`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`MAKO SURGICAL CORP.
`
`Petitioner
`
`V.
`
`BLUE BELT TECHNOLOGIES, INC.
`
`Patent Owner
`
`Case IPR2015-00630
`
`Patent No. 6,205,411 B1
`
`DECLARATION OF ERIC R. KISCHELL
`
`Mail Stop Patent Board
`Patent Trial and Appeal Board
`US. Patent and Trademark Office
`
`PO. Box 1450
`
`Alexandria, VA 22313-1450
`
`Mako
`
`Exhibit 1013 Page1
`
`Mako Exhibit 1013 Page 1
`
`
`
`1.
`
`My name is Eric R. Kischell. I am currently a Senior
`
`Analyst/Programmer at the Mayo Foundation. I hold a MS. in Electrical
`
`Engineering from Texas A&M University, and a BS. in Electrical Engineering
`
`from Northeastern University. My expertise encompasses image
`
`processing/segmentation/display, pattern recognition, 3D visualization, object
`
`oriented architecture, user interface prototyping and code optimization.
`
`2.
`
`I am being compensated at my ordinary rate of $60/hour for my time,
`
`but have no relationship to Mako Surgical Corp. My compensation is not
`
`dependent on the outcome of this case.
`
`3.
`
`4.
`
`A copy of my curriculum vitae is attached hereto.
`
`I began working for Carnegie Mellon University (“CMU”) in July of
`
`1995 as a Research Programmer for The Robotics Institute. I worked at CMU until
`
`December of 1997 in the same capacity. While at CMU, I worked on a team with
`
`Dr. Anthony M. DiGioia III, Dr. Takeo Kanade, Dr. Branislav Jaramaz, David A.
`
`Simon, Robert V. O’Toole, Michael Blackwell, Frederick Morgan, and Bruce
`
`Colgan (“the team”) to develop what we called the “HipNav System.”
`
`5.
`
`Additionally, the team published the following article, which
`
`summarized our development of the HipNav System: “HipNav: Pre-operative
`
`Planning and Intra-operative Navigational Guidance for Acetabular Implant
`
`Mako
`
`Exhibit 1013 Page 2
`
`Mako Exhibit 1013 Page 2
`
`
`
`Placement in Total Hip Replacement Surgery,” 2nd CAOS Symposium, 1996 /
`
`(“DiGioia”) (Ex. 1005).
`
`6.
`
`At the time, it was well understood that malposition of the acetabular
`
`component during total hip replacement surgery caused complications following
`
`surgery, including dislocation and impingement. The HipNav System was
`
`developed to improve placement of the acetabular component during surgery,
`
`through the use of a pre-operative planner, a range of motion simulator, and an
`
`intra-operative tracking and guidance system.
`
`7.
`
`I was tasked with solving the following problem: how to integrate the
`
`intra-operative tracking and guidance system into a graphical user interface that
`
`would allow the surgeon to place the implant in the optimal position.
`
`8.
`
`The data had to be integrated and tracked in real time in a capable
`
`graphical user interface so that the surgeon could utilize the intra—operative
`
`tracking and guidance system to accurately place the implant in the predetermined
`
`optimal position. My solution was the graphical user interface utilized in the
`
`HipNav System as represented in the DiGioia article.
`
`9.
`
`In addition to the graphical user interface issue, we also had to find a
`
`way to process the CT scans so that they could be utilized in the pre-operative
`
`planner.
`
`I was responsible for developing software to perform clean-up of image
`
`Mako
`
`Exhibit 1013 Page 3
`
`Mako Exhibit 1013 Page 3
`
`
`
`contours for 3D object reconstruction. This pre-processing step was also utilized
`
`in the HipNav System as represented in the DiGioia article.
`
`10.
`
`To my knowledge, my solution to these was unique. It took
`
`approximately 14 months to create the graphical user interface, and approximately
`
`20 months to develop the CT scan pre-processing software.
`
`11. After a draft of the DiGioia article was written, I reviewed the draft
`
`and provided further input and revisions before it was published. The development
`
`of the system and the drafting of the article were team efforts, and I was certainly
`
`not the primary researcher on the medical front. Nonetheless, my contributions to
`
`solving the software problems in the HipNav system were important to the
`
`project’s success and were not insubstantial.
`
`12.
`
`In signing this Declaration, I recognize that the Declaration will be
`
`filed as evidence in a contested case before the Patent Trial and Appeal Board of
`
`the United States Patent and Trademark Office. I also recognize that I may be
`
`subject to cross-examination in the case and that cross—examination will take place
`
`within the United States. If cross-examination is required of me, I will appear for
`
`cross—examination within the United States during the time allotted for cross—
`
`examination.
`
`13.
`
`I hereby declare that all statements made herein of my own
`
`knowledge are true and that all statements made on information and belief are
`
`Mako
`
`Exhibit 1013 Page 4
`
`Mako Exhibit 1013 Page 4
`
`
`
`believed to be true, and further that these statements were made with the
`
`knowledge that willful false statements and the like so made are punishable by fine
`
`or imprisonment, or both, under Section 1001 of Title 18 of the United States code.
`
`é/vé' % QM
`
`Eric R. Kischell
`
`2 [/9 Z 29/1?
`
`Date
`
`
`
`Mako
`
`Exhibit 1013 Page 5
`
`Mako Exhibit 1013 Page 5
`
`
`
`ERIC R. KISCHELL
`Southeast of Twin Cities, MN
`keesh@ieee.org
`
`
`
`SUMMARY
`My practical experience and theoretical knowledge in image processing/segmentation/display, pattern recognition, 3D
`visualization, object oriented architecture, user interface prototyping and code optimization have been heavily utilized to
`create very complex medical and military imaging systems and plug‐ins.
`
`
`
`TECHNICAL SKILLS
`Computer Languages
`C, Visual C++, C#, Java, Assembly (MMX/SSE), Visual Basic, HTML, XAML (lite), Perl, Python for data analysis (lite), TCL
`(lite), VBScript, JavaScript (lite), UNIX Shell Scripts, SQL (lite).
`Operating/Window Systems
`MS‐Windows (WIN8.1 (lite), WIN7, Server 2008), Cygwin, LINUX (lite), UNIX, X‐Windows (lite).
`Software Tools
`MATLAB, SPM (lite), Image/J, ITK, VTK, AVS, AVW, N3, BET, NUAGES, .NET, ASP.NET, ,WPF/MVVM, MFC/Win32,
`DirectDraw, XForms, Objective Chart, Accusoft MISDK (DICOM), ActiveX (lite), Eclipse (lite), JBuilder, Visual Café,
`BoundsChecker, VTune, Purify, Insure++, NUnit, CMake, TFS, SourceSafe, CVS, ReSharper, Beyond Compare, Nuget,
`MS‐Word and Project (lite).
`
`
`
`EXPERIENCE
`Aug. 2005 Mayo Foundation, Rochester, MN
`To
`
`
`Present
`
`
` Architected and developed a Cardiac PYP Uptake WPF / MVVM application.
` Enhanced complex WPF / MVVM PACS dashboard to help resolve problems faster. Updated companion
`applications to utilize statistically superior Banker's rounding algorithm.
` Streamlined and improved absolute volume estimates of segmentation subsystem, improved surface/MIP
`volume rendering quality, and added support for new image acquisition protocols in Gastric Accommodation
`application.
` Improved stability and robustness of epileptic brain surgery planning application (SISCOM2). Corrected
`orientation‐related DICOM tags, integrated proprietary surface matching algorithm to handle difficult SPECT
`to MRI registrations, improved legacy registration algorithm (AIR) via more accurate initial translational and
`rotational guesses, and corrected defects in object segmentation pipeline in SISCOM2. This work
`considerably reduced case processing time.
` Corrected critical legacy defect in DMSA renal application in which the wrong kidney was being used to
`calculate the background region of interest (ROI) statistics. Refactored Renal Analysis application to handle
`unexpected acquisitions, protocols, and ROIs.
` Researched, designed, and developed an integrated clinical MRI‐based automated change detection system
`consisting of DICOM receiver, image format conversion, intensity non‐uniformity correction (N3), co‐
`registration (ITK), brain extraction/cleanup (SPM, AVW), change detector, GUI (visualization), and web viewer
`(ASP.NET, VBScript). Optimized system for high performance computing via 64‐bit port, multi‐core
`awareness, OpenMP, and Intel C++ compiler. Partnered with IBM to further improve system workflow and
`performance. Enhanced system with on‐the‐fly image subtraction (with window/level) feature and status
`reporting web service. Extended internal image class library to support per pixel alpha blending. This system
`detected brain defects that were missed by the radiologist.
` Detected and characterized artifacts in Ultrasound phantom images utilizing median filter, Sobel edge detector,
`Hough Transform, Convex Hull, feature extraction (SNR), and alpha‐blending algorithms. Extended MATLAB‐
`based active contour prototype to characterize (target SNR) scanner performance in Ultrasound phantom
`images.
`
`Mako Exhibit 1013 Page 6
`
`
`
`ERIC R. KISCHELL
`
`
`
`
`
`
`
`
`
`
`
`
`
` page
`
` 2
`
`ImageJ Java image processing program (Open Source) http://rsb.info.nih.gov/ij/
`http://ij‐plugins.sourceforge.net/ij‐vtk/index.html
`Various Open Source
`
` Proposed Java Advanced Imaging photo quality improvements for a United Way community card hackathon.
` Prototyped microscopic‐based complex object length analysis plug‐in.
` Contributed level of detail integration (multiresolution modeling) and screen capture to VTK‐ImageJ volume
`and surface rendering plug‐ins. Experimented with color volume rendering via ij‐VTK and TCL.
` Prototyped cycloid arc (stereological method) measurement and batch processor plugins for commercial client.
` Prototyped color image segmentation for pink particles in a sediment profile utilizing basic color space
`transformation.
` Contributed elasticized ROI, Swing‐based measurement results table, and drag and drop plug‐in.
` Developed automated time annotation and meat processor plugins to save user time and effort.
` Prototyped color image processing cell counting methods for a non‐profit application.
` Contributed UML Use Case diagrams to a Linux‐based animal shelter engine.
`
`
`Various
`
`
`
`
`Sept. 1999
`To
`
`July 2005
`
`Terason, Burlington, MA
`
`
`
` Researched and integrated image quality scan conversion interpolation routines to substantially increase image
`quality. This feature helped Terason “make a big splash” at RSNA. Improved gray scale resolution of images
`especially for high frequency probes.
` Hand‐optimized time critical code using Intel SIMD SSE/SSE2 instructions to increase speed 30‐fold.
` Field‐tested a portable, low power, PC‐based Ultrasound device to support clinical trials.
` Developed and integrated gray scale mapping "S‐Curve", calibrated gray scale display curves, gamma
`correction, histogram/statistical, brightness/contrast, M‐mode/Spectral Doppler display controls, and
`Spectral Doppler sound API.
` Designed and developed optimized cardiac time series data and complex measurement scrolling utilizing virtual
`bitmap, BitBlt, and Direct Input techniques.
` Specified, prototyped, and tested biopsy needle guide subsystem.
` Refined and integrated crude near real‐time EKG rendering prototypes from hardware group to speed product
`development.
` Refactored and extended image file input/export (DICOM, TIF), DIB, palette, printing, on‐screen visual
`components, and exam/patient presets.
` Organized hardware/software co‐design meetings to facilitate company communication.
` Researched histogram‐based “one‐touch magic button” image optimization techniques utilizing heuristic
`(greedy), Powell, and simulated annealing algorithms for SBIR phase I grant.
` Researched speckle reduction/edge enhancement filters.
` Prototyped advanced image rendering techniques (DIB sections, True‐color DirectDraw pixel compositing
`utilizing source color keying and hardware overlay plane).
` Researched Time Gain Compensation (TGC) modeling taking into account tissue attenuation/frequency/depth,
`transmit focusing action/beam shape, center frequency variation, and other transducer characteristics.
`
`
`AutoQuant Imaging, Watervliet, NY
`Dec. 1997
`
`To
`
`Aug. 1999 Managed and re‐engineered a 3D visualization Windows 98/NT‐based product to increase sales. Managed, designed,
`and assisted in development of product family reusable code libraries, visualization and GUI components. Researched
`and prototyped a multispectral crosstalk elimination project. Managed and assisted in development of cell counting
`and neuron tracing image segmentation SBIR phase I projects. Researched and integrated a point spread function
`correction scheme for widefield microscopic data to improve product image quality. Integrated Visual Basic front‐end
`code with C/C++ algorithm DLLs via Automation/ActiveX.
`
`Mako Exhibit 1013 Page 7
`
`
`
`ERIC R. KISCHELL
`
`
`
`
`
`
`
`
`
`
`
`
`
` page
`
` 3
`
`Carnegie Mellon University, Pittsburgh, PA
`Medical Robotics and Computer Assisted Surgery
`Prototyped, integrated, and field‐tested the HipNav intraoperative surgical navigational system to support clinical
`trials. Architected, developed, tested, and maintained a semi‐automatic contour extraction system including
`complementary 3D GUI tools for surface model generation of the pelvis and femur. Investigated template‐based
`learning utilizing active contours.
`
`
`July 1995
`To
`
`Dec. 1997
`
`
`Nov. 1993
`To
`
`June 1995
`
`Atlantic Aerospace Electronics Corporation, Waltham, MA
`Image Exploitation Group (secret clearance security level)
`Successfully fulfilled large DOD contract by integrating SAR image‐based change detection system consisting of clutter
`reduction/object detection algorithm, object oriented database, and GUI. Programmed, debugged, and tested target
`recognition algorithms based on multiband IR images. Formulated and developed statistical and fuzzy logic‐based
`feature analysis techniques to discriminate natural clutter objects in SAR imagery.
`
`
`
`EDUCATION
`Texas A&M University, College Station, TX
`M.S. in Electrical Engineering
`Thesis Title
`Classification of Brain Compartments and Head Injury Lesions by Neural Networks Applied to MRI.
`Courses
`Image Processing (Introductory and Advanced Special Topics), Pattern Recognition (Classical and Neural Networks),
`Signal Theory, Statistical Communication Theory, Fuzzy Logic and Intelligent Systems, Parallel Numerical Algorithms
`Projects
`MRI brain boundary detection, image noise removal/enhancement, pattern recognition military targets, character
`recognition, image compression, fuzzy control (Dodge van).
`Publications/Conferences
`Journal of Computer Assisted Microscopy. "Three‐Dimensional Thinning Algorithm…"
`Neuroradiology Journal (1995) 37:535‐541. Houston Conference Biomed. Eng. Res. (Feb. 1993).
`
`
`
`Northeastern University, Boston, MA
`B.S. in Electrical Engineering
`Additional Courses
`OpenCV (short), Kinect (short). Rochester MN .NET Developers and Friends
`
`Agile Training Boot Camp. Mayo / TEKsystems
`Microsoft Developers Conference (HTML5/Javascript for Gaming, SOA, Cross Platform Mobile Apps)
`Applied Statistics and Hypothesis Testing (short). Mayo
`Unix Shell Fundamentals and Unix Shell Advanced (short). Mayo
`Parallel Computing Toolbox and MATLAB Distributed Computing Server (short). MathWorks
`Cell Processor Architecture and Programming (short). IBM
`MATLAB Fundamentals and Programming Techniques (short). MathWorks
`DICOM Basics, IHE Intermediate Level Course (short). Siemens/internal
`Programming the .NET Framework (ASP.NET, Web Services).
`Win‐Dev (WinForms, Graphics/GDI+/Image Rendering, DirectSound/Direct3D). Boston University
`Software Development for Medical Device Manufacturers (short), Design Patterns (short). IEEE
`Medical Imaging Infrastructure (short). MIT
`Java, Data Mining (short), Web Tech. (short). Carnegie Mellon University
`User Interface Software Tools (HCI). Carnegie Mellon University
`Object‐Oriented Analysis and Design (2). Boston University and IEEE
`Prog. MS‐Windows; C I, II, C++. Relational database and flat file reformatter. Northeastern University
`
`
`
`ACTIVITIES/INTERESTS
`IEEE, Rochester MN .NET Developers and Friends, C4.
`Enjoy fine art (drawing/painting), music, indie/science fiction/horror/Mike Judge films, health and fitness.
`
`Mako Exhibit 1013 Page 8
`
`