imaging.

INTRODUCTION: Pyruvate dehydrogenase deficiency is an inherited inborn error of metabolism associated with early neonatal death and long-term neurologic sequelae in survivors. Prenatal diagnosis currently relies on isolation of fetal cells for subsequent genetic and/or biochemical studies. Magnetic resonance imaging and magnetic resonance spectroscopy have been used on occasion for both postnatal diagnosis and management of pyruvate dehydrogenase deficiency. We illustrate a case in which these non-invasive modalities also prove useful for prenatal diagnosis of this condition. CASE: A 31-year-old multipara with a history of two prior infants affected with pyruvate dehydrogenase deficiency presented with a spontaneous dichorionic, diamniotic twin pregnancy. Magnetic resonance imaging and magnetic resonance spectroscopy were performed on both fetuses. Magnetic resonance imaging of the presenting (male) fetus demonstrated mild ventriculomegaly, increased extracerebrospinal fluid, and decreased cortical sulcation and gyration. The non-presenting (female) fetus was structurally normal. Magnetic resonance spectroscopy spectra were obtained for both fetuses, and were normal. The diagnosis of pyruvate dehydrogenase deficiency was made in the presenting fetus after delivery on the basis of subsequent mortality from severe lactic acidosis. CONCLUSION: Prenatal MR imaging of the fetal brain can be used for prenatal diagnosis in fetuses at risk for pyruvate dehydrogenase deficiency. Prenatal MR spectroscopy, although technically feasible, does not appear to have a role in the prenatal diagnosis of this condition.

Isolated calf vein thrombosis: comparison of MR venography and conventional venography after initial sonography in symptomatic patients.

RATIONALE AND OBJECTIVES: The authors performed this study to compare magnetic resonance (MR) venography and conventional venography in the diagnosis of deep venous thrombosis (DVT) in the calf after sonography. MATERIALS AND METHODS: Sonography was performed in 595 patients who were suspected of having lower-extremity DVT. Patients with positive above-knee duplex sonograms, allergy to iodinated contrast material, renal insufficiency, or cardiac pacemakers and patients who were obese were excluded. The remaining 73 patients were asked to undergo MR venography and conventional venography. All studies were to be performed within 48 hours of the clinical diagnosis and according to standard clinical practice. Images were interpreted by radiologists who were blinded to the results of other modalities. Two separate analyses were performed: one in which conventional venography was used as the standard of reference, and one in which the presence of at least two positive studies for thrombus was considered diagnostic. RESULTS: Although 36 patients agreed to participate in the study, only 14 underwent MR venography and conventional venography within 48 hours of the clinical diagnosis. With use of any two positive studies for confirmation, acute DVT was diagnosed in three patients. Conventional venography depicted two of the three cases, whereas sonography and MR venography each depicted all three. The findings were concordant in only five of the 14 patients. CONCLUSION: Moderate discrepancy among modalities was demonstrated. This suggests radiologists should undertake comparisons among these three modalities for the detection of calf DVT. In patients with a high clinical suspicion, a second modality may be useful if the initial study is negative.

Bayesian regression methodology for estimating a receiver operating characteristic curve with two radiologic applications: prostate biopsy and spiral CT of ureteral stones.

RATIONALE AND OBJECTIVES: The authors evaluated two Bayesian regression models for receiver operating characteristic (ROC) curve analysis of continuous diagnostic outcome data with covariates. MATERIALS AND METHODS: Full and partial Bayesian regression models were applied to data from two studies (n = 180 and 100, respectively): (a) The diagnostic value of prostate-specific antigen (PSA) levels (outcome variable) for predicting disease after radical prostatectomy (gold standard) was evaluated for three risk groups (covariates) based on Gleason scores. (b) Spiral computed tomography was performed on patients with proved obstructing ureteral stones. The predictive value of stone size (outcome) was evaluated along with two treatment options (gold standard), as well as stone location (in or not in the ureterovesical junction [UVJ]) and patient age (covariates). Summary ROC measures were reported, and various prior distributions of the regression coefficients were investigated. RESULTS: (a) In the PSA example, the ROC areas under the full model were 0.667, 0.769, and 0.703, respectively, for the low-, intermediate-, and high-risk groups. Under the partial model, the area beneath the ROC curve was 0.706. (b) The ROC areas for patients with ureteral stones in the UVJ decreased dramatically with age but otherwise were close to that under the partial model (ie, 0.774). The prior distribution had greater influence in the second example. CONCLUSION: The diagnostic tests were accurate in both examples. PSA levels were most accurate for staging prostate cancer among intermediate-risk patients. Stone size was predictive of treatment option for all patients other than those 40 years or older and with a stone in the UVJ.

MR imaging-guided prostate biopsy with surgical navigation software: device validation and feasibility.

Magnetic resonance (MR) imaging--guided prostate biopsy in a 0.5-T open imager is described, validated in phantom studies, and performed in two patients. The needles are guided by using fast gradient-recalled echo and T2-weighted fast spin-echo images. Surgical navigation software provided T2-weighted images critical to targeting the peripheral zone and the tumor. MR imaging can be used to guide prostate biopsy.

Three-dimensional magnetic resonance imaging of fetal brains.

We assessed ten prenatal magnetic resonance imaging(MRI) scans for fetal brain anomalies, and identified eight that were suitable for post-processing. Anatomical abnormalities were assessed on three-dimensional (3D) models and compared with two-dimensional (2D) imaging. We calculated the volumes of the intracranial ventricles and of a periventricular haemorrhage. In three cases, additional clinical information was obtained. 3D modelling of the brain in-utero is possible and can be used to plan treatment.

Primary versus secondary ovarian malignancy: imaging findings of adnexal masses in the Radiology Diagnostic Oncology Group Study.

PURPOSE: To analyze ultrasonographic (US), computed tomographic (CT), and magnetic resonance (MR) imaging features of primary and secondary ovarian malignant neoplasms to determine if there is any significant difference in their appearance. MATERIALS AND METHODS: Analysis of the multi-institutional Radiology Diagnostic Oncology Group data revealed 86 patients with primary ovarian carcinoma and 24 patients with a secondary ovarian neoplasm. Numerous imaging features that had been recorded for the adnexal masses with each imaging modality were reviewed and compared between primary and secondary malignant ovarian neoplasms. RESULTS: Of the imaging features assessed with all three modalities, multilocularity as determined at US (P =.02) or MR imaging (P: =.01) was the only significant feature. At US, 30 (37%) of 81 primary ovarian cancers were multilocular, whereas only three (12%) of 24 metastatic neoplasms were multilocular. At MR imaging, 40 (74%) of 54 primary ovarian cancers were multilocular, whereas only five (36%) of 14 metastatic neoplasms were multilocular. Neither a predominately solid appearance nor bilaterality was significantly different between primary and secondary neoplasms. CONCLUSION: For malignant ovarian masses, multilocularity at MR imaging or US favors the diagnosis of primary ovarian malignancy rather than secondary neoplasm, but it is difficult to accurately distinguish between primary and secondary ovarian malignancies.

Advances in biomedical imaging.

In the 21st century, diseases will be diagnosed and treated using increasingly less invasive, more sophisticated imaging and image-guided procedures. During the past 100 years, the field of biomedical imaging has developed from Roentgen's original discovery of the x-ray to the imaging tools of today, such as magnetic resonance imaging, computed tomography, positron emission tomography, and ultrasonography. The benefits of using these sophisticated noninvasive imaging tools are already evident: more accurate and timely diagnosis of disease has translated into improved patient care. Recent advances in imaging research have shown the potential to change many aspects of clinical medicine within the next decade. Major new areas of research focus on development of the molecular, functional, cellular, and genetic imaging tools of the future, aided by new information technology and image fusion/integration capabilities. Image-guided therapy is growing rapidly, with advances in computer science, technology, and noninvasive treatment methods, such as focused ultrasonography. Undoubtedly, these and other new imaging techniques will enhance the ability to accurately diagnose and recognize disease and allow understanding of the molecular mechanisms of diseases and their respective responses to therapy. Given this explosion in new technologies, the next 25 years promise to result in dramatic changes in diagnostic imaging, particularly with respect to detection and recognition of disease.

Evaluation of three-dimensional finite element-based deformable registration of pre- and intraoperative prostate imaging.

In this report we evaluate an image registration technique that can improve the information content of intraoperative image data by deformable matching of preoperative images. In this study, pretreatment 1.5 tesla (T) magnetic resonance (MR) images of the prostate are registered with 0.5 T intraoperative images. The method involves rigid and nonrigid registration using biomechanical finite element modeling. Preoperative 1.5 T MR imaging is conducted with the patient supine, using an endorectal coil, while intraoperatively, the patient is in the lithotomy position with a rectal obturator in place. We have previously observed that these changes in patient position and rectal filling produce a shape change in the prostate. The registration of 1.5 T preoperative images depicting the prostate substructure [namely central gland (CG) and peripheral zone (PZ)] to 0.5 T intraoperative MR images using this method can facilitate the segmentation of the substructure of the gland for radiation treatment planning. After creating and validating a dataset of manually segmented glands from images obtained in ten sequential MR-guided brachytherapy cases, we conducted a set of experiments to assess our hypothesis that the proposed registration system can significantly improve the quality of matching of the total gland (TG), CG, and PZ. The results showed that the method statistically-significantly improves the quality of match (compared to rigid registration), raising the Dice similarity coefficient (DSC) from prematched coefficients of 0.81, 0.78, and 0.59 for TG, CG, and PZ, respectively, to 0.94, 0.86, and 0.76. A point-based measure of registration agreement was also improved by the deformable registration. CG and PZ volumes are not changed by the registration, indicating that the method maintains the biomechanical topology of the prostate. Although this strategy was tested for MRI-guided brachytherapy, the preliminary results from these experiments suggest that it may be applied to other settings such as transrectal ultrasound-guided therapy, where the integration of preoperative MRI may have a significant impact upon treatment planning and guidance.

Optimizing target coverage by dosimetric feedback during prostate brachytherapy.

PURPOSE: Postimplant dosimetry of permanent prostate implants shows a loss of coverage compared to the preplan. One contributing factor is needle misplacement. The significance of needle misplacement and the clinical utility of dosimetric feedback were analyzed in the setting of interventional magnetic resonance (IMR) guided prostate brachytherapy. METHODS AND MATERIALS: Information provided by an intraoperative planning system was analyzed for 10 patients. Needle misplacement was measured and the dosimetric consequences calculated. Additional catheters and sources were placed following the insertion of all planned catheters to compensate for nonideal needle placement. RESULTS: Source misplacement ranged from 0.0 to 1.0 cm (median, 0.3 cm). The resulting loss of coverage ranged from 1% to 13%, and the intraoperative dosimetric feedback allowed a recovery of from 0% to 12% coverage. Between 0 and 3 (median, 2) additional needles and from 0 to 10 (median, 8) additional sources were required to restore coverage of the target. Final planned coverage exceeded 94% for all patients. CONCLUSION: The discrepancy between planned and achieved needle placement leads to a loss of dosimetric coverage of the target volume. Dosimetric feedback allows compensation for needle divergence. The technique of real-time dosimetric feedback does not require an IMR system, and could be generalized to ultrasound-guided implants.