Virtual Library

 x 
My Library Virtual Library FAQ

Start Your Search

Ricardo S Avila



Author of

  • +

    P2.13 - Radiology/Staging/Screening (ID 714)

    • Event: WCLC 2017
    • Type: Poster Session with Presenters Present
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      P2.13-026b - A Novel Ultra Low Cost CT Image Quality Measurement Device (ID 10341)

      09:30 - 09:30  |  Presenting Author(s): Ricardo S Avila

      • Abstract

      Background:
      Assessing CT image quality is becoming of increasing concern in the domain of quantitative imaging. Current calibration devices tend to be time-consuming to use and often require special expertise for analysis. We have developed a novel approach for measuring image quality on CT scanners that is automated and inexpensive.

      Method:
      Three new rolls of 3M 3/4x1000 Inch Scotch Magic tape($1.50 each) were placed radially out from iso-center and CT scanned using standard head, body,and low dose lung protocols on a GE VCT and a Siemens Somatom Definition AS scanner. A Gammex 464 ACR CT Accreditation phantom was also scanned on the same scanners with identical protocols. GE and Siemens scans were reconstructed with 0.625, 1.25,and 2.5mm and 0.6, 1.0,and 2.0mm slice thickness and spacing, respectively. A total of 36 3D CT scans(36=2 objects x 2 scanners x 3 protocols x 3 thicknesses) were used for this study. Automated analysis was performed using Radia Diagnostic Software(Radiological Image Technology, Inc.) for the Gammex scans and Accumetra software for the tape scans. Both software tools produced measurements for CT linearity(air and acrylic HU), in-plane resolution, slice thickness,and image noise. Mean, standard deviation,and difference in measurements was used to evaluate performance.

      Result:

      Gammex Mean, SD Tape Mean, SD (Tape-Gammex) Mean, SD
      Air (HU) -988, 10.4 -995, 4.6 -6.97, 6.38
      Acrylic (HU) 130, 2.0 121, 12.3 -8.90, 12.75
      In-plane Resolution (LP/cm) 6.32, 0.31 6.09, 0.67 -0.23, 0.91
      Slice Thickness (mm) 1.88, 1.13 1.42, 0.57 -0.46, 0.63
      Image Noise (HU SD) 13.39, 9.93 7.05, 2.65 -6.35, 8.47
      Given that mean tape measurements differed from Gammex phantom measurements by <10 for HU density,<0.25 for LP/cm of in-plane resolution,<0.5 for mm of slice thickness, and <10 for HU SD of image noise, scotch tape has the potential to be used as a fast, accurate, and inexpensive tool for assessing CT scanner and protocol image quality.

      Conclusion:
      A new automated and inexpensive method for CT scan image quality assessment that relies on advanced image processing techniques provides results comparable to standard calibration methods thus allowing CT scan calibration to be performed rapidly and inexpensively allowing for more comprehensive integration of quality standards into daily practice.

  • +

    P3.13 - Radiology/Staging/Screening (ID 729)

    • Event: WCLC 2017
    • Type: Poster Session with Presenters Present
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      P3.13-035 - Automatic Estimation of Measurement Error on CT Imaging (ID 10333)

      09:30 - 09:30  |  Presenting Author(s): Ricardo S Avila

      • Abstract

      Background:
      There has been increasing recognition that lung nodule measurement on CT scans is imprecise and that an understanding of the extent of this imprecision is necessary when trying to determine whether actual change in volume has occurred. The various factors that influence this are numerous with two of the most prominent being the overall quality of the CT scan (including all of the adjustable parameters) and the size of the nodule.

      Method:
      We have developed an automated system whereby a calibration device is scanned on a given scanner with a given protocol and then the system can automatically predict the extent of measurement error for a given size solid nodule. We compared this approach to empirically derived results obtained from a database of 117 screen-detected stable nodule ranging in size from 2.2 to 18.7 mm that were scanned twice on the same CT scanner using the same protocol. Automated volumetric analysis was performed using commercial software. This allowed us to determine the relationship between standard deviation of the measurements versus nodule size. We then scanned our calibration device using the same scanning protocol as was used on those nodules to automatically calculate the size and standard deviation relationship.

      Result:
      Predicted solid nodule volume standard deviation compared with empirically derived values across a range of nodule sizes was within 20% (see figure)Figure 1



      Conclusion:
      Results from our automated approach were highly correlated with results obtained from scans obtained in actual clinical practice. The ability to predict extent of error specific to a given scanner and scanning protocol is an essential step in understanding whether change has occurred and has implications for both diagnosis and therapy assessment, including predicting when a follow up scan should be obtained. This type of information will ultimately become a necessary component of all quantitative imaging programs.

  • +

    PL 01 - Prevention, Screening, and Management of Screen-Detected Lung Cancer (ID 586)

    • Event: WCLC 2017
    • Type: Plenary Session
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      PL 01.02 - Major Advances in CT Screening: A Radiologist's Perspective (ID 7838)

      08:35 - 08:55  |  Author(s): Ricardo S Avila

      • Abstract
      • Presentation
      • Slides

      Abstract:
      Advances in CT scanners. CT screening was first introduced when helical CT scanners became available in the early 1990’s (1-4). Since then, there have been remarkable advances in CT scanner technology with concurrent increase in the number of CT examinations per year by approximately 10% annually. More powerful hardware and image reconstruction algorithms have allowed faster scanning at lower radiation doses in today’s multidetector CT (MDCT) scanners. Ultra low-dose techniques are gaining acceptance. With respect to lung cancer screening, thinner collimation now possible has led to the detection of many more small pulmonary nodules. Also, there have been evolutions in diagnostic techniques such as percutaneous biopsies, navigational bronchoscopy, and PET scans and these advances have been integrated into the regimen of screening with a resulting decrease in the frequency of surgical resection of benign nodules (5). Definition of Positive Results. Updates in the definition of positive results have continued to be developed that allow for improvements in the efficiency of workups. One of the major changes has been to update the size thresholds for positive results from 4 to 6 mm and also to avoid rounding errors (6, 7). The NELSON trial introduced the concept that a positive result should be based on the initial CT scan and a follow-up CT scan for small nodules, rather than solely on the initial CT scan and this has been adopted by I-ELCAP (6). The I-ELCAP and NLST databases have been used to provide follow-up strategies for nonsolid and part-solid nodules (6). Considerations as to screening frequency may substantially reduce costs for lower risk individuals. There is increasing recognition that different approaches are needed for baseline and repeat scans where even when nodules might have the same characteristics as they should be managed differently. The management of both nonsolid and part-solid nodules has dramatically changed. For the first time, imaging as a biomarker for aggressiveness has been used to monitor whether a cancer is progressing. Growing nonsolid nodules can be followed on an annual basis and only the emergence of a solid component triggers more aggressive intervention. For the part-solid nodule it has now been recognized that the important component from a prognostic perspective is the solid portion not the overall size. Quantitative assessments. Quantitative assessment of many findings on chest CT scans have been developed (6). In particular, assessment of nodule size and growth as to the probability of malignancy and lung cancer aggressiveness has progressed. Most guideline organizations have moved from a single measurement of length to an average diameter (average of length and width) (6) and to three measurements of volume (7). The errors involved in any of these measurements are influenced by multiple factors including the intrinsic properties of the nodule and the software used to make the measurement (8, 9). Additionally, they are impacted by the variability of CT scanners and their adjustable scan parameters. Advances in incorporating measurement errors into growth assessment by RSNA’s Quantitative Imaging Biomarkers Alliance (QIBA) has led to a web-based calculator. The American College of Radiology (ACR) specifies that growth for a nodule of any size requires “an increase of 1.5 mm or more.” Both approaches allow for large measurement errors for the wide range of CT scanners and the protocols. The I-ELCAP guidelines for solid and the solid component of part-solid nodules is given explicitly in I-ELCAP protocol (6). Each of these approaches has specific technical requirements as measurement error is influenced by both the scanner itself, the choice of various adjustable parameters on the scanner (slice thickness, slice spacing, dose, FOV, pitch, recon kernel etc.) as well as characteristics of the nodule itself. Additional considerations for computer-assisted volume change assessment requires: 1) inspecting the computer scans and the segmentation for image quality (e.g. motion artifacts) and for the quality of the segmentation; 2) the radiologist visually inspecting both nodule image sets side-by-side to verify the quality of the computer segmentation for each image that contains a portion of the nodule; 3) examination of the segmentations for errors such as when a vessel is segmented as part of a nodule in one scan but not in the other; 4) that the scan slice thickness for the purpose of volumetric analysis should be 1.25 mm or less. When using any computer-assisted software, the radiologist must be satisfied with the CT image quality and the computer segmentation results, further substantiating the notion that the decision of whether growth has occurred is ultimately based on clinical judgment. Innovations in use of imaging and genetic information. Radiomics is an emerging field of study on the quantitative processing and analysis of radiologic images and metadata to extract information on tumor behavior and patient survival (10). The hypothesis is that data analysis through automated or semi-automated software can provide more information than that of a physician. Its use has shown improved diagnostic accuracy in discriminating lung cancer from benign nodules. It has been used successfully in breast imaging, with 2017 FDA approval of a computer-aided diagnosis tool which utilizes advanced machine learning analytics. Furthermore, radiomics has been linked with the field of genomics, inferring that imaging features are closely linked to gene signatures such as EGFR expression, a known therapeutic target. In the future, as larger data sets emerge and inter-institutional sharing of images becomes more commonplace, radiomics will become more tightly integrated with lung cancer diagnosis, treatment planning, and patient survival prognostication. References 1. Henschke C, McCauley D, Yankelevitz D, Naidich D, McGuinness G, Miettinen O, Libby D, Pasmantier M, Koizumi J, Altorki N, and Smith J. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999; 354:99-105. 2. The International Early Lung Cancer Action Program Investigators. Survival of Patients with Stage I lung cancer detected on CT screening. NEJM 2006; 355:1763-71 3. Kaneko M, Eguchi K, Ohmatsu H, Kakinuma R, Naruke T, Suemasu K, and Moriyama N. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 1996; 201: 798-802. 4. Sone S, Nakayama T, Honda T, Tsushima K, Li F, Haniuda M, et al. Long-term follow-up study of a population-based 1996-1998 mass screening programme for lung cancer using mobile low-dose spiral computed tomography. Lung Cancer. 2007; 58:329-41. 5. Linek HC, Flores RM, Yip R, Hu M, Yankelevitz DF, Powell CA. Non-malignant resection rate is lower in patients who undergo pre-operative fine needle aspiration for diagnosis of suspected early-stage lung cancer. Am J Respir and Crit Care Med 2015; 191: A3561 6. International Early Lung Cancer Action Program protocol. http://www.ielcap.org/sites/default/files/I-ELCAP%20protocol-v21-3-1-14.pdf Accessed March 27, 2015 7. Van Klaveren RJ et al. Management of Lung Nodules Detected by Volume CT Scanning. N Engl J of Medicine 2009; 361: 2221-9 8. Henschke CI, Yankelevitz DF, Yip R, Archer V, Zahlmann G, Krishnan K, Helba B, Avila R. Tumor volume measurement error using computed tomography (CT) imaging in a Phase II clinical trial in lung cancer. Journal of Medical Imaging 2016; 3:035505 9. Avila RS, Jirapatnakul A, Subramaniam R, Yankelevitz D. A new method for predicting CT lung nodule volume measurement performance. SPIE Medical Imaging 2017: 101343Y 10. Lee G, Lee HY, Park H, et al. Radiomics and its emerging role in lung cancer research, imaging biomarkers and clinical management: State of the art. Eur J Radiol. 2017; 86:297-307.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • +

    WS 01 - IASLC Supporting the Implementation of Quality Assured Global CT Screening Workshop (By Invitation Only) (ID 632)

    • Event: WCLC 2017
    • Type: Workshop
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      WS 01.12 - Planning for USA Registries for CT Screened Images – What Are Their Objectives? (ID 10650)

      10:20 - 10:35  |  Presenting Author(s): Ricardo S Avila

      • Abstract
      • Slides

      Abstract:
      The reimbursement of low dose CT lung cancer screening for high risk populations in the United States by the Centers for Medicare and Medicaid Services (CMS) [1] has been implemented with a requirement to participate in a nationwide registry run by the American College of Radiology (ACR) [2]. This registry’s main purpose is to enable the collection of basic information on lung cancer screening including patients’ demographic information, medical history and risk factors, procedure indications, and follow-up information. Owing in part to the large data sizes of low dose CT lung cancer screening studies, which can exceed 500 MB for each 3D CT scan acquisition, this important US lung cancer screening registry is not collecting CT image data. However, the I-ELCAP study has been collecting international lung cancer screening data, including CT scan images, for over two decades [3]. There are several important benefits to collecting CT lung cancer screening image datasets in addition to basic lung cancer screening information. CT image data provides important information on the quality of actual scans and findings in the field, which can help identify areas of improvement for national screening efforts as well as for the local lung cancer screening site. One of the most important benefits is that expert review of these scans and findings can help train local radiologists on how to improve delivery of lung cancer screening. In addition, many image acquisition characteristics can be automatically evaluated that influence lung cancer screening performance. Determining whether patients are being over scanned (outside the lung region), whether the CT table was properly positioned, and whether the CT reconstruction field of view was properly set can be evaluated are some of the areas that can be evaluated using automated analysis methods provided that the CT scan datasets are available for processing. Also, new image quality standards for CT lung cancer screening data acquisition are becoming available and these requirements can potentially be evaluated against actual scans acquired. Another important benefit that is enabled by CT lung cancer screening image data registries is the potential to identify new imaging biomarkers as well as help improve existing imaging biomarkers. A persistent challenge for lung cancer imaging research groups is to continuously collect lung cancer screening image data obtained from current day patients and using modern CT scanners. Given that CT scanner technology and methods are changing rapidly it is particularly important to have a large continuous source of imaging data, which a large image-based registry can provide. In addition to informed consent to conduct research and patient privacy protections, studies based on registry data can support the lung cancer imaging research community by further collecting additional quantitative metadata with each CT scan. The collection of images allows for retrospective reviews of imaging findings that were not known to be important for the different diseases that may occur in the lungs. One such example is recognition of early interstitial lung disease which can be as deadly as lung cancer [4]. Having the prior images for review once a diagnosis is made allows for future early recognition and for development of follow-up recommendations. Growing recognition of subtypes of nodules (subsolid and solid), both solitary and multiple ones, and review of prior imaging has been important in limiting invasive procedures for certain subtypes [5, 6]. Automated methods can potentially be used by image-based registries to calculate and store the location, surface geometry, and volume of the lungs, suspicious nodules, cancer tumors, and relevant anatomy and pathology. If data transmission bandwidth is a roadblock to collecting image data, automated methods can be employed to at least collect images of identified lung cancers and other targeted areas (e.g. suspicious lung nodule regions). Another opportunity is to document the fundamental image quality characteristics of CT scans, as is becoming available using automated methods. Documenting image quality information within large lung cancer screening image datasets will enable the research community to better understand the relationship between image quality and measures of lung cancer screening success, such as the ability to detect and measure small lung nodules. This data will be critical to help inform the establishment of new minimum imaging standards that are being developed for lung cancer screening studies. Over the next few years several new lung cancer screening initiatives will launch in the United States including an effort to deploy lung cancer screening services at US Department of Veterans Affairs Medical Centers. These lung cancer screening studies will offer a fresh opportunity to collect lung cancer screening image data with modern tools, research targets, and methods. References 1. CMS recommendation to support reimbursement for lung cancer screening, , February 5, 2015. 2. Pederson JH, Ashraf H, Implementation and organization of lung cancer screening, Ann Transl Med. 2016 Apr; 4(8): 152. 3. Yankelevitz DF, Henschke CI, Advancing and sharing the knowledge base of CT screening for lung cancer, Ann Transl Med. 2016 Apr; 4(8): 154. 4. Salvatore M, Henschke CI, Yip R, Jacobi A, Eber C, Padilla M, Koll A, Yankelevitz D. Journal Club: Evidence of Interstitial Lung Disease on Low-Dose Chest CT: Prevalence, Patterns and Progression. AJR AM J Roentgenol 2016: 206:487-94 5. Yankelevitz DF, Yip R, Smith JP, Liang M, Liu Y, Xu DM, Salvatore M, Wolf A, Flores R, Henschke CI. CT screening for lung cancer: nonsolid nodules in baseline and annual repeat rounds. Radiology 2015; 277: 555-64 6. Henschke CI, Yip R, Wolf A, Flores R, Liang M, Salvatore M, Liu Y, Xu DM, Smith JP, Yankelevitz DF. CT screening for lung cancer: part-solid nodules in baseline and annual repeat rounds. AJR Am J Roentgenol 2016; 11:1-9

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • +

    WS 02 - IASLC Symposium on the Advances in Lung Cancer CT Screening (Ticketed Session SOLD OUT) (ID 631)

    • Event: WCLC 2017
    • Type: Symposium
    • Track: Radiology/Staging/Screening
    • Presentations: 1
    • +

      WS 02.15 - Quality Control for Lung Imaging Biomarkers (ID 10628)

      15:30 - 16:00  |  Presenting Author(s): Ricardo S Avila

      • Abstract
      • Presentation

      Abstract:
      Computed Tomography (CT) imaging of the lung has been routinely used over the last few decades to detect and treat early lung cancer and other related diseases. As CT image acquisition technology has improved, the use of CT for quantitative and precise lung imaging clinical applications has greatly expanded. High resolution CT studies, which now easily obtain sub-millimeter resolution of the entire chest within a breath-hold, are now widely used to detect and measure changes in early lung cancer lesions and COPD. Traditionally, several concurrent methods have been used to ensure that the quality of acquired CT images is adequate for general clinical use. This includes regular scanning and analysis of CT quality control phantoms from ACR (as well as from individual CT scanner manufacturers) and visual inspection of acquired images by radiologists for significant image artifacts. While these methods have served the field of radiology well for identifying and correcting major image quality issues, there has not been standard image quality assessment methods available for specific clinical applications that require precise image-based measurements. To improve global quality control of lung imaging studies, several clinical societies and organizations have provided image acquisition and measurement guidance documents intended to be followed by clinical sites [1, 2, 3]. We are entering a new era of quantitative imaging where easy to use tools are available that ensure that precise quantitative image measurements can be routinely and reliably obtained. To achieve this goal, a new set of task-based image quality control measures is being developed by research groups and radiology societies such as the RSNA’s Quantitative Imaging Biomarkers Alliance [4]. Each major quantitative imaging-based clinical task is being extensively studied to determine the fundamental image quality properties needed (e.g. resolution, sampling rate, noise, intensity linearity, spatial warping) to achieve a minimum level of measurement performance. In addition, new low-cost phantoms are being developed that can be quickly scanned and automatically analyzed to estimate these fundamental properties throughout the full three-dimensional CT scanner field of view. Deploying these low-cost phantoms and automated phantom analysis software on the cloud further enables global clinical sites to quickly and easily verify the quality of a CT scanner and acquisition protocol for a specific quantitative clinical task. In addition to providing a fast method for verifying conformance with minimum quantitative imaging performance standards, the reports generated can provide guidance as to the best protocols observed for a particular CT scanner model, thereby allowing a clinical site to optimize image acquisition protocols with the best evidence obtained through crowd-sourcing task-specific image quality information. The QIBA CT lung nodule task force is now preparing to launch a pilot project to evaluate the utility of these new image quality control measures for the quantitative measurement of the change in volume of solid lung nodules (6mm to 10mm diameter) [5]. Over the coming months this new “active” and cloud-based analysis approach will be deployed at international lung cancer screening institutions and use statistics will be assembled. The data collected has the potential not only to inform the lung cancer screening community on the global quality of lung cancer screening imaging, but also to establish early data on whether these new methods can one day serve as a more effective approach to providing quality control for quantitative imaging methods. References 1. Kauczor HU, Bonomo L, Gaga M, Nackaerts K, Peled N, Prokop M, Remy-Jardin M, von Stackelberg O, Sculier JP; European Society of Radiology (ESR); European Respiratory Society (ERS), ESR/ERS white paper on lung cancer screening, ESR/ERS white paper on lung cancer screening. 2. IELCAP, IELCAP Protocol Document, http://www.ielcap.org/sites/default/files/I-ELCAP-protocol.pdf Accessed May 31, 2017. 3. Fintelmann FJ, Bernheim A, Digumarthy SR, Lennes IT, Kalra MK, Gilman MD, Sharma A, Flores EJ, Muse VV, Shepard JA, The 10 Pillars of Lung Cancer Screening: Rationale and Logistics of a Lung Cancer Screening Program, Radiographics. 2015 Nov-Dec;35(7):1893-908. 4. https://www.rsna.org/QIBA/ 5. RSNA QIBA, Draft QIBA Profile: Lung Nodule Volume Assessment and Monitoring in Low Dose CT Screening, http://qibawiki.rsna.org/images/e/e6/QIBA_CT_Vol_LungNoduleAssessmentInCTScreening_2017.05.15.docx, May 15, 2017.

      Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.