Upstate New York Association of Physicists in Medicine, Inc.      (A Chapter of the AAPM)
Spring Meeting -
Wednesday May 2nd, 2012

Weiskotten Hall

SUNY Upstate Medical University

Syracuse, NY


                                MEETING SPONSORS



                                MEETING PROGRAM




Business Meeting


Lunch (Lunch Sponsor: Sun Nuclear Corporation)                                                                                                    


Meeting Introduction

Robert Meiler PhD, UNYAPM President

Proffered Paper Session


Preconditioned Alternating Projection Algorithms for Maximum a Posteriori ECT Reconstruction

Andrzej Krol1

SUNY Upstate Medical University


Dynamic susceptibility contrast MRI for brain tumor: a closer look at the signal change

S Yee

SUNY Upstate Medical University


SBRT Treatment of NSCLC with VMAT in comparison to 3DCR*

Caitlin E. Doring, B.Sc.

Roswell Park Cancer Institute


A robust 2D correction of predicted EPID response for pre-treatment IMRT verification*

Daniel W Bailey, PhD

Roswell Park Cancer Institute


Enhancement of Radiation Dose Effects by Gold Nanoparticles and Iodine for superficial radiation therapy : Experimental study*

Donghyun Kim, MS

Roswell Park Cancer Institute


Beam/Beamlet Weight Optimization with Approximate Dose Values for Enhanced Computational Efficiency*

Jason Spaans, MS

Roswell Park Cancer Institute


Refreshments and Vendor Exhibits                        

* Young Investigators' Competition

Sponsored Talks and Competition Winner Announcement


Calculating Organ Doses for Adult and Pediatric CT Scans Using KERMA Ratios

Kent M. Ogden, PhD

SUNY Upstate Medical University


Topics in CTPA: Computed Tomography for Pulmonary Angiography

Ernest Scalzetti, MD

SUNY Upstate Medical University




Directions to Weiskotten Hall

Coming From the West

• NYS Thruway (I-90) Exit 39 (Syracuse)

• 690 East to 81 South to Adams/Harrison exit (#18)

• Keep left off exit ramp. (Almond Street, under Rt. 81)

• Left at second light (E. Adams Street) - University Hospital is on right. Continue past

University Hospital, uphill to traffic light at Irving Ave.

• Turn Right onto Irving. At first traffic light, Weiskotten Hall is on the right, set back

from the Waverly Ave. intersection.

Coming From the East

• Thruway Exit 34A (Rte. 481)

• 481 South to 690 West to the Downtown/Townsend St. exit (#13)

• Left off the ramp onto Townsend St.

• Left at sixth light (Adams Street)

• Continue past University Hospital, uphill to traffic light at Irving Ave.

• Turn Right onto Irving. At first traffic light, Weiskotten Hall is on the right, set back

from the Waverly Ave. intersection.

Coming From the North:

• Route 81 South to Adams/Harrison exit (#18)

• Keep left. You will go under Rt. 81. Left at second light (Adams Street)

• Continue past University Hospital, uphill to traffic light at Irving Ave.

• Turn Right onto Irving. At first traffic light, Weiskotten Hall is on the right, set back

from the Waverly Ave. intersection.

Coming From the South:

• Route 81 North to Adams/Harrison exit (#18)

• Right on Adams Street

• Continue past University Hospital, uphill to traffic light at Irving Ave.

• Right onto Irving. At first traffic light, Weiskotten Hall is on the right, set back from the

Waverly Ave. intersection.

Walking from Parking Garage: If you are walking from the parking garage to Weiskotten Hall:

Walk up Adams Street to Irving Avenue. Turn right onto Irving. Weiskotten Hall is just past Crouse

Hospital on the right.

Accessible Parking: Available in a metered lot located between Weiskotten Hall and the Veterans’

Administration Hospital









1) Preconditioned Alternating Projection Algorithms for Maximum a Posteriori ECT Reconstruction


Andrzej Krol1, Si Li2, Lixin Shen3, Yuesheng Xu3,2 and David Feiglin1


1Department of Radiology, SUNY Upstate Medical University

2School of Mathematics and Computational Sciences, Sun Yat-sen University,
Guangzhou 510275, China

3Department of Mathematics, Syracuse University


There is a great need to reduce radiation dose to the patients undergoing the emission computed tomography (ECT) examinations. This could be accomplished by lowering the total amount of activity in the radiotracer administered. However, it would lead to very high Poisson noise in the raw ECT data.  In turn, such very noisy data if treated by conventional techniques, such as EM-TV or OSEM, would result in very noisy and clinically unacceptable reconstructed images. To attain good quality ECT reconstructions from low-dose ECT examinations, we introduce a generalized fixed-point formulation of the total variation (TV) regularized maximum a posteriori  (MAP) ECT reconstruction problem.  Based on this formulation, we propose preconditioned alternating projection proximity algorithms for computing the fixed point. We theoretically prove the convergence for special cases of our proposed algorithms.

In numerical experiments, performance of our algorithms, with an appropriately selected preconditioning matrix, is compared with performance of the conventional MAP expectation-maximization (MAP-EM) algorithm with TV regularizer (EM-TV) and performance of the  recently developed nested EM-TV algorithm for ECT reconstruction.


Based on the numerical experiments performed in this work, we observe that the alternating projection proximity algorithm with the EM-preconditioner very significantly outperforms the benchmark EM-TV in both the convergence speed, the noise in the reconstructed images  and the image quality. It also slightly outperforms the nested EM-TV in both the convergence speed and the image quality.


We conclude that the developed alternating projection proximity algorithm with the EM-preconditioner might allow very significant reduction in the radiation dose to the patients imaged using ECT by providing the same contrast-to-noise ratio for hot and cold lesions as conventional EM-TV but with the total administered radiotracer activity 2 to 6 times lower than presently used standard-of-care  ECT examinations.



2) Dynamic susceptibility contrast MRI for brain tumor: a closer look at the signal change


S Yee1 and S Hahn2 (1) Radiology, (2) Radiation Oncology, SUNY Upstate Medical University,

Syracuse, NY


Purpose: The signal change in the dynamic susceptibility contrast (DSC) MRI technique is mainly caused by T2* changes accompanied by the passage of contrast bolus through the local vasculature. However, this assumption easily fails in the vicinity of brain tumor, where, due to the disrupted microvasculature, the contrast molecules can leak into the extravascular, extracellular space, resulting in significant changes in T1. The purpose of this abstract is, instead of discarding the non-complying signal change patterns caused by the reduced T1, to legitimately identify such signal change patterns in the vicinity of  tumor, and to qualitatively make sense of them in terms of the degree of contrast leakage, which might be linked to the disease    progression or to the response to anti-angiogenesis therapy.


Methods: DSC-MRI (with manually injected bolus of multihance, 0.2 ml/kg) was applied for high grade brain tumor. The scan technique was based on T2*-weighted PRESTO implemented on a Philips 1.5T scanner. The region of interest (ROI) was drawn on the post contrast T1 weighted image (Fig 1A) and the signal changes were obtained from the DSC-MRI. A simulation (Fig 1C and 1D) was also performed to include T1 factors in the modeling.


Results: The failure of conventional modeling is clearly identified as in Fig 1B, where the signals in some ROIs overshoot beyond the baseline value, which is theoretically prohibited if considered based on only R2* change as in Eqs 1. The simulation shows the non-complying patterns are in deed related with significant T1 changes, which might be linked to the degree of blood brain barrier disruption.


Conclusion: As the importance of DSC-MRI increases as a tumor therapy response monitoring tool, the new approach to include non typical signal patterns would be a valuable addition to overcome the limitation of conventional DSC- MRI.



3) SBRT Treatment of NSCLC with VMAT in comparison to 3DCRT


Authors: Ms. Caitlin E. Doring, B.Sc., Matthew B. Podgorsak, Ph.D. and Z. Iris Wang, Ph.D.


Purpose: To demonstrate the dosimetric potential of volumetric modulated arc therapy (VMAT) for the treatment of patients with medically inoperable stage I/II non-small cell lung cancer (NSCLC) with stereotactic body radiation therapy (SBRT).

Material and Methods: Fourteen patients treated with 3D-CRT with varying tumor locations, tumor sizes and dose fractionation schemes were chosen for study. The target prescription doses were 48 Gy in 4 fractions, 52.5 Gy in 5 fractions, 57.5 Gy in 5 fractions and 60 Gy in 3 fractions for 2, 5, 1 and 6 patients, respectively. VMAT treatment plans with a mix of 2-3 full and/or partial non-coplanar arcs with 5°-25° separations were retrospectively generated using Eclipse version 10.0. The 3D-CRT and VMAT plans were then evaluated by comparing their target dose, critical structure dose, high dose spillage, and low dose spillage as defined according to RTOG 0813 and RTOG 0236 protocols.


Results: The VMAT treatment plans yielded an average 9.6-33.7% reduction in dose to critical structures and an average 12.0-12.5% increase in conformity compared with the treated 3D-CRT plans. The D2cm improved with VMAT in 11 of 14 cases. The 3 that worsened were still within the acceptance criteria. Of the 14 3D-CRT plans, 7 had a D2cm minor deviation, while only one of the 14 VMAT plans had a D2cm minor deviation. The R50% improved in 13 of the 14 VMAT cases. The 1 case that worsened was still within the acceptance criteria of the RTOG protocol. Of the 14 3D-CRT plans, 7 had an R50% deviation. Only 1 of the 14 VMAT plans had an R50% deviation, but it was still improved compared to the 3D-CRT plan.


Conclusion: In this cohort of patients, no dosimetric compromises resulted and dosimetric improvements were seen from planning SBRT treatments with VMAT relative to the 3D-CRT treatment plans. 






4) A robust 2D correction of predicted EPID response for pre-treatment IMRT verification


Daniel W Bailey, PhD, Lalith Kumaraswamy, MS, Matthew B Podgorsak, PhD


**We request that this submission be considered for the Young Investigator Competition at the UNYAPM Spring 2012 meeting. The presenting author is currently in his first year of the medical physics residency at Roswell Park Cancer Institute, Buffalo NY.


Purpose: Predicted electronic portal imaging device (EPID) response, as calculated by a commercial treatment planning system (TPS), is up to 15% lower than measured EPID response for off-axis IMRT fields. Two original algorithms are presented to correct for EPID prediction errors. The EPID prediction algorithm and a recent image-to-dose conversion algorithm are each tested for ability to identify TPS dose calculation errors.


Method and materials: By comparing test images to respective predictions, correction factors were calculated to modify the EPID diagonal calibration profile (applied via radial symmetry). Secondly, image/prediction comparisons were used to compute a 2D correction matrix for EPID predictions, to account for radially-asymmetric errors. Over 50 IMRT fields of varying complexity were tested with each correction technique, and with a diode array. Absolute dose and beam-profile errors were separately induced into the TPS and a number of IMRT plans were recalculated and measured with three systems – an EPID prediction system, an EPID image-to-dose conversion system, and a diode array – for comparison to verification plans.

Results: With the profile correction, TPS predictions agree much better with EPID measurements, yielding improvement in gamma pass rates (3%,3mm) of over 30% on average for off-axis IMRT fields. Since off-axis prediction errors are not radially-symmetric, the matrix correction further improves pass rates by 5% on average (up to 30%) for fields where the profile correction is limited. The EPID prediction system was unable to catch either induced TPS error, while both the image-to-dose conversion system and the diode array indicated both errors.


Conclusions: Profile correction is effective and efficient though approximate, due to radial symmetry. The matrix correction is comprehensive but requires computational manipulation of DICOM images. Users must be aware that EPID prediction systems may be unable to catch delivered IMRT inaccuracies due to calculation errors downstream from the actual fluence calculation.



5) Enhancement of Radiation Dose Effects by Gold Nanoparticles and Iodine for superficial radiation therapy : Experimental study


Donghyun Kim1,2, Zhou Wang1,2 1 SUNY at Buffalo, 2 Roswell Park Cancer Institute


Purpose: High atomic number (Z) materials have been considered as a method for enhancing radiation dose in tumors. The dose enhancement due to interactions of kilovoltage x-rays with high-Z materials (i.e., gold or iodine) has been well demonstrated through computational works. This study is to experimentally quantify the effect using gold nanoparticles (AuNPs) and iodine solutions, respectively.

Method and Materials: Iodine and AuNPs (AuroVist, Nanoprobes, Yaphank NY) are uniformly distributed in each cylinder phantom (1.6 cm diameter and 2.0 cm depth) separately. Concentrations of Iodine and AuNPs were varied from 40 to 225 mg/ml and 16.0 mg/ml to 62.25 mg/ml, respectively. The Iodine solutions were irradiated with 75 to 150 kVp x-rays from a superficial x-ray therapy machine at doses of 250 to 400 cGy. The AuNPs solutions in the cylinder were irradiated with 40 to 150 kVp x-rays. The phantom was placed at the center of the cone to ensure a uniform radiation field. Radiation doses were measured using GafChromic EBT2 films (International Specialty Products, Wayne NJ). Dose enhancement factors (DEF), i.e., the ratio of dose to high-Z material versus dose to water, were calculated and plot as functions of concentration and kVp.

Results: Experimental DEFs varied between 1.01 and 1.38. The DEFs increased with increasing concentration and varied with changing kVp. The maximum DEF measured for Iodine solution was 1.38 at 225 mg/ml and 150 kVp. The maximum DEF measured for AuNPs was 1.32 at 62.25 mg/ml and 40 kVp. The volume of dose enhancement over the target were decreased with increasing concentration and varying kVp. The target volume covered by dose enhancement is larger with relatively lower concentrations.


Conclusions: The magnitude of dose enhancement due to AuNPs and Iodine presence has a dependency with the concentration and kVp. In clinical applications, the concentration of the high-Z material and kVp should also be selected to suit the depth and volume of the target tumor.



6) Beam/Beamlet Weight Optimization with Approximate Dose Values for Enhanced Computational Efficiency


Jason Spaans, MS, Harish Malhotra, PhD, Minerva Ringland, and Daryl P. Nazareth, PhD


Purpose:  To introduce a novel method that increases the efficiency of beam and beamlet weight and fluence optimization, by employing approximate dose values in the objective function. 


Methods: A least-squares objective function was used, based on clinical dose-volume objectives.  In the initial stage of optimization, the method utilizes dose values determined by the additive dose approximation (ADA)  to evaluate the objective function.   These dose values are based on pre-computed dose indices, along with a novel correction scheme.  During this stage, no calculation of the true dose is required, resulting in a significant reduction in computation time, since the manipulation of large dose matrices is avoided.  The resulting optimized parameters are subsequently used as the starting point in the final optimization step. This method was evaluated  on prostate bed, esophageal, and brain patients with varying numbers of treatment fields, by comparing the CPU time required with that of a standard optimization method.    The downhill simplex algorithm was used for the evaluations. 


Results: Nine prostate bed patients with three different beam configurations were used: 4 field 3D-CRT plans, 16 field 3D-CRT plans and a 64-beamlet IMRT-like plan.  These calculations demonstrated an average reduction in optimization time of  33%, 54%, and 46%, respectively.  Two 3D-CRT esophageal patients were evaluated, one with 14 fields and one with 16 fields, resulting in an average improvement of 29%.  Finally, a 3D-CRT 16 field brain patient was studied and exhibited a 57% decrease in CPU time. 


Conclusions:  We have developed a method to be used in conjunction with existing optimization algorithms, which significantly decreases the number of standard objective function evaluations and, as a result, computation time.  This technique may be applied to any type of optimization algorithm and set of clinical variables requiring pre-computed dose matrices, and an objective function involving dose-volume objectives (e.g., multi-criteria optimization).









7) Calculating Organ Doses for Adult and Pediatric CT Scans Using KERMA Ratios


Kent M Ogden, PhD, SUNY Upstate Medical University


CT scanner-reported dose metrics include the Computed Tomography Dose Index (CTDI) and the Dose-Length Product (DLP), which do not directly represent organ or effective dose and must be converted using factors typically calculated using Monte Carlo methods.  This talk will describe a method of calculating organ doses in patients from newborn to adult, based on KERMA ratio (RK).  RK is defined as RK  = K/KCT, where K is the KERMA measured at specific points in anthropomorphic phantoms, and KCT is the in-air KERMA at the CT scanner isocenter for an equivalent technique.

RK values have been published for anthropomorphic phantoms ranging from newborn to 10 y/o (CIRS) phantoms and in an adult (Rando) phantom.  KERMA measurements were made using thermoluminescent dosimeters (TLD’s) for 16 and 4 slice CT scanners from two manufacturers.  RK values were measured in 21 different organs/tissues and for varying kV values.  At 120 kV, median RK values ranged from 0.92 for newborns to 0.6 for adults.  Values are generally higher in smaller body parts and approach a value of 1.0 in the neck region in pediatric phantoms.

KERMA Ratios provide a simple and elegant method for determining organ dose from easily measure scanner outputs and then scaling for the scan technique used.  Application to the estimation of effective dose (E) will be discussed, as well as limitations present in the currently published data.  Future work will be discussed and includes measurements made using a 320 slice CT scanner.


8) Topics in CTPA: Computed Tomography for Pulmonary Angiography


Ernest Scalzetti, MD, SUNY Upstate Medical University


A. Repeated Referrals for CT Pulmonary Angiography and Cumulative Radiation Dose: Longitudinal Follow-up of a Single-Institution Cohort


Introduction: Many patients have a CT pulmonary angiogram (CTPA) to evaluate for acute pulmonary embolism (PE). We performed a longitudinal followup study of a cohort of such patients who had a CTPA, to determine how often they receive additional CTPAs in successive years and their cumulative radiation dose.


Methods: We received an IRB waiver for review of patient records. We identified all patients referred for CTPA, for evaluation of acute PE, in 2005. All were followed through 2009 for additional CTPAs at our institution. We recorded technical details of each CTPA to permit estimation of an effective radiation dose. The scans were reviewed by an experienced thoracic radiologist, cumulative effective doses were calculated, and additional clinical information was retrieved for all patients who had >2 CTPAs.


Results: 650 patients received at least one CTPA in 2005. Through 2009, 126 of them had >1 CTPA (19.4%), 38 had >2 (5.9%) and 17 had >3 (2.6%). Median effective dose per scan was 4.3 mSv (range 2.6-8.0 mSv). The largest number of CTPAs was 17, in a 33 year old female who had no predisposing condition for PE, for a cumulative effective dose of 60 mSv. The largest cumulative dose was estimated at 112 mSv in a 44 year old female patient who underwent 14 CTPAs; she also had no predisposing condition for PE. Of the 38 patients who had >2 CTPAs, 23 were female (61%, p=NS), the median age was 51 years (range 28-87), median cumulative effective dose was 13 mSv (range 8-112 mSv), and 11 (6%) of the 173 total CTPAs showed an acute PE. Of the 5 patients who had >6 CTPAs, all were female, the median age was 40 years (range 33-48 years), the median cumulative effective dose was 60 mSv (range 29-112 mSv), and none of the 58 total CTPAs was positive for acute PE.


Conclusion: Fewer than 20% of patients received a repeat CTPA during a followup period of at least 4 years. Patients who were referrred repeatedly and incurred the largest cumulative doses tended to be females of younger age; very few of these scans led to a diagnosis of acute PE. Cumulative effective doses exceeded 50 mSv in 4 patients. Multiple CTPAs can lead to clinically meaningful radiation doses with little or no clinical benefit.


B. Feasibility of a Modified Circulation Time Method to Select Prescan Delay Time for CT Pulmonary Angiography


Introduction: CT pulmonary angiography (CTPA) requires scanning during peak enhancement of pulmonary arteries (PAs). The delay between the beginning of contrast infusion and the onset of the CTPA may be based on bolus tracking, or calculated from a set of circulation time images. We demonstrate the feasibility of a modified circulation time (MCT) method.


Methods: The IRB granted permission to review patient records. Data were recorded from consecutive patients referred for CTPA. CTPA was performed on 16 and 64 section helical CT scanners. The MCT method provides a set of circulation time images made under conditions as similar as possible to the subsequent CTPA. MCT images were obtained through the main PA during infusion of 75 cc of contrast: a blend of 25 cc Omnipaque 350 with 50 cc saline using a dual-headed injector at a rate of 5 cc/s, followed by 30 cc saline. The prescan delay for CTPA was calculated by subtracting the expected CTPA duration from the time at which contrast began to wash out from the main PA. The CTPA consisted of 1.25 mm sections made after infusion of 75 cc Omnipaque 350 at 5 cc/s followed by 30 cc saline. Magnitude of peak enhancement and time-to-peak were measured on each MCT series and compared to the main PA enhancement of the CTPA. An experienced chest radiologist subjectively assessed the quality of CTPA enhancement.


Results: There were 50 patients. Mean age was 52 years (range 21-86 years) and mean body weight 87 kg (range 46-168 kg). Peak MCT enhancement (mean ± std dev) was 183 HU ± 36 HU; mean time-to-peak was 15 s ± 3 HU. Prescan delays implemented by the CT technologists were 16 s ± 4 s (range 7-23 s). Prescan delays determined after the fact by the reviewing radiologist were 16 s ± 3 s (range 11-22 s); the median discrepancy was 0 s. Main PA enhancement on CTPA was 437 HU ± 129 HU (range 190-721 HU). Segmental and larger PAs were evaluable in all cases; in 4 cases (8%) subsegmental PAs were not evaluable, corresponding to main PA enhancement of <264 HU.


Conclusions: A modified method of calculating prescan delays for CTPA was implemented successfully in our practice. It allowed assessment of PAs to the subsegmental level in >90% of cases. It appears to translate well into clinical practice.