Browsing by Subject "Tomography, X-Ray Computed"

Now showing 1 - 16 of 16

4D Study of Thoracic Cancer Radiation Treatment
(2007-05-22) Huang, Tzung-Chi; Zhang, Geoffrey
Respiratory motion causes an added uncertainty in the radiation treatment of thoracic malignancies due to an increase in the normal tissue irradiated and an uncertainty in the radiation coverage of the tumor. This results in a potential increase in complications from treatment and may be insufficient to ensure coverage of the tumor. Reduction of the volume of normal tissue irradiation while maintaining tumor coverage is used to accomplish this goal. The application of 4D CT imaging to radiotherapy treatment planning is an active area of research with the goal to reduce the required normal tissue irradiation and improve the tumor coverage. Deformable image registration holds the key to link the information of two images at various phases. The major purpose of this study is to develop and validate the optical flow method (OFM), a method of deformable image registration by which the image content properties are utilized to generate a displacement vector between each voxel in the reference image to the target image for registration. With OFM, we were able to develop and validate an automated method for intrathoracic motion estimation from breath-hold and 4-D computed tomography imaging; demonstrate the path integration of a four-dimensional dose distribution onto the 3-D anatomy; develop an automated target delineation technique; and to develop and implement a method to quantify tumor response and normal tissue damage by comparison of pre- and post-treatment and 18F-FDG-PET scans, all of which constitute meaningful applications and represent substantial progress in radiation treatment.
Addressing Label Variability in Deep Learning-Based Segmentation for Radiation Therapy
(December 2021) Balagopal, Anjali; Wang, Jing; Jiang, Steve B.; Nguyen, Dan; Lin, Mu-Han; Fei, Baowei
Accurate segmentation of tumor and surrounding organs-at-risk (OARs) is important for radiotherapy treatment(RT) planning. Manual segmentation by physicians currently used in clinical practice is time consuming and highly depends on the physicians' skill and experience, leading to large inter-and intra-observer variation. Deep learning(DL) algorithms, such as those used for image recognition, has been promising in the development of automated segmentation tools for medical imaging. Contouring in RT has some specific challenges as opposed to semantic segmentation in other spaces. DL segmentation models are trained with large, annotated datasets. The annotations play such an important role since these labels are the supervisory signals teaching the model how to segment. Majority of the times, a very high-quality dataset is unavailable. The datasets available are usually biased, noisy, and sometimes scarcely annotated. One of the primary sources of noise in RT datasets is annotation variations. This dissertation addresses the challenge of label variability and investigates different methods to deal with various kinds of label variability. The thesis studies the impact that label variability has on automation performance as well as patient outcome and devices ways to reduce or to respect this variability. Computed Tomography based segmentation model for intact and post-operative prostate cancer RT planning are proposed and developed for clinical use. Uncertainty in the automated label prediction for CTV and OARs are evaluated in detail. Impact of inter-physician variability on patient outcome for post-operative prostate cancer is investigated and instead of segmenting a general CTV , PSA-Net is proposed that respects style variations across physicians as well as across institutions. For dealing with systematic label variability in some structures, a prior-guided deep difference meta-learner is proposed that can segment a structure in a new labelling style from just a handful of prior segmented patients. A multi-modality IPA segmentation model is proposed to reduce label variability due to expertise differences among physicians in clinical trials. This model can effectively help inexperienced physicians in producing expert segmentations. The contributions of this thesis are expected to facilitate better understanding of label variability in RT and help in avoiding/respecting label variability when developing deep learning models for RT.
Automated Quantification of Image-Derived Phenotypes and Integration with the Electronic Health Record at Scale in an Academic Biobank
(2020-05-01T05:00:00.000Z) MacLean, Matthew Timothy; Hill, Joseph A.; de Lemos, James; Rader, Daniel
BACKGROUND: Radiographic images obtained during clinical care can yield a tremendous number of quantitative traits that facilitate translational research. Both liver fat and abdominal adipose mass are examples of quantitative traits that are highly relevant to human health and disease and can be quantitated from medical images such as CT scans. The Penn Medicine Biobank has generated genomic and biomarker data from >50,000 participants with consent to access EHR data. Among these patients, there are greater than 160,000 CT scans, representing over 19,000 patients from which these quantitative traits could be derived. However, manual review of imaging data is time-consuming, costly, and can produce variable results. OBJECTIVE: The goal of this research was to develop a fully automated pipeline to quantify hepatic fat and abdominal adipose mass from CT scans that could be run at scale on patients within the Penn Medicine Biobank. METHODS: We developed a fully automated image filtering and analysis pipeline to analyze CT imaging data. Deep learning networks were trained to identify the presence of IV contrast, delineate the borders of the liver and spleen, and detect visceral and subcutaneous fat. To identify CT studies, we queried our biobank of 52,441 patients for all non-contrast chest, abdomen, and all abdomen/pelvis scans and identified 161,748 CT scans from 19,624 patients. All scans were processed in our deep-learning pipeline in less than 96 hours using parallel processing with cloud-computing resources. From the imaging data, we extracted 12 different image-derived phenotypes that were used in association studies with the electronic health record (N>13000). We also performed genetic association studies (N>5000) on our CT-derived measure of liver fat (LF). Liver fat was defined as the difference in attenuation between the spleen and the liver. Receiver operator characteristic analysis of the liver fat metric was conducted by utilizing 135 patients who had both a CT scan as well as a liver biopsy. Finally, we performed principal component analysis to explore the interrelatedness of the image derived metrics in the context of the phenome. RESULTS: Each component of the algorithm was individually validated for accuracy. The first deep learning network (CNN1) functioned to identify IV contrast, and on a testing set of 400 (half with IV contrast), CNN1 classified 399 scans correctly. CNN2 was tasked with identifying the superior and inferior borders of the abdominal cavity and on a testing set of 100 scans was on average within one slice of the superior (1.01±1.11) and inferior (0.70±0.64) borders. Performance of CNN3, which was tasked with labeling liver, spleen, visceral and subcutaneous fat was assessed by computing region-of-interest area overlap ratios (Dice coefficients). Dice coefficients for a validation set of randomly selected CT scans showed high values for liver (0.95±0.02, n=20), spleen (0.92±0.07, n=20), abdominal compartment (0.98±0.01, n=10), subcutaneous fat (1.00±0.00, n=10), and visceral fat (0.99±0.01, n=10). After extracting the liver fat (LF) metric for all patients, LF had a mean of -6.4 ± 9.1 Hounsfield units. Association studies with billing codes in the electronic health record yielded many known associations for LF including with chronic liver disease, diabetes mellitus, obesity, and hypertension. A genetic association study showed significant associations with variants located in PNPLA3 and TM6SF2. ROC analysis using pathology data yielded an AUC value of 0.81 with a balanced cutoff value of -6 HU for LF. CONCLUSION: This paper presents a fully automated AI-based algorithm for the quantification of traits from medical imaging. This high-throughput algorithm was applied to over 160,000 scans to quantify 12 different traits and the results were integrated with phenome, genome, and pathology data in the Penn Medicine Biobank. This process is scalable, fast, and fault-tolerant and can power translational studies when integrated with clinical and genetic data.
Density Analysis of Spontaneous Lower Extremity Fractures Using Computed Tomography: A Case-Control Study
(2020-05-01T05:00:00.000Z) Narayanan, Anish; Chhabra, Avneesh; Chalian, Majid; Pezeshk, Parham
BACKGROUND: Spontaneous fractures are predominantly caused by osteoporosis and have significant morbidity and mortality associated with them. The current gold standard of clinical imaging for these osteopenic patients, Dual Energy X-Ray Absorptiometry (DXA), has a number of inherent deficiencies, including limited scanner availability, inaccuracies of projective areal density measurements, and lag in visualized radiographic change when following disease and treatment progress of osteopenic bone. Computed Tomography (CT) imaging has the potential to address these limitations, but the existing literature which discusses its potential use in this situation is limited in scope. OBJECTIVE: It is hypothesized that the patients with spontaneous fractures will exhibit reduced bone mineral density (BMD) as quantified by Hounsfield (HU) measurements in the trabecular bone on CT imaging when compared to appropriately matched controls. METHODS: A retrospective consecutive series of 522 adult patients with admission for fracture were initially obtained from the hospital electronic medical record (EMR). A number of chart review exclusion criteria were then applied, including traumatic history, evidence of malignancy, known renal disease or other secondary pathology that could be ascribed as the etiology of the bone insufficiency, or surgical hardware placement. These patients with CT imaging of the fracture site were then subdivided by anatomic location to select for femoral fractures and were then age and sex matched to appropriate control patients who had 5underwent KUB (kidney, ureters, and bladder) non-contrast CT scans with clinical indication for nephrolithiasis. Following obtaining the case and control CT studies, elliptical fixed region 3 cm² measurements in the trabecular bone were obtained without cortical sampling at three locations (the site of the fracture, proximally at the femoral head, and distally at the lesser trochanter) on both the fractured and contralateral side in both fracture cases and controls. Inter- and intra-patient comparisons were performed, including Chi-square and t-test analyses. RESULTS: A total of 24 spontaneous fractures and 25 controls were analyzed in this study. No significant differences in all captured demographic parameters, including mean age, gender, height, weight, and body mass index (BMI), were observed. There were statistically significant differences in the recorded BMD between the fracture and contralateral non-fracture sides at (p = 0.0001) and distal (p < 0.0001) to the fracture, with elevation of the trabecular bone density at the fracture site's ROI. Proximal and distal bone density differences existed between case fracture and control non-fracture sites (p < 0.0001, p = 0.0001), and between the case non-fracture and control non-fracture sites (p < 0.0001, p < 0.0001). The reliability for measurements was good to excellent proximal to the fracture site (ICC = 0.63-0.87), moderate to excellent at the fracture site (ICC = 0.43-0.78), and fair to good distal to the fracture site (ICC = 0.24-0.68). Additionally, at the proximal site, the odds ratio of every 50 unit decrease in HU is 1.744 (95% CI: 1.291 to 2.356). CONCLUSION: Patients with spontaneous femoral fractures exhibit reduced BMD when compared to 6asymptomatic controls that can be distinguished on CT imaging per reduced HU density in the trabecular bone. Bone insufficiency is best demonstrated proximal or distal to, rather than at, the fracture site as the site of fracture demonstrates trabecular bone compression and hemorrhage that artificially elevates the BMD and obscures any present osteopenia. Opportunistic use of pre-existing CT imaging could therefore be invaluable in identifying a patient's true osteopenic status, saving patients an additional DXA scan while providing accurate, three-dimensional information regarding the true material density of a patient's skeleton.
Development of Deep Learning Artificial Intelligence to Detect Osteoporosis
(2024-01-30) Fan, Christopher; Scanio, Angelo; Joshi, Parag; Öz, Orhan K.; Peshock, Ronald M.; Kay, Fernando
Osteoporosis poses a substantial social and economic burden, with estimated treatment costs reaching a combined six trillion USD in the USA, Canada, and Europe. Although dual-energy X-ray absorptiometry (DEXA) is the diagnostic gold standard, computed tomography (CT) scans have proven to be reliable proxies for bone density measurement. Opportunistic screening for low bone density using CT obtained for other purposes can potentially reduce complications from osteoporotic fractures and health care costs. In this study, we developed an artificial intelligence (AI) algorithm using neural networks and the MONAI library to estimate DEXA bone density from non-contrast cardiac CT obtained for coronary calcium scoring purposes. A total of 2797 Dallas Heart Study phase 2 participants (39% male, 61% female) were included. The AI algorithm was first developed to automatically segment trabecular bone from cortical bone. This was trained and validated with manual segmentation of the trabecular bone by two medical students, a radiologist, using MONAI 3D autoseg. The ML algorithm achieved a Dice score of 0.97 when compared to human segmentation. A second AI model was developed utilizing segmentations of the first model. This AI was trained utilizing corresponding DEXA bone mineral density (BMD) for thoracic vertebrae. The best performing model was trained for 102 epochs, resulting in a training root mean square error (RMSE) of 0.0628 mg/cm2 and validation RMSE of 0.0842 mg/cm2. The final AI algorithm predictions yielded an R2 value of 0.71 compared to DEXA (Figure 1). Our findings underscore the clinical feasibility for an automated neural network to predict DEXA scores from non-contrast cardiac CT. This approach may help in the early detection of unsuspected low bone mineral density in patients undergoing CT scans for other reasons, allowing for potential improvements in patient outcomes and resourceful utilization of diagnostic imaging.
Diagnostic Yield of Cervical Radiographs in Infants with Deformational Plagiocephaly
(2014-04-11) Cho, Min-Jeong; Kane, Alex; Derderian, Christopher; Smartt, James M.
BACKGROUND: Deformational plagiocephaly is a common condition of infancy in which the child presents with craniofacial asymmetry thought to be due to prenatal/postnatal external molding forces on the cranium. Etiologies may include muscular torticollis, intrauterine constraint (i.e. small maternal pelvis, multiple gestation, or breech position), and supine sleep position. The current standard of care for deformational plagiocephaly is active repositioning or orthotic helmet, and these treatments rely on cranial malleability. Patients with deformational plagiocephaly are frequently referred for cervical radiographs in order to determine whether there is osseous abnormality associated with torticollis. Some clinicians believe that screening cervical radiographs are important to rule out bony abnormalities in this condition, while others believe this is unnecessary due to low prevalence of accompanying osseous deformity. OBJECTIVE: The purpose of this study was to determine the diagnostic yield of cervical radiographs in demonstrating cervical anomalies in a population of infants referred to a tertiary craniofacial center with deformational plagiocephaly (DP). METHODS: After obtaining IRB approval, all patients with diagnosis of DP who underwent cervical radiographs between the years of 2010 to 2012 were reviewed. Cervical radiographic findings as determined by radiologist report, perinatal data, and physical exam findings were recorded, and descriptive statistics were generated. RESULTS: Electronic medical records of 730 patients with diagnosis of DP were reviewed. Abnormal findings were recorded in 6.71% of cervical radiograph reports (n=49/730). Of those with abnormal findings, 59% (n=29/49) demonstrated osseous abnormalities including: clavicle abnormality (n=4), bony fusion (n=10), cervical ribs (n=4), joint instabilities (n=5), and vertebral abnormalities (n=11). Those with non-osseous abnormalities (n=26/49) included head tilt (n=4), abnormal curvature (n=16), and soft tissue prominence (n=6). The other 96% of the study population were without osseous abnormalities. CONCLUSION: There is a fairly low diagnostic yield in ordering cervical radiographs in patients with deformational plagiocephaly. Considering the radiation exposure and cost associated with the practice of ordering routine cervical radiographs in all patients presenting with this DP, an inspection of its inclusion as a necessary step in the diagnostic algorithm is warranted.
Donald W. Seldin, M.D., Research Symposium finalist presentations
(2022-04-29) Almonte, Matthew; Duvalyan, Angela; McAdams, Meredith; Onyirioha, Kristeen; Saez-Calveras, Nil; Triana, Taylor
This edition of the UT Southwestern Internal Medicine Grand Rounds features presentations by the six Foster Fellows selected as finalists from the Seventh Annual Donald W. Seldin, M.D. Research Symposium, which was held on April 21, 2022. These Foster Fellows presented work that spanned the breadth and depth of scholarly activity across the department, and at the close of Grand Rounds, one will be selected as the 2022 Seldin Scholar, in honor of Dr. Donald W. Seldin. The Grand Rounds presentation includes additional award presentations recognizing Clinical Vignettes, as well as the Award for Research in Quality and Education at Parkland Hospital and the Social Impact Award.
HCC Surveillance Is Associated with Potential Harms
(2015-01-26) Muffler, Adam; Atiq, Omair; Yopp, Adam; Singal, Amit G.
BACKGROUND: Hepatocellular carcinoma (HCC) is the 3rd leading cause of cancer death worldwide and leading cause of death in patients with cirrhosis. HCC surveillance is recommended in patients with cirrhosis to improve early detection rates. A comprehensive assessment of HCC surveillance should weigh both benefits and harms; however, no study to date has assessed potential harms. Although ultrasound and alpha fetoprotein (AFP) have minimal direct harms, there are potential downstream harms from follow-up tests that should be considered. Objective: To quantify and characterize potential harms of HCC surveillance among a large cohort of patients with cirrhosis METHODS: We conducted a retrospective cohort study among patients with cirrhosis followed at a large safety-net health system. We recorded all surveillance abdominal imaging and/or alpha fetoprotein (AFP) testing between January 2010 and December 2013. We defined a false positive surveillance test as a suspicious liver mass on ultrasound or AFP >20 ng/mL, without HCC development during follow-up. We recorded CT or MRI scans, biopsies, or any procedures performed as a direct result of surveillance testing. Predictors of harm were identified using logistic regression, with significance being defined as p<0.05. RESULTS: We identified 571 patients with cirrhosis, with median follow-up of 2.8 years. HCC surveillance was performed at least once in 551 (96%) patients. Surveillance testing led to diagnostic CT or MRI testing in 123 (21.5%) patients - 74 with one CT/MRI and 49 with multiple studies. Rates of unnecessary diagnostic testing increased from 15% if followed for ≤1 year to 25% if followed for 1-2 years to 37% if followed for ≥2 years. An additional two patients had a biopsy and one patient angiogram for false positive surveillance tests. Follow-up tests were performed due to false positive ultrasound in 47 cases, false positive AFP in 35 cases, and indeterminate ultrasound results in 41 cases. In multivariate analysis, surveillance harm was associated with viral liver disease (OR 1.60, 95%CI 1.04-2.46), receipt of hepatology subspecialty care (OR 2.32, 95%CI 1.52-3.59), and coverage by Parkland Health Plus (OR 2.21, 95%CI 1.45- 3.40). CONCLUSION: This study is the first to demonstrate HCC surveillance can be associated with potential harms. One in five patients have at least one unnecessary diagnostic test, and nearly 10% have multiple tests. Better HCC surveillance tools, with a higher positive predictive value, are urgently needed.
I'm sorry to bother you at a party like this, but should I get one of those scans?”: evaluating the coronary arteries using CT and MRI in 2006 and beyond
(2006-02-02) Peshock, Ronald M.
Masking Enhances Accuracy of Bladder Deformation in Multi-Fraction Adaptive Brachytherapy as a First Step Toward Composite Dose Estimation
(2014-02-04) Barclay, Jennifer; Albuquerque, Kevin; Pompos, Arnold; Gu, Xuejun
BACKGROUND: GEC-ESTRO guidelines for cervix HDR brachytherapy advocate measurement of the minimum dose to the 2cc volume of organs at risk (OAR) receiving the highest amount of radiation and summation across multiple treatment fractions to give a worst-case-scenario cumulative dose estimate. If the OAR from different fractions could be accurately co-registered using deformation, then a more accurate composite dose could be obtained. OBJECTIVE: As a first step toward composite dose estimation, we sought to assess and improve the quality of bladder deformation using a technique called masking, which involves resetting the pixel values within a contour. METHODS: CT scans from nine cervical cancer patients with bladders contoured by radiation oncologists were obtained, and the urethra near the bladder was contoured using the catheter as a fixed reference point. Three copies of each CT were made: the first was unaltered, the second had the bladder masked, and the third had the bladder masked at one pixel value and the rest of the body masked at a different pixel value. Using VelocityAI 2.8.1, the bladder was deformed onto the target (Fraction 1) planning CT from subsequent planning CTs in an attempt to match up the tissue from different fractions. To assess the accuracy of the deformation, several indices were used: the percent error of the deformed bladder volume from the expected volume, the conformality index, the Hausdorff distance, and the distance between the centers of the deformed urethra and the target urethra. RESULTS: Deformation quality improved with masking. The standard deviation of the percent error was reduced from 18.1% with no masking to 4.3% with masking. Mean conformality increased from 0.83 with no masking, to 0.91 with the bladder masked, to 0.93 with body and bladder masked (P<0.001). The mean Hausdorff distance decreased from 13.8mm without masking, to 9.1mm with the bladder masked, to 5.7mm with body and bladder masked (P<0.001). The mean error in the urethra deformation increased from 4.3mm without masking, to 5.2mm with the bladder masked, to 6.6mm with body and bladder masked (P=0.08). CONCLUSION: The accuracy of bladder deformation can be significantly improved by masking. With masking, the volume and location of the deformed bladder more closely approached that of the target bladder than without masking. Thus masking has the potential to improve the accuracy of dose deformation and composite dose calculation in adaptive brachytherapy.
Mechanical Signals for Compensatory Lung Growth Assessed by High Resolution Computed Tomography
(2008-09-18) Ravikumar, Priya; Hsia, Connie C.W.
This dissertation involves the use of high resolution computed tomography (HRCT) to understand the role of intra-thoracic mechanical force and its distribution in regenerative growth in dogs i.e. to quantify lobar lung volumes and density gradients in normal and post-pneumonectomy (following lung resection) lungs. HRCT was used to quantitatively assess regional distribution of lung volume and density gradients among lobes of the lung in order to follow the expansion of remaining lobes following lung resection with a high degree of anatomical precision, and to determine the relationships between lung expansion and alveolar tissue growth. I also extended this work by relating regional lung expansion and growth assessed by radiology to regional alveolar tissue growth assessed by detailed quantitative histology under light and electron microscopy. This study illustrates for the first time a powerful and novel use of in vivo imaging to quantify regional lung distortion and changes in local volume, lung compliance as well as soft tissue density. These changes can be followed non-invasively and serially in a wide range of clinical and investigational applications, such as a) assessing the extent and progression of regional heterogeneity in lung disease or injury; b) assessing local response to treatment or surgical intervention; or c) assessing normal or abnormal patterns of lung growth.
Physician Estimated Depth: Can It Reduce Unnecessary CT Scan Evaluation of Pectus Excavatum?
(2019-01-22) Postma, Heather Elizabeth; Mokdad, Ali; Fike, Frankie; Alder, Adam; Hicks, Barry; Schindel, David; Qureshi, Faisal; Burkhalter, Lorrie; Pandya, Samir
BACKGROUND: The severity of pectus excavatum is determined by computed tomography (CT) derived indices such as the Haller Index (HI) and the Correction Index (CI). Physician estimated depth (PED) as an alternative measurement may preclude for CT. We retrospectively evaluated PED as a screening tool to identify surgical candidates. METHODS: Patients with a diagnosis of pectus excavatum between 1/1/2009 and 3/30/2018 were extracted from the electronic health record for review. Patients without available imaging were excluded. HI and CI were calculated from CT images. CT derived measurements acted as an approximation of PED. Using ROC analysis, we estimated the optimal PED cut-off for identifying surgical candidates according to an HI ≥ 3.25. RESULTS: A total of 94 patients were identified and all met inclusion criteria. Patients were predominantly males (82%) with a median age of 15 (IQR=14-16). Almost half (46%) were underweight (BMI < 18.5). The median HI was 4.1 (IQR=3.7-5.1) with 89% ≥ 3.25. The median CI was 39 (IQR=29-47). CI was highly correlated with HI (r=0.77, area under the curve, AUC=0.88). A CI of 27 was best at correctly classifying HI above/below 3.25 (sensitivity, Se=89%, specificity, Sp=90%). Median PED was 2.5 (IQR=2.1-3.0). Overall, a PED of 2cm correctly classified 86% of HI above/below 3.25 (Se=88%, Sp=70%, AUC=0.84). Among underweight patients, ROC parameters improved (93% correctly classified, Se=95%, Sp=75%, AUC=0.92). PED was highly correlated with CI (r=0.77, AUC=0.94). Among underweight patients, a PED of 2cm correctly classified 95% of CI above/below 27 (Se=97%, Sp=80%, AUC=95%). CONCLUSION: PED over 2cm can accurately identify patients who require CT imaging and pectus correction. Our findings show that PED may be employed easily in the clinic as a screening tool, thereby minimizing unnecessary CT scans. A prospective evaluation of PED is underway at our center.
Quantifying Differences in Femoral Head and Neck Asphericity in the Common Hip Conditions of Femoroacetabular Impingement (FAI) and Hip Dysplasia Versus Controls Using Radial 3DCT Imaging and Volumetric Segmentation
(2018-01-23) Schauwecker, Natalie; Xi, Yin; Slepicka, Chenelle; Dessouky, Riham; Fey, Nicholas; Chatzinoff, Yonatan; Chopra, Rajiv; Wells, Joel; Chhabra, Avneesh
AIM: 3DCT analysis of femoral head and bump anatomy in quantifying pathology in common hip conditions of FAI and hip dysplasia versus controls. Material and Methods: Consecutive patients who obtained 3DCT imaging for hip dysplasia or FAI were compared to asymptomatic controls. Alpha angles on radial CT and 3D volumetric femoral head and bump segmentations were obtained by two readers. Inter- and intra-patient comparisons were performed including inter-reader and ROC analyses. RESULTS: 25 FAI patients, 16 hip dysplasia patients and 38 controls were analyzed. FAI and dysplasia patients exhibited higher alpha angles and higher bump-head volume ratios than the controls (p<0.05). FAI patients exhibited larger bumps than dysplasia and the contralateral hips of FAI were also different than the controls. Alpha angle at 2 oﾒclock and bump to head ratio showed the highest area under the curve for cases versus controls. The reader reliability was better for volumetric segmentation (ICC= 0.35-0.84) as compared to the alpha angles (ICC= 0.11-0.44). CONCLUSION: Patients with both FAI and dysplasia exhibit different femoral head anatomy than the asymptomatic controls. Volumetric segmentation of femoral head and bump is more reliable and better demonstrates the bilateral femoral head anatomy differences of cases versus controls.
The Radiation Footprint on the Pediatric Trauma Patient
(2013-01-22) Farzal, Zainab; Fischer, Anne C.; Wilson, Sarah; Brown, Erica; Burkhalter, Lorrie
BACKGROUND: Overuse of radiation in patients has recently become an important topic of discussion within the medical community. Pediatric patients are most at risk. A recent study has proven the incidence of 1 brain tumor for every 10,000 head computed tomography (CT) scans among patients under the age of 10 in the ten years following a single scan1. Although there have been attempts to limit radiation dosage, there is no published data on the quantity of imaging done on the pediatric trauma patient. The high index of suspicion in trauma has created a paradigm of comprehensive imaging independent of symptoms. Our hypothesis was to identify the current amount of radiation used in a pediatric trauma patient for one visit and correlate the levels of imaging with trauma activation status in order to identify the subset of pediatric trauma patients most at risk for over-radiation. METHODS: This IRB-approved retrospective review of pediatric trauma patients at an ACS verified Level 1 independent children's hospital reviewed three levels of trauma activation (Stats, Alerts, Consults) from June 2010 to January 2011. Charts were analyzed for demographics, mechanism of injury, injury severity score (ISS), imaging modalities, and radiation dosages. Our study included the total number of Stats with representative cohorts from Alerts and Consults. 215 patients (N=1050) met inclusion criteria with complete dosimetry data available. RESULTS: The demographics include: gender (143M, 72F); age range <1-16 years (median 5.5), and activation status with average ISS score (Consults, 7.7 ± 9; Alerts, 8.8 ± 7; and Stats, 17 ± 14, respectively). Non-accidental trauma (NAT) and Stat activations exceeded all others in radiation exposure. Per Stat the number of CTs ranged from 0-10 with 2-3 CTs in 35% and 4-10 CTs in 40% for one admission. The studies most often repeated were head CTs (45%), face/sinus CTs (13.8%) and neck CTs (10%). The majority (66%) of outside CTs delivered more radiation, of which 50.0% of the doses were at least double the dosage delivered at the children's hospital. CONCLUSION: This study is the first to correlate the amount of radiation exposure with trauma activation status. Most of the repetitive imaging was utilized in Stat activations and NATs and used 2-3 times as many CT scans. The identified factors associated with the most radiation include suspected NATs, Stat activations, and outside imaging. To minimize the radiation footprint, we may need to change the current practice of imaging to identify all possible injuries regardless of symptoms.
Readmission and Imaging Outcomes in Pediatric Complicated Appendicitis: A Matched Case-Control Study
(2017-01-17) Murphy, Kristen; Foglia, Robert; Wolf, Steven; Alder, Adam
INTRODUCTION: Currently, the treatment guidelines for perforated appendicitis generally include primary appendectomy or non-operative management followed by an interval appendectomy 6 to 12 weeks post discharge (first-line antibiotics). First-line antibiotics along with abscess drainage and deferred appendectomy is selected with the intent to minimize complications of surgical management. However, investigation of specific, clinically-relevant outcomes identified that primary appendectomy reduced time away from normal activities and was associated with higher family satisfaction, fewer CT scans, and fewer visits to the emergency department. The benefits of each continue to be debated. The aim of this study was to compare clinically-relevant outcomes such as length of stay, imaging rate and readmission between patients selected for first-line antibiotics and first-line appendectomy using a matched case-control approach. METHODS: The electronic medical record system at Children's Medical Center was queried for all patients diagnosed with perforated appendicitis who underwent an appendectomy. A total of 3,491 were identified over 4 years. Among 905 patients with perforated appendicitis, 105 underwent first-line antibiotic therapy. The patients were grouped by intervention, first-line antibiotics vs. first-line appendectomy. No standardized protocol currently exists for management of delayed appendectomy at our institution. The 291 patients were matched with a ratio of 1:2 and based on age, gender, and presence of a fecalith on imaging. Data points including length of stay (LOS), total number of imaging scans, and number of visits to the ED and readmission to the hospital were collected. The values are reported as mean and standard deviation. RESULTS: The first-line antibiotic group had significantly longer primary hospitalization (LOS) in addition to a longer total LOS (158.40 ± 129.10 vs. 108.19 ± 97.91, p < 0.0001; 199.72 ± 142.65 vs. 118.80 ± 93.217, p < 0.0001). These were readmitted more often (0.21 ± 0.48 vs. 0.08 ± 0.311, p = 0.0026) though ED visits were statistically similar to primary appendectomy (0.17 ± 0.40 vs. 0.10 ± 0.35, p = 0.1024). Re-hospitalization LOS was not longer (p = 0.2000). The first-line antibiotic group also underwent more imaging scans during their initial hospital visit as well as after the primary diagnostic scan (1.02 ± 0.46 vs. 0.54 ± 0.57, p < 0.0001; 0.26 ± 0.52 vs. 0.06 ± 0.28, p < 0.0001). CONCLUSIONS: In this study, we found that delayed appendectomy is associated with longer hospital stays, increased hospital admissions, and more imaging scans. Readmissions are also higher. These outcomes may be related to selection bias as well as lack of a standardized approach outlining when to scan patients and in access to outpatient surgical care.
Utility of Cardiac CT vs. MRI in Mitral Valve Architecture Evaluation
(2018-04-04) Chauhan, Siddharth; Bajona, Pietro; Kay, Fernando; Vela, Ryan
INTRODUCTION: New minimally invasive and percutaneous techniques are being developed to repair and replace the mitral valve. These procedures rely upon the knowledge of the exact structure of the mitral valve and its surrounding structures. However, there has not been a standard method identified to accurately assess the mitral apparatus. METHODS: Perform an anatomic study to evaluate various structures associated with the mitral valve and mitral valve apparatus in 40 cadaveric hearts. The anatomic study will serve as baseline for comparison with a subsequent pilot imaging study, consisting of retrospective in vivo assessment of mitral valve and associated structures with cardiac computed tomography (CT) and magnetic resonance imaging (MRI) in four subjects. Main anatomic structures were qualitatively assessed on both imaging methods using a 5-point Likert scale. Imaging measurements were compared with anatomic study measurements. RESULTS: After analyzing the cardiac CT and cardiac MRI images, we found the cardiac CT was the only modality reliable to obtain measurements of the mitral valve apparatus. The cardiac MRI was unreliable in assessing the mitral valve apparatus and is supported by the quality assessment, which demonstrates that cardiac CT images are better to read for the mitral valve structures that were analyzed. CONCLUSION: Cardiac CT is the superior method in identifying the mitral valve apparatus in comparison to cardiac MRIs. The use of cardiac CTs may benefit physicians to formulate pre-op treatment plans but needs further research and development.