COMPARISON OF THREE MAGNETIC RESONANCE IMAGING SEQUENCES FOR MEASURING CARTILAGE THICKNESS IN THE CANINE STIFLE by Lawrence Alexander Brown A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama August 4, 2012 Copyright 2012 by Lawrence Alexander Brown Approved by Merrilee Holland, Chair, Department of Clinical Sciences Judith Hudson, Department of Clinical Sciences John T Hathcock, Department of Clinical Sciences Calvin Johnson, Department of Pathobiology ii Abstract Magnetic Resonance Imaging (MRI) using 3 sequences; Proton Density Spin Echo (PDSE), 3 Dimensional Spoiled Gradient with Fat Suppression (3D SPGR-FS) and Steady State Free Precision (SSFP) in the sagittal plane was carried out on 11 normal cadaver stifles. The stifles were imaged within 24hrs postmortem. Three readers obtained cartilage thickness measurements from the images of all sequences. These measurements were compared to those obtained from histological prepared specimens in the sagittal plane. Histology specimens were prepared from the location where MRI images had been obtained. The values for mean cartilage thickness from MRI and histology were compared using regression, correlation, and student?s t test. Pearson?s Correlation was used to assess correlation of cartilage thickness values obtained from magnetic resonance images and histology as well as to assess correlation among readers. There was no significant difference between MRI and histology measurements in 20.2% of the images combined for all sequences. Of these, the Balanced Steady State Free Precision (BSSFP) sequence had the highest number of images that were not significantly different from the histologic measurements (24%). MRI measurements generally overestimated cartilage thickness when compared to histology. There was moderate correlation between histology measurements and MRI images for all readers (R values: 0.25 to 0.77). Correlation between readers for all sequences varied from moderate to poor (R values: 0.25 to 0.77). There was poor coefficient of variations for all image types when compared to histology. All sequences had sufficient contrast resolution to identify the articular cartilage of the femoral condyles. Further investigation of iii these sequences using a higher field strength magnet may allow more accurate evaluation of cartilage thickness by allowing further manipulation of the matrix size and field of view to more accurately represent cartilage thickness in the canine stifle. iv Acknowledgments The author would like to thank Dr. Merilee Holland and Dr. Judith Hudson for their patience in carrying out this project. He would also like to thank Dr. John T. Hathcock for his time spent contributing to the results. A special thanks go to Dr. Calvin Johnson for his time and help explaining the histology preparation techniques. Thanks also go to Dr. James Wright for his time performing the statistical analysis and explaining the results. This project would not have been possible without the technical assistance of Kim Bryan and Terrell Lynch; thank you both. The author also wished to acknowledge the support of his mother Sandra, brother Nicholas for their endless support and sacrifices. I would not be where I am if it were not for you both. The support of Dr. Lauren Reid over the last few years, in good times and bad, has been invaluable. v Table of Contents Abstract ......................................................................................................................................... ii Acknowledgments........................................................................................................................ iv List of Figures ............................................................................................................................. vii I.INTRODUCTION .................................................................................................................... 1 II. LITERATURE REVIEW ....................................................................................................... 3 III. MATERIALS AND METHODS .......................................................................................... 8 Animal Subjects ................................................................................................................ 8 Magnetic Resonance Imaging ............................................................................................ 8 Image Analysis................................................................................................................. 10 Histology Sampling ......................................................................................................... 11 Statistical Analysis ........................................................................................................... 15 IV. RESULTS ............................................................................................................................. 16 V. DISCUSSION ........................................................................................................................ 21 Study Limitations ............................................................................................................. 22 VI. CONCLUSION..................................................................................................................... 26 REFERENCES ......................................................................................................................... 27 APPENDICES ........................................................................................................................... 30 APPENDIX I ............................................................................................................................ 31 Pearson?s Correlation Coefficients ................................................................................. 31 APPENDIX II ........................................................................................................................... 37 vi Analysis of Variance and Regression ............................................................................. 37 APPENDIX III .......................................................................................................................... 40 Histology measurements and Student t test ...................................................................... 40 vii List of Figures Figure 1. Left: PDSE, 3D SPGR FS and SSFP MRI sequences were obtained of the stifles in a sagittal plane. Right: Gross femoral condyles were sectioned in the sagittal plane. .... 9 Figure 2. Cartilage was identified as the layer of increased signal intensity between the subchondral bone which emitted no radiofrequency signal and the synovial fluid which had low signal intensity. ............................................................................................. 11 Figure 3. Gross sagittal sections were cut through each femoral condyle ................................. 13 Figure 4. Histologic section of the articular cartilage of a femoral condyle stained with hematoxylin and eosin (H&E) and toluidine blue to distinguish cartilage and bone. Cartilage thickness measurements, as shown, included the calcified zone. ............... 14 Figure 5. Measurements (mm) from the femoral condyle made using histology and a 3D SPGR FS MRI imaging sequence ......................................................................................... 17 Figure 6. Measurements (mm) from the femoral condyle made using histology and PDSE FS MRI imaging sequence ............................................................................................... 18 Figure 7. Measurements (mm) from the femoral condyle made using histology and a SSFP FS MRI imaging sequence. .............................................................................................. 19 1 I. INTRODUCTION Osteoarthritis (OA) and other diseases affecting the cartilage in the stifle joint of small animals can dramatically affect their quality of life. It is the most common arthropathy in the dog.10 Great emphasis has been placed on non-invasive evaluation of cartilage in human medicine, where the disease processes are similar to those in veterinary patients. Evaluation of cartilage thickness is an important indicator for staging arthritic disease. Studies have been conducted on horses to evaluate the accuracy of magnetic resonance imaging for estimating cartilage thickness;1 however, little research has been published for small animals. The project goal was to assess the diagnostic value of three magnetic resonance imaging sequences, currently used in human medicine, for imaging the canine stifle. The plan was to compare and contrast the ability of these sequences to allow accurate assessment of cartilage thickness in canine stifle cadaver specimens. Measurements of cartilage thickness from each image sequence were made at standardized locations on the femoral condyles. These values were compared to measurements of cartilage thickness obtained from gross dissection and histological analysis of the stifle joints. Statistical analysis using correlation and the student?s t test were carried out to compare MRI measurements of cartilage thickness with those from histological sections. From the results of this study, the goal was to find an MRI sequence which can be used to accurately assess cartilage thickness, and therefore recommend a magnetic resonance imaging sequence and technique that can be added to a standard stifle imaging protocol to best evaluate cartilage degeneration with canine stifle disease. This study tested the following hypotheses: 2 1. Cartilage thickness in dogs measured using MRI is similar to that measured using histopathology. 2. There is no difference in measurements of cartilage thickness obtained using 3 techniques: Proton Density Spin Echo (PDSE), 3 Dimensional Spoiled Gradient with Fat Suppression (3D SPGR-FS) and Steady State Free Precision (SSFP). Specifically, the study sought to assess the ability of three MRI sequences to accurately cartilage thickness in the canine stifle joint and to determine if there is any significant difference between the accuracy of the three sequences. 3 II. LITERATURE REVIEW Lameness is a common clinical presentation in small animal medicine. A common cause of lameness is degenerative joint disease, which includes chronic changes to the cartilage, joint capsule and synovium.2 Typically, degenerative joint disease is diagnosed in small animal patients using the combination of a complete history, physical exam and radiographs of affected joints. Radiographs are excellent at evaluating bone; however, they are extremely poor at allowing cartilage assessment. Clinicians and students at veterinary schools are taught to evaluate radiographs for secondary changes associated with cartilage abnormalities, e.g. narrowing of joint spaces, osteophyte formation and subchondral bone changes. Contrast arthrography can be used to evaluate surface defects, but without invasive arthroscopic procedures or surgical techniques, joint cartilage cannot be directly evaluated. Magnetic Resonance Imaging (MRI) is used in human orthopedic medicine to evaluate cartilage pathology in joint disease3,5,7. This modality - like radiography and computed tomography - is a non-invasive imaging technique; however, it differs by using the concepts of magnetic induction and magnetic resonance to form multiple 2D images with excellent soft tissue contrast. This allows evaluation of both bone and soft tissue structures and contributes to the understanding of the pathophysiology for joint diseases. Degenerative joint disease leads to a decrease in the thickness of articular cartilage. Therapy for this disease process attempts to slow down the rate of degeneration by direct or indirect means. Therapies in small animal medicine include weight control, physical therapy and pharmaceutical / neutraceutical use. Owners and clinicians evaluate patient response through 4 activity levels, pain and body condition score improvement and radiographic assessment of affected joints. The response of cartilage to therapy is not evaluated directly. If an MRI protocol that allows accurate and precise measurement of cartilage thickness can be defined, we would have a method to directly evaluate progression of degenerative joint disease and possibly to determine the efficacy of various treatment options over time. This point is echoed in current human literature where it has been stated that MRI imaging of cartilage may allow for large scale studies of OA progression to be conducted, as well as clinical trials for investigating the efficacy of structure modifying OA drugs.3 MRI research on cartilage evaluation in veterinary medicine is limited compared to human medicine. There has been recent interest in Large Animal (LA) medicine for investigating the ability of this modality to evaluate cartilage. A 2010 study1 evaluated MRI in its ability to assess osteoarthritis in the metacarpophalangeal joint of standardbred and thoroughbred racehorses. Twenty cadaver legs were imaged at the level of the metacarpophalangeal joint using MRI with a single pulse sequence. Specific locations in the joint were selected to evaluate correlation between MRI images and gross measurements. This study reported good precision and moderate correlation (r =0.44; p < 0.0001) between of the two methods of measuring cartilage thickness. In a similar 2005 study,4 32 horse cadaver legs were imaged at the level of the carpal joint using two different pulse sequences. Cartilage thickness measurements of the radial, intermediate and ulnar carpal bones from these sequences were compared with gross measurements. Significant correlation (r=0.96; P<0.001) was reported between the images from different pulse sequences and gross measurements; however, variation between MRI and gross thickness depended on whether or not the histology measurement included the calcified cartilage 5 layer. When the calcified cartilage layer was not included, MRI measurements were significantly greater than gross measurements suggesting that MRI better estimates cartilage thickness when measurements include the calcified cartilage layer. A number of studies have been carried out in human medicine comparing the ability of various pulse sequences to evaluate cartilage thickness in assorted joints. An extensive review of MRI cartilage imaging protocols was presented by the Osteoarthritis Research Society International in 20063. This review concluded that high accuracy and adequate precision for quantitative assessment of cartilage morphology can be obtained using fat-suppressed gradient echo MRI sequences, and appropriate image analysis techniques for cross-sectional and longitudinal studies in osteoarthritis (OA) patients.3 There are a number of pulse sequence used in human medicine that have been found to yield the most accurate results for measuring cartilage thickness. A common sequence used for morphological imaging of cartilage is a three dimensional spoiled gradient echo sequence with fat suppression (3D SPGR-FS). 5,6 Other MRI sequences have been also been investigated. A 2006 publication7 outlined several sequences which address two imaging problems: 1. Conventional sequences provide insufficient spatial resolution and inadequate data about cartilage physiology. 2. Long image acquisition times required for 3D-SPGR sequences require longer periods of patient general anesthesia and increased risk of image degradation secondary to motion artifacts. 6 This study suggested that DEFT (Driven Equilibrium Fourier Transformation), BSSFP (Balanced Steady State Free Precision) and fat water separation sequences allow good assessment of cartilage with acceptable spatial resolution and decreased acquisition times 7. A 2008 study evaluated cartilage thickness and volume changes over time in dogs as a model for OA in humans.18 This study was conducted on five dogs that had surgical transection of their cranial cruciate ligaments. Serial MRI studies of the stifle were then carried out post immediately and at 4, 8 and 26 weeks post cranial cruciate ligament transection. The dogs were euthanized at 26 weeks and the joint cartilage was evaluated grossly for defects. This study reported a general loss of cartilage thickness throughout the time period of the study. Cartilage loss was most pronounced on the tibial plateau. The paper reported good correlation between gross cartilage observations and cartilage volume changes (r = -0.81 p < 0.001) although there was no reported assessment of accuracy of MRI at evaluating cartilage thickness compared to histology measurements. A 2011 study evaluated cartilage volume changes in the canine stifle over time after cranial cruciate ligament transection and extracapsular surgical stabilization19. Thirty one dogs had surgical transection of the cranial cruciate ligament. After 4 weeks, surgical extracapsular stabilization was performed and the dogs were divided into 2 treatment groups for the purpose of evaluating if tiludronic acid was able to decrease the progression of OA. MRI examination of the stifle joints was carried out at 10, 26,91, 210 and 357 days post cranial cruciate ligament transection. This study reported an initial decrease in cartilage volume post cranial cruciate ligament transection which stabilized after extracapsular stabilization. The dogs were euthanized on day 364 and cartilage was examined grossly. This study did not evaluate the accuracy of MRI at evaluating cartilage thickness. 7 To the author?s knowledge, although studies have been performed to assess cartilage volume change over time,18,19 no studies have been carried out to evaluate the ability of MRI to accurately and precisely measure cartilage thickness in dogs. The Auburn University College of Veterinary Medicine currently has an MRI unit with sufficient field strength (1.5T) to obtain images using pulse sequences used in human and veterinary medicine to evaluate cartilage thickness and morphology. 8 III. MATERIALS AND METHODS Animal Subjects Eleven normal pelvic limbs were disarticulated at the level of the coxofemoral joint from 6 cadaver dogs of various breeds, age and gender, with a body weight >15kg. Using power analysis, a population of eleven dogs with a proposed correlation of 0.8 yielded an acceptable power of >0.9. The dogs were previously enrolled in a separate IAACUC approved study at Auburn University College of Veterinary Medicine. The limbs were obtained within 1 hour of euthanasia and immediately refrigerated at approximately +4?C. Image acquisition occurred within 48hrs of euthanasia. Each limb was radiographed to evaluate for evidence of gross joint abnormalities that would necessitate exclusion from the study (e.g. neoplastic disease within the joint or presence of a metallic bone plate). Magnetic Resonance Imaging Each stifle was imaged in extension with the caudal surface in contact with the table. Each stifle joint was placed at the isocenter of a 1.5T MRI unit (Philips Medical Systems, Milwaukee) in a human extremity coil (Quadrature Lower Extremity, Philips Medical Systems, Milwaukee, WI, Model 473P1-64H, Frequency 63.73 MHz). Three separate image sequences were acquired in the sagittal plane through each stifle (Figure 1): 1. Proton Density Spin Echo (PDSE) (slice thickness 2mm, gap 0mm, time to echo (TE) 7.7ms, repetition time (TR) 150.0ms, flip angle 90?, number of acquisitions 2, field of view 220mm, matrix 384 ? 512, pixel size 0.57 ? 0.69mm) 9 2. 3 dimensional spoiled gradient fat suppression (3D SPGR-FS), (slice thickness 2mm, gap 0mm, time to echo (TE) 7.0ms, repetition time (TR) 42.0ms, flip angle 20?, number of acquisitions 2, field of view 220mm, matrix 384 ? 512, pixel size 0.57 ? 0.69mm). 3. Steady State Free Precision (SSFP) (slice thickness 2mm, gap 0mm, time to echo (TE) 8.0ms, repetition time (TR) 500.0ms, flip angle 90?, number of acquisitions 2, field of view 220mm, matrix 384 ? 512, pixel size 0.57 ? 0.69mm). Figure 1. Left: PDSE, 3D SPGR FS and SSFP MRI sequences were obtained of the stifles in a sagittal plane. Right: Femoral condyles were sectioned in the sagittal plane. 10 Image Analysis Images were analyzed at a diagnostic workstation by the author (LB) together with two board certified veterinary radiologists (MH, JTH) independently. Cartilage thickness was measured using the Osirix DICOM viewer software measuring tool. Measurements of cartilage thickness were made at the mid-sagittal point of the medial and lateral femoral condyles. The reason for measuring cartilage thickness at these specific sites was because their spatial location was easy to identify and section during preparation of histological samples. Cartilage was identified as a layer of increased signal intensity located between the subchondral bone which emitted no radiofrequency signal and the synovial fluid which had low signal intensity (Figure 2) as suggested by Olive et al. (2010) in a previous study.20 Measurements were made perpendicular to the bone surface. Three measurements were made at each site and an average value calculated. Repeatability of MRI measurements were determined using standard deviation and a repeatability index based on replications and the residual error. The repeatability index was calculated by subtracting the measurement error percentage from 100%. 11 Figure 2. Cartilage was identified as a layer of increased signal intensity between the subchondral bone which emitted no radiofrequency signal and the synovial fluid which had low signal intensity. Histology Sampling The stifle joints were then prepared for gross cartilage thickness evaluation. After magnetic resonance imaging was completed, the stifles were refrigerated at approximately +4?C with histologic preparation beginning within 48 hours to minimize post mortem changes to cartilage which could affect thickness. Histologic samples were taken from the approximate sites analyzed with MRI. Sample preparation was carried out by the primary investigator after training and supervision from a board certified pathologist (CJ). Four-millimeter-thick osteochondral slices were made through the medial and lateral condyles in a sagittal plane (Figure 3). Samples of bone were histologically processed as described in previous studies8,9. 12 Slices were fixed in 10% buffered formaldehyde (pH 7.4) for 48 hours and then decalcified in a Decalcifier II? (Surgipath Medical Industries, Inc., Richmond, IL). The samples were processed routinely for histology, embedded in paraffin, and sectioned at 5?m thickness, followed by staining with hematoxylin and eosin (H&E) and toluidine blue to distinguish cartilage and bone (Figure 4). 13 Figure 3. Gross sagittal sections were cut through each femoral condyle. 14 Figure 4. Histologic section of the articular cartilage of a femoral condyle stained with hematoxylin and eosin (H&E) and toluidine blue to distinguish cartilage and bone. Cartilage thickness measurements, as shown, included the calcified zone. The samples were then examined by light microscopy and images of cartilage were captured (Q-Capture software, Q-Imaging Corp, Burnaby, BC, Canada) and cartilage thickness measured by digital morphometry (Image J, National Institutes of Health, Bethesda, Maryland, USA). These measurements were carried out by the primary investigator with guidance and training from a board certified veterinary pathologist (CJ). Ten independent measurements were 15 conducted for each specimen and expressed as the median and interquartile range. Data from each group was compared using a two-tailed Mann-Whitney U test. Statistical Analysis Data was entered into a spreadsheet using Microsoft Excel 2010 and analyzed using Statistical Analysis System software (SAS, release version 9.2, Carey N.C.) The values for mean cartilage thickness from MRI and histology were compared using regression, correlation, and student?s t test. Pearson?s Correlation was used to assess correlation of cartilage thickness values obtained from magnetic resonance images and histology as well as to assess correlation among readers. 16 IV. RESULTS MRI images of all eleven stifles using all three of the image sequences (PDSE, 3D SPGR-FS, SSFP) were obtained in the sagittal plane. All images were subjectively assessed as having sufficient signal to noise ratio (SNR) for image analysis. All images had hyperintense signal in the region of the articular cartilage of the femoral condyles. A total of 66 magnetic resonance images were obtained for analysis by the three readers (LB, JTH, MH). Acceptable histological samples were obtained from all condyles of all stifles. Five condyles (B1med, B2 lat, C2 lat, D2 med and F1 lat) were re-sectioned from the samples obtained due to folding artifact on the initial slides. Values for the paired student t test using pooled measurements of the readers showed significant differences between the mean cartilage thickness measured from histologic section and the MRI thicknesses for all sequences (p < 0.05). For individual readers, there was no significant difference in 14/66 images (reader 1), 25/66 images (reader 2) and 1/66 images (reader 3) when comparing the mean cartilage thickness measurements between histology and MR (p>0.05). Of the 40 images for all readers with no significant difference between the cartilage thickness measured by histology and MRI, 16 were SSFP images, 12 PDSE and 4 3D SPGR-FS images (p>0.05). The mean cartilage thickness values for all three MRI sequences pooled for all readers were greater than the mean value obtained from the histological measurements (See figures 5-7). 17 Figure 5. Measurements (mm) from the femoral condyle made using histology and a 3D SPGR-FS MRI imaging sequence. The mean cartilage thickness measured using 3D SPGR-FS images was 0.949mm compared to a mean histological measurement of 0.734mm. 18 Figure 6. Measurements (mm) from the femoral condyle made using histology and a PDSE MRI sequence. The mean cartilage thickness measured using PDSE images was 0.872mm compared to a mean histological measurement of 0.734mm. 19 Figure 7. Measurements (mm) from the femoral condyle made using histology and an SSFP MRI imaging sequence. The mean cartilage thickness measured using SSFP images was 0.924mm compared to a mean histological measurement of 0.734mm. Pearsons rank correlation coefficient values, that were statistically significant (p<0.05), showed poor to moderate correlation between histology measurements and MRI images for all readers (R values: 0.25 to 0.77) (appendix I). Good correlation (R>0.78 (p<0.05)) between readers was present for 3D SPGR-FS images of medial condyles; between readers 1 and 2 for SSFP images of medial condyles and for all readers for SSFP images of the lateral condyle. There was poor to moderate correlation between readers for all MRI images of the other condyles. Moderate correlation was present between readers 1 and 2 for 3D SPGR-FS images of 20 the lateral condyles; between readers 1 and 3 for PD images of the lateral condyles and between all readers for PD images of the medial condyle. All other R values between readers were not statistically significant. Regression of all three MRI sequence images showed a poor coefficient of variation (R2: 0.40). Regression for individual sequences compared to histology showed poor coefficient of variations for all image types (3D SPGR-FS: 0.11, PDSE: 0.25 and SSFP: 0.35) (appendix II) with SSFP having the highest value. 21 V. DISCUSSION MRI has revolutionized musculoskeletal imaging in human medicine with its ability to differentiate cartilage and periarticular soft tissue structures because of increased contrast resolution compared to other modalities. There continues to be substantial research assessing and improving quantitative evaluation of cartilage as cartilage thickness may prove to be a powerful marker for osteoarthritis (OA).11 Non-invasive evaluation of cartilage thickness has potential in veterinary medicine to objectively assess progression of articular disease as well as response to drug and neutraceutical therapy over time. This modality therefore represents a useful tool in quantitative evaluation of the canine stifle joint. To the author?s knowledge, this is the first study in veterinary medicine that has investigated the efficacy of MRI to quantitatively evaluate cartilage thickness in small animals. A recent study by Olive et al. evaluated the ability of 3D SPGR-FS images for evaluating cartilage thickness and detecting full thickness denuration in the metacarpophalangeal joint in horses1. Olive et al. reported moderate correlation (0.44 p<0.0001) between MR thickness measured with 3D SPGR-FS images (0.90mm ? 0.17mm) and histology (0.79mm ? 0.16mm)1. The results from this study differ from Olive et al. showing poor correlation between 3D SPGR- FS images and histology with R values ranging from -0.26 to 0.41 for all readers. Olive et al reported 3D SPGR-FS images significantly overestimated cartilage thickness <1mm in thickness, which is consistent with the results from this current study. Correlation of MRI measurements for all sequences between readers was variable. The highest significant correlation (p<0.05) between most readers was with SSFP images of the lateral condyle suggesting that this sequence is the most precise of the three sequences for evaluating cartilage thickness. The coefficient of variation between the MRI images and 22 histology measurement was highest for SSFP images (0.30) suggesting this sequence to be the most accurate of the sequences for evaluating cartilage thickness, however; this value is considered low. Study limitations There are several factors in the study design which may have contributed to error between MRI and histological measurements of cartilage thickness: Images were obtained on cadaver specimens. To minimize potential post mortem changes on image quality and variability, MRI images were obtained within 48 hours after euthanasia. A study by Bolen et al. (2010) studied how refrigerating equine cadaver limbs at 4 0C would affect the subjective quality and signal changes of various anatomic structures for magnetic resonance imaging. Bolen et al. reported no subjective change in image quality, and there was no reported changes in cartilage signal. The author believes that the changes from refrigeration in the current study were likely minimal, however; future study design could include image acquisition prior to euthanasia to avoid influence from refrigeration. The parameters used during the acquisition of the MRI images for all sequences contribute to the signal to noise ratio (SNR) and spatial resolution, which are inherent in the MRI sequence properties. The aim of sequence design and planning is to gain maximal spatial resolution with acceptable SNR for image interpretation. Volume averaging associated with voxel dimension of the MRI images is a function of the image field of view (FOV) and matrix size.12 The smallest voxel size achievable is most desirable, but results in decreased SNR of the image. A small image FOV is desirable to minimize the voxel size but this in turn increases the risk of aliasing artifacts creating unacceptable anatomic representation. In this study, FOV and 23 matrix size were standardized for all stifles with the stifle orientated as if a patient was in dorsal recumbency. In a clinical setting FOV and matrix size may have to be varied depending on the size of the leg and individual positioning needs of the patient. SNR of a MRI image is influenced by magnetic field strength by a factor of B1 to B1.5 where B represents the factor of increase in magnetic field strength.12 This means an increase in the field strength of the magnet from 0.5T to 1.5T would result in an increase in SNR from 3-5 times. The field strength of the magnet in this study was 1.5T, a common high field strength magnet at many academic referral veterinary institutions. Higher strength magnets are becoming available for clinical use (e.g. 3T) and research (e.g. 7T). These units potentially will improve SNR and allow smaller voxel sizes to be used; however, T1 relaxation times increase with higher field strength resulting in longer image acquisition times. Longer image acquisition times translate to increased artifact production and increased saturation of short T1 tissues which in turn lead to decreased contrast sensitivity.12 Future studies with higher strength magnets must therefore consider these other factors to optimize image quality and practicality for acquisition with clinical patients. A study by Pepin et al. (2009)22 used a 3D SPGR MRI sequence with a 7T magnet. The total imaging time for one stifle in the coronal plane using a 7T was 15 minutes22 compared with approximately 8 minutes for this study using a 1.5T magnet. There was likely error associated with the decalcification procedure and histological processing of the condyle specimens. To the author?s knowledge, no study has been performed investigating the alteration of cartilage thickness in vivo versus ex vivo and the effect of decalcification prior to processing with a microtome. Cano-S?nchez et al. (2005) describe a procedure for slide preparation of bone samples which does not involve decalcification. They describe a cutting and grinding technique known as the EXAKT system (EXAKT Vetriebs, 24 Norderstedt, Germany). This technique is used to minimize sample shrinkage by not using a decalcification procedure and may be a suitable alternative study design. However, the equipment is expensive, not readily available and utilizes dehydration techniques which may still alter cartilage thickness. The stifle joint is a complex condylar synovial joint with rounded condyles14. Because of the conformation of the condyle, there is error associated with angle variation of the sagittal plane through the femur constructed during the MRI image acquisition and slice angle through the femur during preparation of the histological samples. Therefore, there is potential artificial cartilage thickening on a histological section or MRI image if the plane of cut is <90? from the subchondral bone surface. There is also error from the accuracy of correlating the MRI image and histological slice locations. All three readers differ in their amount of career experience reading musculoskeletal MR images although MRI is a relatively new modality. Reader 1 was a 2nd year radiology resident and readers 2 and 3 were ACVR board certified radiologists. It may have been interesting to include readers with different levels of experience reading MRI images (e.g. radiologists, orthopedic surgeons and residents) and compare the differences between readers. The project did not aim to address variation in cartilage thickness between legs or between patients. The primary concern was to compare the ability of the different MRI sequences in their ability to assess cartilage thickness. Olive et al.1 evaluated the ability of 3D SPGR-FS sequence images at identifying full thickness cartilage denurations with moderate sensitivity (0.56) and high specificity (0.92). However, these lesions were artificially introduced post mortem to the articular cartilage of equine metacarpophalangeal joints. Identifying full thickness denuration in the canine stifle in a similar study may also be worthwhile. Although in vivo cartilage loss 25 results in subchondral bone changes and has been studied in human medicine,11,15 fewer studies have been performed in veterinary medicine.16,17 Identification of full thickness lesions in clinical cases may be more relevant but study design may be difficult. 26 VI. CONCLUSION This study was designed to evaluate the ability of MRI to accurately represent cartilage thickness in the canine stifle and to determine if there was a difference in measurements between the sequences. The majority of MRI cartilage thickness measurements were statistically different from those measured by histology and so objectively the null hypothesis that measurements of cartilage thickness in dogs using MRI are identical to measurements taken using histopathology can be rejected. However, subjectively the MRI sequences investigated were able to clearly represent articular cartilage of the canine stifle and further studies with a larger number of samples and higher field strength magnets may be able to more accurately depict cartilage thickness. As BSSFP sequences had the highest number of images that showed cartilage thickness to be significantly similar to histology measurements, further investigation into its use may be warranted for cartilage imaging in small animals The good correlation between readers for MRI measurement using the 3DSPGR and BSSFP images for the medial and lateral condyles respectively suggest MRI may offer a precise tool for evaluating cartilage thickness and may warrant further investigation with a larger sample size. 27 REFERENCES 1. Olive J, d?Anjou MA, Girard C, et al. Fat-suppressed spoiled gradient-recalled imaging of equine metacarpophalangeal articular cartilage. Vet Radiol Ultrasound 2010; 51:107-115. 2. Ettinger SJ, Feldman EC. Textbook of Veterinary Internal Medicine. 6th Edition 2005. Elsevier Saunders. p. 83 3. Eckstein F, et al. Magnetic resonance imaging (MRI) of articular cartilage in knee osteoarthritis (OA): morphological assessment. Osteoarthritis Cartilage 2006;14:A46-A75. 4. Murray RC, Branch MV, et al. Validation of magnetic resonance imaging for measurement of equine articular cartilage and subchondral bone thickness. Am J Vet Res 2005;66:No11: 1999- 2005 5. McCauley TR, Disler DG. Magnetic resonance imaging of articular cartilage of the knee. J Am Acad Orthop Surg 2001;9:2e8. 6. Disler DG. Fat-suppressed three-dimensional spoiled gradient recalled MR imaging: assessment of articular and physeal hyaline cartilage. AJR Am J Roentgenol 1997;169; 1117e23. 7. Gold GE, et al. MRI of articular cartilage in OA: novel pulse sequences and compositional/functional markers. Osteoarthritis and Cartilage 2006;14,A76eA86. 8. Ytrehus B, Ekman S, et al. Focal changes in blood supply during normal epiphyseal growth are central in the pathogenesis of osteochondrosis in pigs. Bone 2004;35(6): 1294-1306. 9. Olson EJ, Lindgren BR, et al. Effects of long-term estrogen replacement therapy on the prevalence and area of periarticular tibial osteophytes in surgically postmenopausal cynomolgus monkeys. Bone 2007; 41(2): 282-289. 10. Fossum TW, Hedlund CS, Hulse DA, Johnson AL, Seim III HB, Willard MD, Carroll GL. Small animal surgery. 2nd Edition. Mosby. p1024. 2002 28 11. Eckstein F, Worth, W. Quantitative cartilage imaging in knee osteoarthritis. Review Article. Arthritis 2011. Volume 2011, Article ID 475684. Hindawi Publishing Corporation. 19 pages. 12. Bushburg JT, Seibert JA, Leihdoldt EM Jr, Boone JM. The essential physics of medical imaging. 2nd Edition. 2002. Lippincott Williams & Wilkins. pp 439-442. 13. Cano-S?nchez J, Campo-Trapero J, Gonzalo-Lafuente JC, Moreno-L?pez LA, Bascones- Mart?nez A. Undecalcified bone samples: a description of the technique and its utility based on the literature. Med Oral Patol Oral Cir Bucal 2005;10:E74-E87 14. Evans HE. Miller?s Anatomy of the dog. 3rd Edition. 1993. Saunders. p246. 15. Jans LBO. MR imaging findings of lesion involving cartilage and bone in the paediatric knee: a pictorial review. Organe de la Soci?t? royale belge de radiologie. 2011;94:247-253 16. Martig S, Boisclair J, Konar M, et al: MRI characteristics and histology of bone marrow lesions in dogs with experimentally induced osteoarthritis. Vet Radiol Ultrasound 48:105?112, 2007 17. D?Anjou MA, Moreau M, Troncy E, Martel-Pelletier J, Abram F, Raynauld JP, Pelletier. Osteophytosis, subchondral bone sclerosis, joint effusion and soft tissue thickening in canine experimental stifle osteoarthritis: comparison between 1.5T magnetic resonance imaging and computed radiography. Vetsurg 2008;37:166-177 18. Boileau C, Martel-Pelletier J, Abram F, et al. Magnetic resonance imaging can accurately assess the long-term progression of knee structural changes in experimental dog osteoarthritis. Ann Rheum Dis. 2008; Jul;67(7):926-32 19. Pelletier JP, Troncy E, Bertaim T, Thibaud et al. Treatment with tiludronic acid helps reduce the development of experimental osteoarthritis lesions in dogs with anterior cruciate ligament transection followed by reconstructive surgery: a 1-year study with quantitative magnetic resonance imaging. J Rheumatol 2011; Jan;38(1):118-128 20. Olive J, D?Anjou MA, Alexander K, Laverty S, Theoret C. Comparison of magnetic resonance imaging, computed tomography, and radiography for assessment of non-cartilagenous changes in equine metacarpophalangeal osteoarthritis. Vet Radiol Ultrasound 51: 267-279, 2010 29 21. Bolan G, Haye D, Dondelinger R, Busoni V.Magnetic resonance signal changes during time in equine limbs refrigerated at 40C. Vet Radiol Ultrasound 51: 19-24, 2010 22. Pepin SR, Griffith CJ, Wijdicks CA, et al. A comparative analysis of 7.0-tesla magnetic resonance imaging and histology measurements of knee articular cartilage in a canine posterolateral knee injury model. The American Journal of Sports Medicine 37: S1 pp 119S-124S 30 APPENDICES 31 APPENDIX I Pearson?s Correlation Coefficients ------------------------------------ image=3DSFPG condyle=lat ------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 0.86215 0.15958 9.48367 0.63967 1.11000 brown holland 11 0.69739 0.18878 7.67133 0.37467 0.99767 holland hathcock 11 0.75506 0.15848 8.30567 0.59033 1.02600 hathcock histo 11 0.68664 0.18476 7.55303 0.34494 1.11027 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.61478 0.42353 0.18106 brown 0.0441 0.1943 0.5942 holland 0.61478 1.00000 0.29085 0.05462 holland 0.0441 0.3856 0.8733 hathcock 0.42353 0.29085 1.00000 -0.25671 hathcock 0.1943 0.3856 0.4461 histo 0.18106 0.05462 -0.25671 1.00000 histo 0.5942 0.8733 0.4461 32 ------------------------------------ image=3DSFPG condyle=med ------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 1.20718 0.21197 13.27900 0.87367 1.54667 brown holland 11 1.20118 0.25355 13.21300 0.80467 1.62000 holland hathcock 11 0.97030 0.24031 10.67333 0.61300 1.29667 hathcock histo 11 0.78150 0.28835 8.59649 0.42233 1.28036 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.94933 0.81998 0.41438 brown <.0001 0.0020 0.2051 holland 0.94933 1.00000 0.78038 0.37582 holland <.0001 0.0046 0.2547 hathcock 0.81998 0.78038 1.00000 0.26325 hathcock 0.0020 0.0046 0.4341 histo 0.41438 0.37582 0.26325 1.00000 histo 0.2051 0.2547 0.4341 33 -------------------------------------- image=PD condyle=lat --------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 0.78036 0.10679 8.58400 0.60367 0.95400 brown holland 11 0.75976 0.16758 8.35733 0.55600 1.18000 holland hathcock 11 0.60676 0.09887 6.67433 0.45767 0.80133 hathcock histo 11 0.68664 0.18476 7.55303 0.34494 1.11027 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.74928 0.68614 0.32598 brown 0.0079 0.0197 0.3279 holland 0.74928 1.00000 0.57513 -0.02502 holland 0.0079 0.0642 0.9418 hathcock 0.68614 0.57513 1.00000 0.59433 hathcock 0.0197 0.0642 0.0538 histo 0.32598 -0.02502 0.59433 1.00000 histo 0.3279 0.9418 0.0538 34 -------------------------------------- image=PD condyle=med --------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 1.15942 0.28529 12.75367 0.72867 1.56667 brown holland 11 1.13503 0.28110 12.48534 0.81400 1.76667 holland hathcock 11 0.82724 0.28092 9.09967 0.47700 1.29667 hathcock histo 11 0.78150 0.28835 8.59649 0.42233 1.28036 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.74983 0.63039 0.25234 brown 0.0079 0.0376 0.4541 holland 0.74983 1.00000 0.77649 0.54952 holland 0.0079 0.0049 0.0799 hathcock 0.63039 0.77649 1.00000 0.54284 hathcock 0.0376 0.0049 0.0844 histo 0.25234 0.54952 0.54284 1.00000 histo 0.4541 0.0799 0.0844 35 ------------------------------------- image=SSFP condyle=lat -------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 0.90991 0.26843 10.00900 0.64233 1.55000 brown holland 11 0.79976 0.21872 8.79733 0.56067 1.37667 holland hathcock 11 0.71812 0.23625 7.89933 0.49433 1.32667 hathcock histo 11 0.68664 0.18476 7.55303 0.34494 1.11027 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.90742 0.97967 0.64690 brown 0.0001 <.0001 0.0315 holland 0.90742 1.00000 0.93626 0.77078 holland 0.0001 <.0001 0.0055 hathcock 0.97967 0.93626 1.00000 0.68314 hathcock <.0001 <.0001 0.0205 histo 0.64690 0.77078 0.68314 1.00000 histo 0.0315 0.0055 0.0205 36 ------------------------------------- image=SSFP condyle=med -------------------------------------- The CORR Procedure 4 Variables: brown_ holland hathcock histo Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label brown_ 11 1.09042 0.19315 11.99467 0.85033 1.58667 brown holland 11 1.10315 0.18790 12.13467 0.76100 1.51667 holland hathcock 11 0.99852 0.34277 10.98367 0.65567 1.86333 hathcock histo 11 0.78150 0.28835 8.59649 0.42233 1.28036 histo Pearson Correlation Coefficients, N = 11 Prob > |r| under H0: Rho=0 brown_ holland hathcock histo brown_ 1.00000 0.89773 0.35565 0.46196 brown 0.0002 0.2831 0.1526 holland 0.89773 1.00000 0.54722 0.54504 holland 0.0002 0.0815 0.0829 hathcock 0.35565 0.54722 1.00000 0.23767 hathcock 0.2831 0.0815 0.4816 histo 0.46196 0.54504 0.23767 1.00000 histo 0.1526 0.0829 0.4816 37 APPENDIX II Analysis of Variance data and Regression The SAS System 16:20 Monday, March 5, 2012 1 Obs code Condyle Histo PD _DFSSPGR SSFP 1 A1 med med 1.09273 1.46778 1.43778 0.91611 2 A1 lat lat 1.11027 0.77478 0.75167 1.41778 3 A2 med med 1.28036 1.36333 1.45667 1.45889 4 A2 lat lat 0.72737 0.81722 0.72844 0.92167 5 B1 med med 0.47445 1.03156 1.14122 0.97189 6 B1lat lat 0.65208 0.66456 0.64933 0.95389 7 B2 med med 0.42233 0.81133 0.98656 0.98200 8 B2 lat lat 0.77945 0.63722 0.70544 0.60789 9 C1 med med 0.72771 0.86056 1.14533 1.08200 10 C1 lat lat 0.63769 0.97844 0.83033 0.90567 11 C2 med med 1.11507 1.13500 0.96856 1.14522 12 C2 lat lat 0.34494 0.64344 0.66189 0.61956 13 D1 med med 0.53672 0.72600 0.79789 0.75567 14 D1 lat lat 0.70375 0.63978 0.60322 0.66567 15 D2 med med 0.97412 0.69500 0.80667 0.90178 16 D2 lat lat 0.63706 0.71600 0.70933 0.73444 17 E1 med med 0.70214 0.95856 1.13889 1.04100 18 E1 lat lat 0.74723 0.75144 0.99578 0.66211 19 E2 med med 0.60150 1.20111 1.21444 1.04311 20 E2 lat lat 0.68417 0.64667 0.84500 0.72344 21 F1 med med 0.66937 1.05933 1.29444 1.12756 22 F1 lat lat 0.52903 0.59844 1.00644 0.68978 The SAS System 16:20 Monday, March 5, 2012 2 The REG Procedure Model: MODEL1 Dependent Variable: Histo Histo Number of Observations Read 22 Number of Observations Used 22 Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model 3 0.49919 0.16640 4.14 0.0214 Error 18 0.72312 0.04017 Corrected Total 21 1.22231 Root MSE 0.20043 R-Square 0.4084 Dependent Mean 0.73407 Adj R-Sq 0.3098 Coeff Var 27.30443 Parameter Estimates Parameter Standard Variable Label DF Estimate Error t Value Pr > |t| Intercept Intercept 1 0.17329 0.19651 0.88 0.3895 PD PD 1 0.45921 0.34726 1.32 0.2026 _DFSSPGR 3DFSSPGR 1 -0.29328 0.31189 -0.94 0.3595 SSFP SSFP 1 0.47488 0.23480 2.02 0.0582 38 The SAS System 16:20 Monday, March 5, 2012 3 The REG Procedure Model: MODEL1 Dependent Variable: Histo Histo Number of Observations Read 22 Number of Observations Used 22 Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model 1 0.30203 0.30203 6.56 0.0186 Error 20 0.92028 0.04601 Corrected Total 21 1.22231 Root MSE 0.21451 R-Square 0.2471 Dependent Mean 0.73407 Adj R-Sq 0.2095 Coeff Var 29.22194 Parameter Estimates Parameter Standard Variable Label DF Estimate Error t Value Pr > |t| Intercept Intercept 1 0.31243 0.17081 1.83 0.0823 PD PD 1 0.48370 0.18880 2.56 0.0186 The SAS System 16:20 Monday, March 5, 2012 4 The REG Procedure Model: MODEL2 Dependent Variable: Histo Histo Number of Observations Read 22 Number of Observations Used 22 Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model 1 0.12941 0.12941 2.37 0.1395 Error 20 1.09291 0.05465 Corrected Total 21 1.22231 Root MSE 0.23376 R-Square 0.1059 Dependent Mean 0.73407 Adj R-Sq 0.0612 Coeff Var 31.84487 Parameter Estimates Parameter Standard Variable Label DF Estimate Error t Value Pr > |t| Intercept Intercept 1 0.44238 0.19599 2.26 0.0353 _DFSSPGR 3DFSSPGR 1 0.30741 0.19976 1.54 0.1395 The SAS System 16:20 Monday, March 5, 2012 5 39 The REG Procedure Model: MODEL3 Dependent Variable: Histo Histo Number of Observations Read 22 Number of Observations Used 22 Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model 1 0.42843 0.42843 10.79 0.0037 Error 20 0.79388 0.03969 Corrected Total 21 1.22231 Root MSE 0.19923 R-Square 0.3505 Dependent Mean 0.73407 Adj R-Sq 0.3180 Coeff Var 27.14102 Parameter Estimates Parameter Standard Variable Label DF Estimate Error t Value Pr > |t| Intercept Intercept 1 0.17445 0.17556 0.99 0.3323 SSFP SSFP 1 0.60568 0.18436 3.29 0.0037 40 APPENDIX III Histology measurements and Student t test A1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 1248 1.066667 2 1353 1.15641 3 1236 1.05641 4 1356 1.158974 5 1119 0.95641 6 1362.003 1.164105 7 1329 1.135897 8 1368.053 1.169276 9 1218.015 1.041038 10 1368.082 1.169301 Mean 1299.016 1.11027 St Dev 87.66918 0.074931 A1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 1314.003 1.123079 2 1260.004 1.076926 3 1269 1.084615 4 1242 1.061538 5 1229.477 1.050835 6 1284.897 1.098203 7 1317.342 1.125933 8 1327.401 1.134531 9 1291.687 1.104006 10 1281.594 1.095379 Mean 1278.496 1.092732 St Dev 29.9721 0.025617 41 Dr Brown A1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.846 0.875 1.42 2 0.894 0.859 1.55 3 0.858 0.845 1.68 Mean 0.866 0.859666667 1.55 St Dev 0.024979992 0.015011107 0.13 T test 3.8166E-06 8.69454E-07 0.019738206 T test Pooled data 4.57267E-06 6.30856E-07 0.000214791 Dr Brown A1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.32 1.38 0.914 2 1.4 1.59 0.916 3 1.3 1.51 0.924 Mean 1.34 1.493333333 0.918 St Dev 0.052915026 0.105987421 0.005291503 T test 0.010380479 0.021045291 1.42846E-09 T test Pooled data 0.000119871 0.00075266 0.001584916 42 Dr Holland A1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.7 0.594 1.37 2 0.712 0.599 1.4 3 0.709 0.623 1.36 Mean 0.707 0.605333333 1.376666667 St Dev 0.006245 0.015502688 0.02081666 T test 2.09E-08 6.16883E-10 5.66803E-07 Dr Holland A1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.19 1.52 0.973 2 1.16 1.53 0.941 3 1.18 1.52 0.952 Mean 1.176667 1.523333333 0.955333333 St Dev 0.015275 0.005773503 0.016258331 T test 0.000447 1.10085E-13 3.54713E-05 43 Dr Hathcock A1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.812 0.824 1.35 2 0.751 0.815 1.27 3 0.691 0.731 1.36 Mean 0.751333 0.79 1.326667 St Dev 0.060501 0.051293 0.049329 T test 0.001085 0.000469 0.001849 Dr Hathcock Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.23 1.26 0.911 2 1.32 1.36 0.981 3 1.34 1.27 0.733 Mean 1.296667 1.296667 0.875 St Dev 0.058595 0.055076 0.127859 T test 0.021313 0.018038 0.094156 44 A2 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 887.788 0.758793 2 885.664 0.756978 3 972.801 0.831454 4 1000.931 0.855497 5 710.68 0.607419 6 714.146 0.610381 7 867.747 0.741664 8 888.324 0.759251 9 910.113 0.777874 10 825.66 0.705692 Mean 851.0253 0.727372 St Dev 106.7432 0.091234 A2 Medial Measurement Thickness (Pixels) Thickness (mm) 1 1623.136 1.387296 2 1715.465 1.466209 3 1562.842 1.335762 4 1500.075 1.282115 5 1421.851 1.215257 6 1437.704 1.228807 7 1455.003 1.243592 8 1458 1.246154 9 1428.154 1.220644 10 1395.306 1.192569 Mean 1498.024 1.280362 St Dev 103.9515 0.088847 45 Dr Brown A2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.893 0.974 1.09 2 0.92 0.935 1.06 3 0.907 1.04 1.12 Mean 0.906667 0.983 1.09 St Dev 0.013503 0.053075 0.03 T test 9.11E-05 0.001496 5.88E-07 T test Pooled data 9.79E-14 5.47E-13 6.6E-15 Dr Brown A2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.47 1.47 1.6 2 1.56 1.55 1.57 3 1.65 1.62 1.59 Mean 1.56 1.546667 1.586667 St Dev 0.09 0.075056 0.015275 T test 0.014995 0.007541 7.56E-07 T test Pooled data 4.9E-13 0.000832 3.96E-09 46 Dr Holland A2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.908 0.58 0.793 2 0.904 0.592 0.782 3 0.901 0.595 0.804 Mean 0.904333 0.589 0.793 St Dev 0.003512 0.007937 0.011 T test 0.000127 0.00021 0.076104 Dr Holland A2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.42 1.63 1.53 2 1.39 1.6 1.5 3 1.42 1.63 1.52 Mean 1.41 1.62 1.516667 St Dev 0.017321 0.017321 0.015275 T test 0.001273 2.53E-07 8.69E-06 47 Dr Hathcock A2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.653 0.63 0.854 2 0.632 0.664 0.951 3 0.637 0.546 0.841 Mean 0.640667 0.613333 0.882 St Dev 0.01097 0.06074 0.060108 T test 0.003726 0.037857 0.02574 Dr Hathcock A2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.16 1.04 1.31 2 1.18 1.3 1.29 3 1.02 1.27 1.22 Mean 1.12 1.203333 1.273333 St Dev 0.087178 0.142244 0.047258 T test 0.059229 0.444495 0.833701 48 B1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 753.054 0.643636 2 798.141 0.682172 3 693.006 0.592313 4 762.053 0.651327 5 693 0.592308 6 807 0.689744 7 789.006 0.674364 8 792.023 0.676943 9 768.006 0.656415 10 756.006 0.646159 Mean 762.9289 0.652076 St Dev 42.06097 0.03595 B1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 513.035 0.438491 2 447.644 0.382602 3 462.243 0.395079 4 511.269 0.436982 5 577.319 0.493435 6 610.66 0.521932 7 642.764 0.549371 8 546.033 0.466695 9 618.029 0.52823 10 654.11 0.559068 Mean 555.1002 0.474445 St Dev 70.21601 0.060014 49 Dr Brown B1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.945 0.793 1.14 2 0.688 0.829 1.09 3 0.688 0.855 0.937 Mean 0.773667 0.825667 1.055667 St Dev 0.148379 0.031134 0.105765 T test 0.286333 0.00171 0.0193 T test Pooled data 0.008728 0.000526 0.004653 Dr Brown B1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.16 1.35 0.984 2 1.12 1.26 1.12 3 1.19 1.22 1.1 Mean 1.156667 1.276667 1.068 St Dev 0.035119 0.066583 0.07343 T test 1.69E-07 0.000248 0.00117 T test Pooled data 6.95E-08 0.00283 0.000395 50 Dr Holland B1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.64 0.486 0.954 2 0.658 0.517 0.993 3 0.654 0.528 0.973 Mean 0.650667 0.510333 0.973333 St Dev 0.009452 0.021779 0.019502 T test 0.991768 0.000251 5.31E-07 Dr Holland B1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.31 1.32 1.01 2 1.28 1.35 1.03 3 1.3 1.36 1.06 Mean 1.296667 1.343333 1.033333 St Dev 0.015275 0.020817 0.025166 T test 5.92E-13 1.74E-12 1.99E-09 51 Dr Hathcock B1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.663 0.594 0.88 2 0.558 0.672 0.971 3 0.487 0.57 0.647 Mean 0.569333 0.612 0.832667 St Dev 0.088546 0.053329 0.167106 T test 0.250161 0.337233 0.198321 Dr Hathcock B1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.673 0.971 0.915 2 0.695 0.923 0.894 3 0.556 0.517 0.634 Mean 0.641333 0.803667 0.814333 St Dev 0.074715 0.249418 0.156526 T test 0.042664 0.149081 0.058706 52 B2 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 1000.058 0.85475 2 977.029 0.835068 3 994.359 0.849879 4 1031.311 0.881462 5 921.703 0.78778 6 870 0.74359 7 855.047 0.730809 8 864.005 0.738466 9 789.365 0.674671 10 801.679 0.685196 Mean 911.9509 0.779445 St Dev 85.72359 0.073268 B2 Medial Measurement Thickness (Pixels) Thickness (mm) 1 489.009 0.417956 2 522.078 0.446221 3 498.009 0.425649 4 465 0.397436 5 510.035 0.435927 6 468.01 0.400009 7 492.009 0.420521 8 498.036 0.425672 9 510.009 0.435905 10 513.035 0.438491 Mean 494.1231 0.422327 St Dev 20.85184 0.017822 53 Dr Brown B2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.674 0.901 0.72 2 0.682 0.775 0.612 3 0.648 0.832 0.595 Mean 0.668 0.836 0.642333 St Dev 0.017776 0.063095 0.067796 T test 0.001248 0.255246 0.046857 T test Pooled data 0.000329 0.000149 0.00085 Dr Brown B2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.03 1.1 1.08 2 1.03 1.05 1.12 3 1.11 1.09 1.09 Mean 1.056667 1.08 1.096667 St Dev 0.046188 0.026458 0.020817 T test 0.001277 0.000153 3.37E-05 T test Pooled data 0.00249 0.000548 0.000955 54 Dr Holland B2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.675 0.692 0.693 2 0.695 0.68 0.676 3 0.692 0.665 0.692 Mean 0.687333 0.679 0.687 St Dev 0.010786 0.013528 0.009539 T test 0.003739 0.002214 0.003623 Dr Holland B2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.889 1.09 1.11 2 0.913 1.11 1.1 3 0.899 1.1 1.14 Mean 0.900333 1.1 1.116667 St Dev 0.012055 0.01 0.020817 T test 1.91E-07 6.04E-10 3.11E-05 55 Dr Hathcock B2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.563 0.605 0.524 2 0.623 0.667 0.5 3 0.483 0.532 0.459 Mean 0.556333 0.601333 0.494333 St Dev 0.070238 0.067575 0.032868 T test 0.012739 0.021659 9.52E-06 Dr Hathcock B2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.684 0.893 0.828 2 0.504 0.734 0.813 3 0.243 0.712 0.557 Mean 0.477 0.779667 0.732667 St Dev 0.221736 0.098764 0.152317 T test 0.720991 0.02405 0.072139 56 C1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 777.006 0.664108 2 714.006 0.610262 3 759.053 0.648763 4 765 0.653846 5 708.915 0.60591 6 768.586 0.656911 7 726.223 0.620703 8 705.517 0.603006 9 727.789 0.622042 10 766.504 0.655132 Mean 746.0962 0.637689 St Dev 28.48539 0.024346 C1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 828.049 0.707734 2 840.434 0.71832 3 868.328 0.742161 4 888.182 0.75913 5 855.258 0.73099 6 843.021 0.720531 7 936.481 0.800411 8 825.136 0.705244 9 903.603 0.77231 10 819.198 0.700169 Mean 851.423 0.727712 St Dev 27.71323 0.023687 57 Dr Brown C1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.01 0.864 1.12 2 0.926 0.843 1.09 3 0.926 0.77 1.15 Mean 0.954 0.825667 1.12 St Dev 0.048497 0.049339 0.03 T test 0.004865 0.016104 0.000203 T test Pooled data 0.001039 0.001687 0.000366 Dr Brown C1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.892 1.12 1.02 2 1.01 1.19 1.02 3 0.831 1.12 1.02 Mean 0.911 1.143333 1.02 St Dev 0.091 0.040415 0 T test 0.074037 0.000722 5.46E-10 T test Pooled data 2.88E-05 0.000591 0.001722 58 Dr Holland C1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.18 0.89 0.806 2 1.16 0.91 0.835 3 1.2 0.91 0.826 Mean 1.18 0.903333 0.822333 St Dev 0.02 0.011547 0.014844 T test 3.05E-06 8.17E-09 6.21E-06 Dr Holland C1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.02 1.23 1.19 2 0.961 1.28 1.22 3 1.12 1.29 1.19 Mean 1.033667 1.266667 1.2 St Dev 0.080376 0.032146 0.017321 T test 0.019259 6.11E-05 9.96E-09 59 Dr Hathcock C1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.875 0.8 0.856 2 0.859 0.833 0.817 3 0.67 0.653 0.651 Mean 0.801333 0.762 0.774667 St Dev 0.114019 0.095828 0.108859 T test 0.124395 0.144481 0.152891 Dr Hathcock C1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.672 1.07 0.999 2 0.656 1.1 0.92 3 0.583 0.908 0.671 Mean 0.637 1.026 0.863333 St Dev 0.047445 0.103286 0.171185 T test 0.053614 0.036296 0.32533 60 C2 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 461.279 0.394256 2 436.147 0.372775 3 411.962 0.352104 4 376.449 0.321751 5 370.218 0.316426 6 421.807 0.360519 7 360.624 0.308226 8 411.558 0.351759 9 355.269 0.303649 10 369.305 0.315645 Mean 403.5801 0.34494 St Dev 34.23121 0.029257 C2 Medial Measurement Thickness (Pixels) Thickness (mm) 1 1329.487 1.136314 2 1249.042 1.067557 3 1308.88 1.118701 4 1297.53 1.109 5 1377.209 1.177102 6 1338.013 1.143601 7 1347.084 1.151354 8 1299.003 1.110259 9 1245.925 1.064893 10 1263.175 1.079637 Mean 1304.628 1.115066 St Dev 42.7777 0.036562 61 Dr Brown C2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.764 0.706 0.755 2 0.811 0.572 0.675 3 0.696 0.679 0.757 Mean 0.757 0.652333 0.729 St Dev 0.057819 0.070868 0.046776 T test 0.003665 0.012902 0.001934 T test Pooled data 0.002044 0.002843 1.76E-11 Dr Brown C2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.18 1.09 1.29 2 1.31 1.09 1.18 3 1.33 1.02 1.07 Mean 1.273333 1.066667 1.18 St Dev 0.081445 0.040415 0.11 T test 0.070641 0.153367 0.419351 T test Pooled data 0.000922 0.005509 0.002836 62 Dr Holland C2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.721 0.38 0.552 2 0.721 0.376 0.561 3 0.705 0.368 0.569 Mean 0.715667 0.374667 0.560667 St Dev 0.009238 0.00611 0.008505 T test 2.48E-12 0.006446 4.87E-10 Dr Holland C2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.21 1.03 1.2 2 1.21 1.11 1.18 3 1.23 1.11 1.22 Mean 1.216667 1.083333 1.2 St Dev 0.011547 0.046188 0.02 T test 1.48E-05 0.350416 0.001582 63 Dr Hathcock C2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.422 0.756 0.547 2 0.542 1.01 0.642 3 0.409 1.11 0.518 Mean 0.457667 0.958667 0.569 St Dev 0.073323 0.182497 0.064861 T test 0.10131 0.026918 0.01996 Dr Hathcock C2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.88 0.988 1.16 2 1.02 0.764 1.09 3 0.845 0.515 0.917 Mean 0.915 0.755667 1.055667 St Dev 0.092601 0.23661 0.125085 T test 0.057901 0.1177 0.493474 64 D1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 789.051 0.674403 2 738.024 0.63079 3 744.024 0.635918 4 813.089 0.694948 5 786.023 0.671815 6 789.023 0.674379 7 798.141 0.682172 8 900 0.769231 9 892.635 0.762936 10 993.018 0.848733 Mean 823.391 0.703753 St Dev 80.32413 0.068653 D1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 687 0.587179 2 579 0.494872 3 654.007 0.55898 4 648.007 0.553852 5 567.389 0.484948 6 603.067 0.515442 7 552 0.471795 8 564.008 0.482058 9 648.007 0.553852 10 726.025 0.620534 Mean 627.9577 0.536716 St Dev 53.72749 0.045921 65 Dr Brown D1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.617 0.678 0.756 2 0.633 0.63 0.726 3 0.634 0.611 0.703 Mean 0.628 0.639667 0.728333 St Dev 0.009539 0.03453 0.026577 T test 0.006397 0.061882 0.391319 T test Pooled data 2.68E-14 5.5E-06 8.88E-06 Dr Brown D1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.724 0.825 0.921 2 0.792 0.896 0.825 3 0.826 0.9 0.805 Mean 0.780667 0.873667 0.850333 St Dev 0.051936 0.042194 0.062011 T test 0.004146 0.00035 0.00479 T test Pooled data 0.000154 5.72E-08 0.000868 66 Dr Holland D1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.738 0.568 0.718 2 0.739 0.583 0.737 3 0.729 0.588 0.703 Mean 0.735333 0.579667 0.719333 St Dev 0.005508 0.010408 0.017039 T test 0.190745 0.000224 0.545711 Dr Holland D1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.809 0.802 0.751 2 0.822 0.804 0.763 3 0.811 0.808 0.769 Mean 0.814 0.804667 0.761 St Dev 0.007 0.003055 0.009165 T test 8.89E-09 2.7E-08 4.77E-08 67 Dr Hathcock D1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.583 0.556 0.578 2 0.548 0.654 0.541 3 0.537 0.561 0.529 Mean 0.556 0.590333 0.549333 St Dev 0.024021 0.055194 0.025541 T test 0.000165 0.040407 0.000158 Dr Hathcock D1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.56 0.724 0.709 2 0.618 0.711 0.682 3 0.572 0.711 0.576 Mean 0.583333 0.715333 0.655667 St Dev 0.030616 0.007506 0.070302 T test 0.077803 5.13E-07 0.076733 68 D2 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 765.376 0.654168 2 753.597 0.6441 3 726.75 0.621154 4 693.785 0.592979 5 742.026 0.63421 6 738.39 0.631103 7 667.323 0.570362 8 786.092 0.671874 9 726.099 0.620597 10 783.092 0.669309 Mean 745.3597 0.63706 St Dev 27.96201 0.023899 D2 Medial Measurement Thickness (Pixels) Thickness (mm) 1 1144.275 0.978013 2 1091.79 0.933154 3 1105.971 0.945274 4 1132.435 0.967893 5 1059.718 0.905742 6 1101.2 0.941197 7 1212.728 1.03652 8 1210.075 1.034252 9 1217.35 1.04047 10 1233.157 1.05398 Mean 1139.717 0.974117 St Dev 60.17079 0.051428 69 Dr Brown D2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.75 0.689 0.805 2 0.755 0.708 0.804 3 0.761 0.712 0.737 Mean 0.755333 0.703 0.782 St Dev 0.005508 0.012288 0.038974 T test 2.46E-07 0.000195 0.010033 T test Pooled data 5.45E-13 1.17E-18 0.000897 Dr Brown D2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.924 1.13 0.973 2 0.61 0.939 0.913 3 0.652 0.877 0.966 Mean 0.728667 0.982 0.950667 St Dev 0.170462 0.131867 0.032808 T test 0.118257 0.984889 0.244286 T test Pooled data 0.008713 0.008096 0.000194 70 Dr Holland D2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.785 0.793 0.788 2 0.799 0.809 0.803 3 0.787 0.787 0.768 Mean 0.790333 0.796333 0.786333 St Dev 0.007572 0.011372 0.017559 T test 1.73E-08 7.37E-08 2.07E-05 Dr Holland D2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.853 0.811 1.1 2 0.849 0.83 1.11 3 0.869 0.834 1.04 Mean 0.857 0.825 1.083333 St Dev 0.010583 0.012288 0.037859 T test 2.49E-05 3.21E-06 0.016959 71 Dr Hathcock D2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.628 0.637 0.607 2 0.596 0.61 0.694 3 0.583 0.639 0.604 Mean 0.602333 0.628667 0.635 St Dev 0.023159 0.016197 0.051118 T test 0.153054 0.87086 0.907271 Dr Hathcock D2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.529 0.721 0.67 2 0.507 0.638 0.739 3 0.462 0.48 0.605 Mean 0.499333 0.613 0.671333 St Dev 0.034152 0.12243 0.06701 T test 4.32E-06 0.028769 0.006339 72 E1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 882.326 0.754125 2 919.102 0.785557 3 950.091 0.812044 4 942.936 0.805928 5 891.182 0.761694 6 837.048 0.715426 7 861.084 0.735969 8 846.261 0.7233 9 850.038 0.726528 10 786.572 0.672284 Mean 874.2604 0.747231 St Dev 52.34316 0.044738 E1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 846.048 0.723118 2 792.091 0.677001 3 864.047 0.738502 4 825 0.705128 5 801.006 0.684621 6 819.005 0.700004 7 822.005 0.702568 8 804.09 0.687256 9 807.089 0.68982 10 837.65 0.71594 Mean 821.5031 0.702139 St Dev 22.3459 0.019099 73 Dr Brown E1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.896 1.04 0.774 2 0.872 0.984 0.768 3 0.825 1.18 0.768 Mean 0.864333 1.068 0.77 St Dev 0.036116 0.100955 0.003464 T test 0.010246 0.026452 0.169681 T test Pooled data 0.000311 0.003148 8.76E-05 Dr Brown E1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.07 1.19 1.1 2 1.1 1.16 1.07 3 1.07 1.17 1.03 Mean 1.08 1.173333 1.066667 St Dev 0.017321 0.015275 0.035119 T test 1.39E-05 1.17E-06 0.001532 T test Pooled data 0.002398 2.61E-05 0.001236 74 Dr Holland E1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.782 0.884 0.637 2 0.82 0.904 0.649 3 0.83 0.892 0.637 Mean 0.810667 0.893333 0.641 St Dev 0.025325 0.010066 0.006928 T test 0.022052 1.13E-06 1.7E-05 Dr Holland E1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.935 1.12 1.08 2 0.96 1.12 1.1 3 0.936 1.13 1.08 Mean 0.943667 1.123333 1.086667 St Dev 0.014154 0.005774 0.011547 T test 6.94E-06 4.56E-15 1.97E-08 75 Dr Hathcock E1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.658 0.858 0.63 2 0.551 1.06 0.571 3 0.529 1.16 0.525 Mean 0.579333 1.026 0.575333 St Dev 0.06901 0.153844 0.052634 T test 0.038365 0.086043 0.01518 Dr Hathcock E1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.01 1.15 1.07 2 0.889 1.13 0.993 3 0.657 1.08 0.846 Mean 0.852 1.12 0.969667 St Dev 0.179385 0.036056 0.113808 T test 0.285245 0.001227 0.054336 76 E2 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 756.292 0.646403 2 774.145 0.661662 3 810 0.692308 4 831.087 0.710331 5 873.046 0.746193 6 831 0.710256 7 732.006 0.625646 8 747.006 0.638467 9 735.006 0.62821 10 816.199 0.697606 Mean 800.4781 0.684169 St Dev 44.96381 0.038431 E2 Medial Measurement Thickness (Pixels) Thickness (mm) 1 729.099 0.623162 2 681.423 0.582413 3 684.237 0.584818 4 690.059 0.589794 5 693 0.592308 6 717.006 0.612826 7 684.007 0.584621 8 699.058 0.597485 9 720.225 0.615577 10 706.435 0.603791 Mean 703.7548 0.6015 St Dev 16.58956 0.014179 77 Dr Brown E2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.797 0.982 0.869 2 0.698 1.05 0.755 3 0.605 0.91 0.759 Mean 0.7 0.980667 0.794333 St Dev 0.096016 0.07001 0.064694 T test 0.707459 0.0108 0.072317 T test Pooled data 0.00191 0.003851 0.000783 Dr Brown E2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.6 1.21 1.31 2 1.25 1.17 1.1 3 1.17 1.3 0.863 Mean 1.34 1.226667 1.091 St Dev 0.228692 0.066583 0.223636 T test 0.03016 0.003346 0.062235 T test Pooled data 0.00359 8.05E-05 0.006666 78 Dr Holland E2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.707 0.752 0.729 2 0.717 0.73 0.734 3 0.724 0.746 0.705 Mean 0.716 0.742667 0.722667 St Dev 0.008544 0.011372 0.015503 T test 0.015043 0.00079 0.014476 Dr Holland E2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.1 1.21 0.976 2 1.12 1.23 0.979 3 1.12 1.23 1.03 Mean 1.113333 1.223333 0.995 St Dev 0.011547 0.011547 0.030348 T test 2.47E-07 1.11E-07 0.001065 79 Dr Hathcock E2 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.551 0.789 0.636 2 0.521 0.97 0.727 3 0.5 0.676 0.597 Mean 0.524 0.811667 0.653333 St Dev 0.025632 0.148305 0.066711 T test 0.000361 0.251586 0.627801 Dr Hathcock E2 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.24 1.25 1.17 2 1.18 1.18 1.11 3 1.03 1.15 0.85 Mean 1.15 1.193333 1.043333 St Dev 0.108167 0.051316 0.170098 T test 0.012243 0.002024 0.045203 80 F1 Lateral Measurement Thickness (Pixels) Thickness (mm) 1 594.348 0.50799 2 626.26 0.535265 3 617.111 0.527445 4 606.898 0.518716 5 615.731 0.526266 6 621.007 0.530775 7 627.029 0.535922 8 654.028 0.558998 9 654.11 0.559068 10 579.117 0.494972 Mean 618.9617 0.529027 St Dev 23.2845 0.019901 F1 Medial Measurement Thickness (Pixels) Thickness (mm) 1 609.738 0.521144 2 792.023 0.676943 3 819.005 0.700004 4 777.006 0.664108 5 783.006 0.669236 6 780.092 0.666745 7 831.049 0.710298 8 807.557 0.69022 9 838.552 0.716711 10 844.541 0.72183 Mean 783.1612 0.669369 St Dev 65.76595 0.05621 81 Dr Brown F1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.592 1.09 0.668 2 0.587 1.07 0.821 3 0.632 1.17 0.753 Mean 0.603667 1.11 0.747333 St Dev 0.024664 0.052915 0.076657 T test 0.019887 0.001943 0.036563 T test Pooled data 1.23E-05 1.24E-08 0.000824 Dr Brown F1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.12 1.44 1.14 2 1.12 1.44 1.17 3 1.11 1.37 1.19 Mean 1.116667 1.416667 1.166667 St Dev 0.005774 0.040415 0.025166 T test 7.51E-10 2.87E-06 1.11E-08 T test Pooled data 2.08E-05 0.000145 1.04E-05 82 Dr Holland F1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.648 0.985 0.705 2 0.644 0.998 0.723 3 0.626 1.01 0.717 Mean 0.639333333 0.99766667 0.715 St Dev 0.011718931 0.01250333 0.009165151 T test 2.15771E-05 1.6813E-08 1.32599E-08 Dr Holland F1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 1.14 1.29 1.2 2 1.14 1.32 1.18 3 1.12 1.29 1.18 Mean 1.133333333 1.3 1.186666667 St Dev 0.011547005 0.01732051 0.011547005 T test 1.36278E-10 8.7452E-12 4.28427E-11 83 Dr Hathcock F1 Lateral Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.525 0.925 0.629 2 0.582 0.967 0.619 3 0.55 0.843 0.573 Mean 0.552333 0.911667 0.607 St Dev 0.028572 0.063066 0.029866 T test 0.299147 0.007513 0.033011 Dr Hathcock F1 Medial Measurement PD (mm) 3DFSPG (mm) SSFP (mm) 1 0.88 1.16 1.12 2 1 1.23 1 3 0.904 1.11 0.968 Mean 0.928 1.166667 1.029333 St Dev 0.063498 0.060277 0.080133 T test 0.00786 0.000796 0.008345 84