Digital Imaging in Anatomic Pathology

Similar documents
SPOT PathSuite Solutions

MIRAX SCAN The new way of looking at pathology

Digital Pathology at Johns Hopkins Practical Research and Clinical Considerations

MacroPATH The new line of Digital Imaging Systems for Grossing

Teaching Digital Histology

Digital Microscopy. Past, Present, and Future. Cyrus V. Hedvat, MD, PhD

InScape: Making Virtual Pathology a Reality

Light Microscopy. Upon completion of this lecture, the student should be able to:

A Practical Guide to Frozen Section Technique

Color aspects and Color Standardization in Digital Microscopy

PathStation TM Fits Your Lab

MODULE No. 34: Digital Photography and Enhancement

DICOM-compatible compression of WSI and diagnostic evaluation

Digital Pathology Update

THEORY AND APPROACHES TO AUTOMATED IMAGE ANALYSIS IN DIGITAL PATHOLOGY

Compressive Through-focus Imaging

FRAUNHOFER INSTITUTE FOR INTEGRATED CIRCUITS IIS. MANUAL PANORAMIC MICROSCOPY WITH istix

Novel machine interface for scaled telesurgery

Digital Imaging in Pathology

RISE OF THE HUDDLE SPACE

Arcturus XT Laser Capture Microdissection System AutoScanXT Software Module. User Manual

Digital Pathology and Tissue-based Diagnosis. How do they differ?

GE Healthcare. Senographe 2000D Full-field digital mammography system

Understanding Infrared Camera Thermal Image Quality

Automated Digitization of Gram Stains. Centralized Reading. Decentralized Assessment. Improved Quality Management.

An Image Processing Approach for Screening of Malaria

Technical Aspects in Digital Pathology

Stereotopix Research. Precision Pathology. Highthroughput. pathology. powered by newcast. Advantages of Stereotopix : RUO

SECTION I - CHAPTER 2 DIGITAL IMAGING PROCESSING CONCEPTS

Positive Pixel Count Algorithm. User s Guide

Feasibility and Design for the Simplex Electronic Telescope. Brian Dodson

The Trend of Medical Image Work Station

SCIENTIFIC INSTRUMENT NEWS. Introduction. Design of the FlexSEM 1000

IMAGE PROCESSING PAPER PRESENTATION ON IMAGE PROCESSING

GE Healthcare. Essential for life. Senographe Essential Full-Field Digital Mammography system

THE ART OF THE IMAGE: IDENTIFICATION AND REMEDIATION OF IMAGE ARTIFACTS IN MAMMOGRAPHY

CLEMEX intelligent microscopy

Image Extraction using Image Mining Technique

34X ZOOM. PathStand TM Routine. Superior Image Quality. PathStand Key Features. 1X Zoom. 8.33X Zoom

IHE Anatomic Pathology Redesign. Sardinia, Italy Nov , 2017

Advancing the Art of Endoscopy

Introduction Approach Work Performed and Results

Image Display and Perception

Multilayer scanning enhances sensitivity of artificial intelligence-aided Mycobacterium tuberculosis detection

CR Basics and FAQ. Overview. Historical Perspective

Thermography. White Paper: Understanding Infrared Camera Thermal Image Quality

DMETRIX S (FUTURE) PERSPECTIVES ON DIGITAL IMAGING & DIGITAL PATHOLOGY SYSTEMS

MECOS-C2 microscopy systems

Digital radiography: Practical advantages of Digital Radiography. Practical Advantages in image quality

4321 FIRST STREET LOREM ET $525,000

Centre for Computational and Numerical Studies, Institute of Advanced Study in Science and Technology 2. Dept. of Statistics, Gauhati University

Virtual 3D Microscopy Using Multiplane Whole Slide Images in Diagnostic Pathology

Quality and GLP for Histology and Pathology of Drug Safety Studies

DIGITAL PHOTOGRAPHY Camera and image capture

Regulatory Forum. Society of Toxicologic Pathology Position Paper on Pathology Image Data: Compliance with 21 CFR Parts 58 and 11

34X ZOOM. PathStand TM Routine. Superior Image Quality. PathStand Key Features. 1X Zoom. 8.33X Zoom

USING SIGNATURE IDENTIFICATION FOR RAPID AND EFFECTIVE X-RAY INSPECTION OF BALL GRID ARRAYS

Second Announcement Call for Participation. (Evaluation Criteria added)

Project Lead the Way: Civil Engineering and Architecture, (CEA) Grades 9-12

Quantitative analysis and development of a computer-aided system for identification of

Leica DMi8A Quick Guide

Digital Pathology and Image Analysis. Queen s University Department of Pathology and Molecular Medicine Shakeel Virk

Virtual Microscopy: Potential Applications in Medical Education and Telemedicine in Countries with Developing Economies

8.2 IMAGE PROCESSING VERSUS IMAGE ANALYSIS Image processing: The collection of routines and

AUTOMATED MALARIA PARASITE DETECTION BASED ON IMAGE PROCESSING PROJECT REFERENCE NO.: 38S1511

CPSC 4040/6040 Computer Graphics Images. Joshua Levine

A Laser-Based Thin-Film Growth Monitor

The SENSE Ghost: Field-of-View Restrictions for SENSE Imaging

What you should have learned from the microscope labs.

1. Editorial. N 9 June Content

Scanning. Records Management Factsheet 06. Introduction. Contents. Version 3.0 August 2017

Scopis Hybrid Navigation with Augmented Reality

Real-Time in vivo Observation of Cells and Nuclei Opens New Possibilities for Diagnostic Endoscopy

Title: Thinking with the Eyes Author(s): Elizabeth Haggerty Hutton Date Created: 8/5/2011 Subject: Biology Grade Level: 9 th Grade Honors Standards:

Biology 29 Cell Structure and Function Spring, 2009 Springer LABORATORY 1: THE LIGHT MICROSCOPE

Dynamic Phase-Shifting Microscopy Tracks Living Cells

2 nd generation TOMOSYNTHESIS

used for low power magnification of a sample image is 3 dimensional

Trust the Colors with Olympus True Color LED

DICOM Correction Proposal

INFORMATION FOR RELATIVES

BodyViz fact sheet. BodyViz 2321 North Loop Drive, Suite 110 Ames, IA x555 www. bodyviz.com

USTGlobal. VIRTUAL AND AUGMENTED REALITY Ideas for the Future - Retail Industry

X o- System. Chiro- Systems. Chiropractic X-Ray Systems and Digital Solutions. Digital Radiography Computerized Radiography Chiropractic Systems

preface Motivation Figure 1. Reality-virtuality continuum (Milgram & Kishino, 1994) Mixed.Reality Augmented. Virtuality Real...

1. Redistributions of documents, or parts of documents, must retain the SWGIT cover page containing the disclaimer.

Slide 1. Slide 2. Slide 3 ACR CT Accreditation. Multi-Slice CT Artifacts and Quality Control. What are the rules or recommendations for CT QC?

The TRC-NW8F Plus: As a multi-function retinal camera, the TRC- NW8F Plus captures color, red free, fluorescein

Using the Microscope for a NANSLO Remote Web-based Science Lab Activity

Robert B.Hallock Draft revised April 11, 2006 finalpaper2.doc

Digital Microscopy: New Paradigm's for Teaching Microscopic Anatomy and Pathology

Keywords: Data Compression, Image Processing, Image Enhancement, Image Restoration, Image Rcognition.

Y N C R O S C O P Y A DIVISION OF THE SYNOPTICS GROUP

You won t be able to measure the incident power precisely. The readout of the power would be lower than the real incident power.

DISTORTlONS DUE TO THE SLIDING MICROTOME

Applications of Optics

PACS Fundamentals. By: Eng. Valentino T. Mvanga Ministry of Health and Social Welfare Tanzania

Digital images and image compression

XM: The AOI camera technology of the future

II. Basic Concepts in Display Systems

Transcription:

your lab focus CE update [generalist histology] Digital Imaging in Anatomic Pathology Ted F. Beals, MD From the Department of Veteran Affairs, Veterans Health Administration, and Department of Pathology, University of Michigan Medical School, Ann Arbor, MI After reading this article, the reader should be able to describe the applications of digital images in pathology, specify some of the factors that influence the quality of digital images, identify values of digital images in pathology, and list essential features of practical applications of digital images. Generalist exam 0102 questions and the corresponding answer form are located after the Your Lab Focus section, p. 335. Applications of digital imaging in pathology Factors influencing image quality Essential features for practical application Current and future uses Digital imaging in anatomic pathology has become a recognized frontier. The technical capabilities are reaching a level that for some, but not all uses, the currently available cameras, software, processing speeds, storage, and displays are acceptable. Telepathology is no longer a fringe concept. Over the past decades, pathologists pictures have evolved from heavy 4" 5" glass lantern slides transported in large, wood cases, to 35-mm transparencies, to digital-image computer files. Early uses of captured digital image files in anatomic pathology were of highly magnified single cells, such as in hematology or cytopathology; the modest resolutions of the early capture devices yielded pleasing images only at high magnifications. Now images of whole tissue sections, captured through the full section depth, have been realized, and soon it will be possible to merge images from serial sections of a biopsy specimen into a single 3-dimensional image file. Such reconstructions will enable the viewing of specimens at any magnification from any direction and in any portion of the specimen with the aid of a navigation device. Capturing the Image A whole spectrum of objects in anatomic pathology is now routinely captured digitally (see sidebar on page 330). For the purposes of this review, digital image is defined as an image captured and stored as an electronic file. Some digitized images in pathology are captured with a document-scanning device. Examples include pathology reports, specimen drawings, pathology reports from other institutions that have wet signatures, and reports that contain graphics. As medical facilities move toward computerbased patient medical records, electronic scanning of documents becomes essential. Nearly all other images for anatomic pathology are captured by a camera. The expanding array of cameras can be simplistically divided into video and direct digital devices. Costs range from tens of dollars to tens of thousands of dollars, and, as might be expected, the quality of the captured images varies proportionately. The digital cameras use a variety of different detectors intended to maximize resolution, the area of the viewed material, and flexibility in viewing conditions and to reduce light requirements (reducing capture times minimizes specimen drift and the effects of workstation vibration on the image). Output from the cameras must be processed into files or, for some telepathology applications, as streams of electronic data. Some of the direct digital cameras process the captured information to enhance perceived resolution. Video camera images can be captured on recording tape, but for this review digital images from video cameras are usually grabbed in time to generate a single file. For the purposes of telepathology, it is useful to define images as static or dynamic. Images captured on 35-mm transparencies are static, while images on videotape are dynamic. Snapshots are static; dynamic images continue to be recorded while moving over the specimen. It is possible to generate dynamic imaging with a direct digital camera if the capture time is rapid enough that a sequence of images can be displayed. Dynamic images are usually expressed as frames per second. Attributes of Digital Files Important attributes of digital image files include: Spatial resolution Area of specimen captured Color quality (technically referred to as color depth) File size Permanence of image content when files are repeatedly opened and closed Innovative technical advances that influence each of these attributes and allow manipulations of the digital information during capture and processing are introduced regularly. There are also a variety of digital image file formats. It is not the intent of this article to discuss this complex topic; it is sufficient to say that not every file can be seen on every platform and that many of the file formats have evolved to conserve file size, but with loss of image attributes. Although it may seem counterintuitive, there is an inverse relationship between the working magnification of 327

your lab focus 328 Glossary Depth of color Quantity of information for a pixel that describes the color attributes (usually in shades of red, green, and blue) Digital image Image stored as a computer file Pixel Rectangular-shaped, smallest defined area of a digital image Reference digital image Image of a lesion selected by a subject matter expert as characteristic of that lesion Spatial resolution Minimal distance between discernable objects in an image Telepathology Transmission of images of patient specimens to pathologists at remote locations for purposes of consultation, diagnostic confirmation, or primary diagnosis Wet signature Handwritten signature on a source document. (As opposed to an electronic signature, which is certified true and validated as being part of an actual, unalterable document electronically.) the view being imaged and the file size (amount of digital information) necessary to capture a faithful rendition. Thus images of single cells filling the viewing screen (as in cytopathology or hematology) require significantly less digital information than do low-magnification views of an inflammatory lesion. A thin optical plane of a cell requires less information than that needed to store the entire image of a 1- mm diameter core biopsy that would enable viewing of any area of the biopsy specimen at any working magnification. File sizes may vary tremendously depending on the spatial resolution, color depth, compression algorithms, and viewing area of the system selected. Color, a quality that nearly all pathologists consider essential, adds technical requirements for the imagecapture process and significantly increases the size of the resulting digital files. It also introduces one more variable in the characteristics of the digital images, since the depth of color is complex and the amount of information in bits required to define the color of each pixel in the image ranges from 256 shades of red, green, and blue, to 16.7 million shades and going up. Image Quality Image quality, which is influenced by a number of variables [T1], is an important and often subjective characteristic. The goal is to capture an accurate depiction of the specimen, but the accuracy needed for different uses varies considerably. For display of Aspergillus hyphae in a transbronchial biopsy specimen to a patient s clinical care team, color is important and modest resolution is adequate. For debate at a hematopathology conference on the subtle differences in a lymphoma that had been in remission but is now rapidly enlarging requires every bit of resolution; modest color rendition would not be objectionable. Rare studies of the characteristics of pathology digital image files demonstrate that bigger may not be necessary or even better. 1 In an attempt to skirt the complexity and ongoing technological changes in these variables, the Veterans Health Administration adopted standards for telepathology that use, as the image quality standard, the ability of those pathologists who will be using the system to diagnose, with comparable accuracy, specimens viewed by standard (on site) practice and those viewed with the system. This is, after all, the only valid criterion for diagnostic uses of digital images. And it is important to remember that one of the most important variables is the diagnostic challenge of the lesions being examined, which is independent of the image-capture system. The claims of digital image quality must be weighed with a recognition that although we all know well the detrimental effects of poor specimen fixation, processing, and staining, and the variability of the optical properties of the microscopes in use, there is a paucity of professional standards for specimen processing, microscope quality, or even individual pathologist s optical acuity. Why not capture images with the maximum available resolution and color depth? As stated, file size grows as the information captured increases. File space is cheap; the real problem is time. As the settings for image quality are increased, it takes longer for the camera to capture the image, slightly longer to process the digital information, and significantly longer to bring the image onto a display monitor. And for telepathology, larger files present technical, cost, and transmission-time challenges. Images should be tailored to the intended use. The Potential to Alter Digital Images Radiologists have become concerned about the ability to alter, augment, or enhance diagnostic radiologic digital images. Pathologists have devoted less attention to this because pathology images are secondary, not primary, material. But when the digital images are used in diagnosis, the ethical problems are similar. It has always been possible to alter photographs, but the ease and sophistication of altering digital images electronically makes this activity more potentially troubling in the medical community. Areas of interest may be highlighted by the direction, intensity, and focus of lighting. Images can be manipulated to ensure that the colors are true. Circles and arrows are superimposed to point the viewer to areas of interest. More ethical concerns relate to electronically smoothing out or sharpening edges or specifically changing the pixel characteristics of a specimen s image to make some characteristic of the lesion more obvious. There is concern whether annotations are superimposed on the digital image or replace the underlying image detail. Most careful pathologists retain a file with the

your lab focus unaltered digital image, and original slides are retained. This issue will become increasingly important as digital images become standard practice in pathology and are used for primary diagnosis or consultations. If digital images are retained for documentation of specimens or slides that are discarded, electronic alteration will become a major concern. Applications Clinical An early use of digital images in pathology had origins in the concept that the ability to diagnose lesions accurately and consistently would be enhanced if morphologic information could be quantified. This led to quantitative cytopathology and quantitative histopathology. Workstations were designed to digitally capture images and then, with the use of increasingly sophisticated algorithms, render precise information about the morphologic features of lesions and in some cases render diagnostic conclusions. These attempts have failed to provide more clinically relevant or reliable information than that provided by pathologists. There continues to be a large volume of research with this technology and advances emerge, but, if anything, the current trend is toward replacing imprecise Images and Communication Variables Influencing Image Quality Characteristics of the specimen and specimen preparation Resolution and quality of the microscope and the lens system Matching of the glass slide and the microscope design Optical alignment and settings of the microscope Spatial resolution and color rendition of the camera Stability of the setup to minimize vibration during capture Graphic capture hardware and software Processing of files File format Display computer characteristics Display monitor characteristics Conditions in the viewing room Visual characteristics of the observer morphologic criteria for identifying lesions with nonmorphologic attributes that predict appropriate therapeutic interventions. One predictable clinical application of digital image analysis is in the screening of uterine cervical cells for the early detection of cancer. Insurmountable pressures are building to substitute dedicated digital image workstations for cytotechnologists and pathologists reviewing the classic Papanicolaou smear, even though the use of T1 this technology will probably require a universal change in the practice of specimen collection and a significant increase in the cost of the service. Educational The use of digital images for publication is skyrocketing. Authors are being encouraged to submit their images in digital form; some publishers require it. Increasing numbers of publications are Choice and level of communication format depends on the (1) level of communication; (2) precision of essential communication; and (3) available formats. To use a real-estate analogy to illustrate the kind of image that is appropriate to a medical event: If you wanted to: And you chose to: Show friends the new Show them some snapshots you summer cabin you purchased took of the cabin and the view of the lake from the front porch? Illustrate a new book on the architectural styles of southwestern ranch homes Review prospective houses with a realtor Buy a $400,000 house Have a housing inspector advise you on the potential problems with the house you are ready to buy Use full-page prints of color photos taken by a professional photographer? Look at a video tour of the neighborhoods and several photos of the exterior and interior of houses on the market? View 25 photos the seller made of the house and property? Show them 25 photos you took going from the basement to the attic with your new, $3,000 high-tech 35- mm camera, using the remarkable new zoom lens and the highest resolution film available Would this be satisfactory? Probably Yes Yes Probably not Definitely not This is an example of image use in: Patient medical record; clinical team review of case Professional publication Clinical conference Pathologist's second opinion T2 Expert consultation on case you do not feel experienced enough to diagnose 329

your lab focus 330 Uses of Digital Images in Anatomic Pathology Gross images of surgical specimens Fresh whole After blocks are removed to document locations After dissections showing nature and extent of lesion Microscopic images of surgical specimens Cytology specimen to document character of material Cytology, microscopic images Hematology smears Electron microscopy Of light microscopy sections to document areas selected for transmission electron microscopy Transmission and scanning images Autopsy In situ of dissection Gross organs and lesions Microscopic images Scanned documents Patient signed consent forms Documents with wet signatures Graphic information from flow cytometry, cytogenetics, and immunochemistry studies Consultation reports with full letterhead and signatures Matrix of staining patterns Lesion location documentation In situ of tumors aspirated for cytology Dermatology in situ images of lesions Endoscopic images of location of biopsy or brushings Laparoscopic images of lesions sampled Surgical procedure in situ images Assembled images for clinical conferences or teaching Telepathology Gross surgical specimens to assist remote dissection Microscopic Consultation Second opinion/confirmation of diagnosis Annotation of any images as an overlay Atlases for comparison, teaching, and reference Macroscopic images of small objects Bacterial colonies Parasites available on the Internet or as files on compact disks (CDs). It is not unrealistic to predict that nearly all published material will soon be accessible in digital format. One advantage of this change is that digital publications provide images, especially full-color images, more cost-effectively. Pathology atlases are now appearing on CD that contain more images, or more color images, than in the corresponding print publication. Some pathology atlases are available exclusively in digital format. One disadvantage to this advance is that the images are subject to degradation by the limitations of the individual display workstation. By combining Internet access and digital-imaging capability, new pathology Web sites are appearing monthly. Many allow interactive sessions and weekly or monthly cases of interest. They are intended to be used for teaching and consultations and to enhance the visibility of the pathology departments that publish them. They take advantage of the accessibility of the Internet and are excellent examples of the variability in the quality of digital images. It is not uncommon today to see medical schools substituting the requirement that incoming students acquire a microscope with the requirement that they have a laptop computer with CD port. Having quality digital images is essential to a pathology teaching program that uses the Internet, online interaction, or CD files. It is also apparent that welldesigned interactive digital teaching modules can serve, at least in part, as a substitute for the traditional pathology curriculum. But having taught for many years, I am constantly reminded that pathology has often been singled out as the most personally satisfying course for medical students because of the ongoing interaction with spirited pathologists who make pathology an exciting topic. I am concerned that creators of interactive digital pathology courses, as talented as they are, will not fill our students need for role models. Much has been written about the adequacy of images for different medical purposes; the scenarios in T2 focus on the diversity and complexity of the subject and help illustrate the fact that one image does not fit all. There has been an interest in using digital images in pathology competency examinations, much of it promoted by those with financial interest in digital image devices. But it should be remembered that the principal standard for assessing professional proficiency must be comparison of individual ability with that of one s peers, using challenges that closely mimic those of the current practice of pathology. And until pathologists routinely rely on digital images for diagnosis, it is inappropriate to challenge them with digital images. Conclusion So what is the promise of digital imaging in pathology? The answer is ready accessibility. Although the quality of a digitally captured pathology image using the latest and best technology and optimal setup does not yet match that of the 35- mm color slide, accessibility overshadows the imperfections. When a clinician drops in to review a case, or one is in the radiology department reviewing the radiologic images of a lesion and seeking to compare the gross or microscopic features, the ability to retrieve the digital images rather than hunt up slides in the storage room becomes a blessing. Digital images are unquestionably becoming a staple in the practice of pathology. The technology has passed the threshold of acceptable and is rapidly expanding to reach the current gold standard of photographic images. For digital images to become an integral part of daily practice, several criteria must be met: The quality of the images must be appropriate for the intended use. The cost of the technology for capture and display must be within the limited resources of a practicing pathologist. The act of image capture must be seamless enough that it is not perceived as an impediment to getting daily work completed, or at least mitigated by the ease of accessibility of the image.

your lab focus For most images of diagnostic material, the interpreting pathologist should be responsible for the capture. The image files should be stored on a networked source and indexed by patient identification number, accession number, and usually by some coding system to enable searches for related lesions. Only when these conditions are met can the value of digital images in anatomic pathology, and in medicine in general, be realized. It is easy to predict that digital images will become, within the next decade, standard in medical school and continuing education, scientific meetings, publications, medical conferences, and medical records. Telepathology will be commonplace for second opinions, consultations, and even primary diagnoses. Reference digital images will be available at readout workstations for assistance in differential diagnosis and comparison. Perhaps digital displays will supersede optical microscopes for the viewing of cases. 1. Doolittle MH, Doolittle KW, Winkelman Z, et al. Color images in telepathology: how many colors do we need? Hum Pathol. 1997;28:36-41. Downloa