Faculty Bibliography

PURPOSE: To report design of a simplified external transmit-receive coil array for 7 Tesla (T) prostate MRI, including demonstration of the array for tumor localization using T2-weighted imaging (T2WI) at 7T before prostatectomy. MATERIALS AND METHODS: Following simulations of transmitter designs not requiring parallel transmission or radiofrequency-shimming, a coil array was constructed using loop elements, with anterior and posterior rows comprising one transmit-receive element and three receive-only elements. This coil structure was optimized using a whole-body phantom. In vivo sequence optimization was performed to optimize achieved flip angle (FA) and signal to noise ratio (SNR) in prostate. The system was evaluated in a healthy volunteer at 3T and 7T. The 7T T2WI was performed in two prostate cancer patients before prostatectomy, and localization of dominant tumors was subjectively compared with histopathological findings. Image quality was compared between 3T and 7T in these patients. RESULTS: Simulations of the B1 + field in prostate using two-loop design showed good magnitude (B1 + of 0.245 A/m/w1/2 ) and uniformity (nonuniformity [SD/mean] of 10.4%). In the volunteer, 90 degrees FA was achieved in prostate using 225 v 1 ms hard-pulse (indicating good efficiency), FA maps confirmed good uniformity (14.1% nonuniformity), and SNR maps showed SNR gain of 2.1 at 7T versus 3T. In patients, 7T T2WI showed excellent visual correspondence with prostatectomy findings. 7T images demonstrated higher estimated SNR (eSNR) in benign peripheral zone (PZ) and tumor compared with 3T, but lower eSNR in fat and slight decreases in tumor-to-PZ contrast and PZ-homogeneity. CONCLUSION: We have demonstrated feasibility of a simplified external coil array for high-resolution T2-weighted prostate MRI at 7T.J. Magn. Reson. Imaging 2013. (c) 2013 Wiley Periodicals, Inc.

Introduction and Objectives: With improving accuracy of prostate cancer detection and localization using multi-parametric MRI (mpMRI), there is an increasing interest in mpMRI guidance of diagnosis and surveillance biopsy. Widespread application of such concept must address the challenging issues, including the ability to perform MRI-guided biopsy in realtime with adequate accuracy, and in the simple urology office environment. We propose and assess a new approach a new approach to directly coregister 2D standard TRUS to MRI. Materials and Methods: The developed concept is to use a raw 2D Ultrasound (US) (B&K 8848 device) prostate image and register it to the corresponding MRI slice. Pre-acquired MRI data represents the whole gland in 3D as an ordered collection of 2D MRI slices of known thickness. We have developed software for USMRI coregistration based on image intensity, texture, and boundaries. The power parameter P is directly proportional to algorithm speed and accuracy. The result is source US overlaying target MRI for visualization. The system was prospectively tested on 8 data sets corresponding to US images matching with prostate MRI. These data come from patients who underwent MRI prior to biopsy. Coregistration results were evaluated as success/failure by an expert urologist who interactively adjusted the transparency of US-MRI overlays and recorded the alignment of anatomical landmarks in both modalities, especially the veru montanum. Results: The system was able to find the matching slice of T2WI for all single 2D standard US slice. In all cases the location of the veru montanum confirmed the coregistration accuracy in axial plan. The median rank of the T2WI slice was the fifth one among median number of 10 T2WI slices. We tested the power parameter P=1; P=5 and P=20. Twenty three coregistrations over 23 were correct. There was a significant positive correlation between prostate volume and time of computation for each P value. The image similarity reached a 0.93 mean Dice index u!

Introduction and Objectives: Focal therapy (FT) is an emerging approach for treatment of localized prostate cancer. Multi-parametric (mp) MRI demonstrated capability to monitor the effect of FT procedures. At time of followup, the ablated zone (AZ) clearly undergoes local shrinkage. Accurate definition of AZ at follow-up will improve FT evaluation and oncologic safety. We analyzed the volume and shape changes of the gland between pre and post FT MRI by developing a 3D coregistration method to compensate for deformation of the gland in response to FT. Materials and Methods: We studied 10 patients who underwent FT (interstitial laser ablation and photodynamic therapy) within IRB approved trials. All patients underwent preoperative, early control and late postoperative 3T MRI which included T2, T1, diffusion and perfusion weighted sequences. We have developed image registration software to analyze, transfer and model shape changes using a deformable and a rigid body transformation. Alignment between pre- and post-op images of AZ was assessed using the overlap index or Dice index (Di) and the maximum boundary distance, or Hausdorff distance (HD). Correction for deformation was measured using the HD normalized by the volume to transform in mm/ cc and automated feature of the software. Results: There was a significant volume decrease D of the gland that averaged 6.49 cc (p=0.017) between preoperative and postoperative gland. D was directly correlated (=0.738, p=0.014) with the ablated volume (7.88, p=0.04). We successfully co registered pre operative to post operative MRI in each cases. There was a significant increase D computed with deformable versus rigid transform. Deformable model achieved a significantly more accurate match of pre- vs. post-FT AZ with deformable (Di=0.88, HD=1.98 mm) vs. rigid transformation (Di=0.95 HD=3.83mm). The deformable approach also yielded a higher (p=0.019) correction of deformation (0.72mm/cc) compared to the rigid model (0.15mm/cc). Conclusion: We described a novel !

INTRODUCTION: To compare qualitative and quantitative imaging features from conventional and diffusion-weighted (DW) magnetic resonance imaging (MRI) in detection of metastatic pelvic lymph nodes in bladder cancer patients undergoing cystectomy. MATERIALS AND METHODS: Thirty-six patients who had undergone cystectomy for bladder cancer with preoperative MRI with DWI sequence prior to surgery were included. Imaging features on conventional and DW-MRI were compared with histopathology at cystectomy. RESULTS: Nodal features associated with metastatic lymphadenopathy were short axis (AUC = 0.85, p < 0.001; when SA > 5 mm: sensitivity = 88%, specificity = 75%), long axis (AUC = 0.80, p < 0.001; when LA > 6 mm: sensitivity = 88%, specificity = 71%), apparent diffusion coefficient (ADC) on DWI, normalized to muscle (AUC = 0.66, p = 0.113; when nADC < 1.35: sensitivity = 75%, specificity = 68%), and absence of fatty hilum on conventional imaging (AUC = 0.73, p = 0.012; when fatty hilum absent, sensitivity = 75%, specificity = 71%). ADC without normalization was not associated with metastasis (p = 0.303). CONCLUSIONS: Imaging findings from conventional MRI and DWI achieved reasonable accuracy for detecting metastatic lymph nodes in bladder cancer, although sensitivity was higher than specificity. A short axis greater than 5 mm on conventional MRI had the highest accuracy of any individual finding. When using DWI, normalization of ADC values to muscle ADC may improve diagnostic performance.

OBJECTIVE: To assess recent national specialty trends in the use of diagnostic ultrasound (US) services in the Emergency Department (ED) setting. METHODS: We searched aggregated 1998-2012 Medicare Part B Master Files for ED diagnostic US studies, excluding cardiac and ophthalmic examinations. Studies were classified by body part and interpreting specialty. Subsequent analysis was performed for higher-volume services rendered by emergency physicians for which discrete codes were present longitudinally for complete vs limited examinations. National trends were analyzed. RESULTS: From 1998 to 2012, paid ED US studies interpreted by radiologists, emergency physicians, and all other physicians increased by 332% (from 221 712 to 735 858 examinations), 4454% (from 561 to 24 992), and 251% (from 26 961 to 67 787), respectively. The fraction of ED US examinations interpreted remained around 90% for radiologists, increased from 0.2% to 3% for emergency physicians, and decreased from 11% to 8% for other physicians. The fraction of complete abdominal and complete retroperitoneal studies interpreted by emergency physicians remained less than 1% from 1998 through 2012. However, emergency physicians experienced disproportionate growth in limited examinations (from 1% to 9% for abdominal studies and from <1% to 20% for retroperitoneal studies). Likewise, the fraction of (typically targeted) chest studies interpreted by emergency physicians increased from less than 1% to 63%. CONCLUSION: From 1998 to 2012, there was substantial growth in ED US studies for Medicare beneficiaries interpreted by radiologists and emergency physicians alike. For more commonly performed services distinguishable as complete vs limited in nature, growth in services by emergency physicians was most dramatic for less complex services.

AIM: To compare the size and shape of the prostate between in-vivo and fresh ex-vivo magnetic resonance imaging (MRI), in order to quantify alterations in the prostate resulting from surgical resection. MATERIAL AND METHOD: Ten patients who had undergone 3 T prostate MRI using a phased-array coil and who were scheduled for prostatectomy were included in this prospective study. The ex-vivo specimen underwent MRI prior to formalin fixation or any other histopathological processing. Prostate volume in vivo and ex vivo was assessed using planimetry. Prostate shape was assessed by calculating ratios between the diameters of the prostate in all three dimensions. RESULTS: Mean prostate volume was significantly smaller ex vivo than in vivo (39.7 +/- 18.6 versus 50.8 +/- 26.8 cm3; p = 0.008), with an average change in volume of -19.5%. The right-to-left (RL)/anteroposterior (AP) ratio of the prostate, representing the shape of the prostate within its axial plane, was significantly larger ex vivo than in vivo (1.33 +/- 0.14 versus 1.21 +/- 0.12; p = 0.015), with an average percent change in RL/AP ratio of the prostate of +12.2%. There was no significant difference between in-vivo and ex-vivo acquisitions in terms of craniocaudal (CC)/AP (p = 0.963, median change = -2.1%) or RL/CC (p = 0.265, median change = +1.3%) ratios. CONCLUSION: The observed volume and shape change following resection has not previously been assessed by comparison of in-vivo and fresh ex-vivo MRI and likely represents loss of vascularity and of connective tissue attachments in the ex-vivo state. These findings have implications for co-registration platforms under development to facilitate improved understanding of the accuracy of MRI in spatial localization of prostate tumours.