The Intersection of Big Data and Interactive Mixed Reality Ronechopixeltechcom Sergioechopixeltechcom CARS 62620 Its All About Intuition 1 DICOM amp IHE CARS 62620 2 DICOM amp IHE Surgery ID: 930759
Download Presentation The PPT/PDF document "The Knowledge Model Based on" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
The Knowledge ModelBased onThe Intersection of Big Data and Interactive Mixed Reality
Ron@echopixeltech.comSergio@echopixeltech.com
CARS - 6-26-20
It’s All About Intuition
1
Slide2DICOM & IHE
CARS - 6-26-20
2
Slide3DICOM & IHE Surgery:
Tasks
Partnership of Engineering and Medicine
To facilitate and promote purpose-driven design, modeling, and architecture in the operating room ;
through the application of ergonomics, engineering, and information technology
achieving consolidation and coordination of components.
To ensure communication, modularity, efficiency, visualization, knowledge and decision management, safety, and
IMPROVED OUTCOMES THROUGH INTEROPERABILITY
for optimal performance of the specific tasks of any given surgical specialty.
CARS - 6-26-20
3
Slide4It’s All About IntuitionEinstein –
“There comes a leap in consciousness , call it intuition – a truly valuable thing.”Jobs – “Have the courage to follow your heart and intuition – a very powerful thing, more powerful than intellect.”
CARS - 6-26-20
4
Slide5Present Concerns
Challenges within radiology (e.g. undetectable objects) negatively impact patient outcomes.Surgery demands visualization with an intuitive focus and ease of protocol development.Interoperability/Enterprise requires excellent communication – no gaps.
Goal –
increased enterprise and patient outcomes.A Solution – Interactive Mixed Reality – A Direct Path to Intuition
CARS - 6-26-20
5
Slide6Thought Leader – ”Can’t See It”
Meaningful data and insights are embedded within medical images.Often undetectable via routine visual analysis – 2D/2.5D* views.
Therefore, valuable information is being overlooked.
(Gillies, Kinahan, Hricak
, 2016; Lee, 2017
2.5D = 2D views of 3D volume data sets
CARS - 6-26-206
Slide7Thought Leader – “Can’t Digest It”
Growing complexity and volume of imaging radiology data. Exceeding the individual radiologist’s ability to make fully informed decisions.
(Thrall et al., 2016)
CARS - 6-26-20
7
Slide8Thought Leader – Is 3D the Answer?
A diagnosis of pathology requires consideration for 3-dimensional resolution.An approach not offered by current diagnostic methods.
(
Ghassemi et al., Kaiser, 2016).
CARS - 6-26-20
8
Slide9Thought Leader – For Brain, 3D is the Answer!
“It is no longer adequate to analyze complex brain and skull-base tumors with 2D views.”
“The future of neuroradiology is in advanced techniques such as virtual reality and augmented reality, which produce a dynamic, interactive 3D view of the patient’s imaging.”
DYNAMIC & INTERACTIVE = INTUITION GENERATION
(Kumar, MD Anderson, RSNA 2018 Highlights)
CARS - 6-26-20
9
Slide10Thought Leader – For Spine, 3D is the Answer!
Scoliosis is a 3D problem. IMR could be utilized for the surgeon to accurately position the placement of rods to achieve the best efficacy for treatment – e.g. addressing the twist in the spine.
IMR technology takes treatment from “best guess” to “I know intuitively I am correct.” IMR is then used by spine, gastro & pulmonary experts to confirm the ideal solution.
Ross & Burnett, AMB Surgical
CARS - 6-26-20
10
Slide11Thought Leader – Surgery Requirement
“Interactive Mixed Reality allows for examination of every layer of a patient’s anatomy.”“IMR facilitates a virtual run-through of the surgical procedure from inside the OR.”Chan,
Frandics, Stanford case study, 2018
CARS - 6-26-20
11
Slide12Radiology Agrees to Shift Focus
Radiologists need to shift focus to outcomes at the enterprise level and use the radiology environment and enterprise imaging as tools to affect the enterprise outcome.Radiology has been the dominant imaging player in the enterprise
. As holographic 3D imaging becomes available, surgery, oncology, and other disciplines are beginning to directly utilize imaging in “open 3D space.”Dr. Cheryl
Petersilge, UPMC (CCF) Sept, 2019
CARS - 6-26-20
12
Slide13Radiology Intuition - Lenny Berliner
"Interactive Mixed Reality Provides a Direct Path to Intuition!The viewing of 2.5D images creates an impediment to Intuition! Radiologists waste time and effort processing and converting 2.5D to 3D; structures have to be thought out; the spontaneity of thought that allows for intuition is impacted; thus
creating roadblocks to Intuition!IMR with presents the 3D world directly to the brain and the mind immediately responds freely with intuition and insights.
Thus, IMR provides a direct path to Intuition!"
CARS - 6-26-20
13
Slide14Surgery Intuition – Hamm & Abernathy
Intuition is as valuable & important as analytical thinking – both analysis and intuition are necessary. Experts possess a developed repertoire of SCRIPTS (Protocols) for use in clinical situations.
Scripts are both complex & flexible – as opposed to algorithms.
CARS - 6-26-20
14
Hamm & Abernathy, Harvard
Slide15The 3D Printing Question?
Bardo – “Seeing specific angles or views, or separating out anatomy, can be done virtually on the fly. I believe augmented virtual reality will become a dominant technique.”Tappa
– Surgeons interest: “3D augmented reality taking over 3D print modeling for surgical planning.”Author – “Perhaps 3D Printing and Interactive Mixed Reality can serve complementary roles in the over arching goal of improving Interoperability across Clinical Care Pathways.”
CARS - 6-26-20
15
Slide16CARS - 6-26-20
16
The Knowledge Model
Cognition - Cognitive Computing Elements (CCE); Machine Intelligence (MI); BD/AI; “Doing Things Right.” Think – vertical axis.
Intuition – Einstein/Jobs - more powerful than Cognition; Human Intelligence (HI);
Berliner – IMR direct path; “Doing the Right Things (1
st).” Think – horizontal axis. The Key: Cognition/Intuition; drives Clinical Efficacy (CE) + Work Flow (WF) = Patient Outcomes. IBM/Watson – CCE drives “How you behave,” & “What you do.” – Doing Things Right - Best Practices. Protocols – Clinical/Technical tie – driven by HI – “Intuition Engine” – provides requiredstructure to minimize Cognitive Bias and maximize Intuitive Process.
Slide17IMR Optimizes Knowledge CARS - 6-26-20
Knowledge = Cognition + Intuition
Cognition
(BIG DATA DRIVEN)
Current
3D Medical Visualization
(2D/2.5D)Intuition(INTERACTIVE MIXED REALITY DRIVEN)IMR - True 3D
(A new way)
IMR (2D/2.5D)
(A better way)17
Slide18Watson – A Cognitive Computing Element
Collaboration of man and machine – machine intelligence. Watson learns from “how you behave” and “what you do” Big Data has the potential to decrease “cognitive bias”
NOTE: Decreasing cognitive bias can increase false positives
CARS - 6-26-20
Ginny
Rometti
, CEO IBM18
Slide19The Key Ingredient – the Protocol
The Clinical/Technical Tie – Physician/Scientist – deep understanding using human intelligenceThinking “outside the box” – driven by Intuition.Dreaming in open 3D space, creating together.
For Virtual Colonoscopy – dividing the colon into linear sections, studying one section at a time to find polyps = a new way of thinking!
These results “CAN’T” be created by studying an “infinite” array of 2D/2.5D images.
CARS - 6-26-20
19
Slide20Protocol Example – Virtual ColonoscopyCARS - 6-26-20
Real Time Interactive Virtual Reality
Work with patient specific anatomy in an open 3D space
Effortless Interpretation
Required visual context, with no extraneous information
Optimal Image Strategy
Procedure specific representation of image dataSharing of Patient Specific 3D ReportsWork with patient specific anatomy in an open 3D spaceEngaging User InterfaceDirectly reach into data and trigger 3D quantification algorithms
20
AI
AIAI
Slide21Virtual Colonoscopy – Increasing CE & WF = PO
Optical Colonoscopy
2D Virtual Colonoscopy
IVR Virtual Colonoscopy
Image Type
VideoCTCTDetection Can miss up to 40% of cancers in right side of colon (most are flat lesions)Flat lesions between 20-80% Flat lesions approximates 100%* UCSF trialInterpretation Time45 – 90 minutes 30 – 40 minutes5 – 10 minutes*UCSF trialPreparation Laxative + Anesthesia Laxative and non laxative preparation – No AnesthesiaLaxative and non laxative preparation – No AnesthesiaCost $2,500 U.S. Average$500 - $750 U.S. Average with reimbursement$500 - $750 U.S. Average with reimbursementCARS - 6-26-20
21
Slide22Helen Devos Children's Hospital Surgery Example
PA – RV Conduit obstructed by sternumNo clarity as to how to expand space for relieving compression of PA-RV conduit
Treatment denied to patient in several sites
CARS - 6-26-20
22
Slide23Helen Devos Children's Hospital Surgery Example
IMR enabled surgical view of sternum and PAIMR surgical view provided verification of needed 3D space for new conduit path (see yellow path)
Treatment and surgical outcome was excellent
CARS - 6-26-20
23
Slide24Protocol Example – Left Atrial Appendage
Introduction – patients with Atrial Fibrillation can be seriously harmed by clots coming out of the LAA. The Watchman implant device from Boston Scientific blocks the opening to the LAA. The goal is to accurately size the implant for breadth and depth of placement.
Solution – get to the face of the Ostium and measure for implant size. With 2.5D views it takes at least 20 min. to search for the optimal 2.5D view. With Interactive Mixed Reality it takes approximately 1.5 min – no guessing – reach the target and click. The Landing Zone is equally simple to determine.
Intuition – for Virtual Colonoscopy, intuition was key for developing the protocol process. For the LAA intuition plays a key role in leveling the playing field in dealing with patients of various sizes and shapes – Interactive Mixed Reality provides the intuitive capability.
CARS - 6-26-20
24
Slide25A PLAN – The Intersection of BD & IMR
STARTING POINT –Interactive Mixed Reality – drives the Intuitive element of KnowledgeBig Data – drives the Cognitive element of Knowledge –
reducing cognitive biasPOTENTIAL FOR A BIG WIN -1. IMR –
initial protocol for increasing Patient Outcomes (Doing the Right Things)2. BD – using 1, provides Best Practices to optimize the IMR Protocol (Doing Things Right)
CARS - 6-26-20
25
Slide26Market Dynamics – Surgeon
Surgeon – “I never opened a patient and saw a 2D/2.5D view.” Immediately recognizes the key benefit of IMR.
Case: could not adequately visualize due to a complex surgery – IMR made the difference. Will utilize IMR for surgical planning. Need
: improvement of daily workflow and resolution of complex cases.
CARS - 6-26-20
26
Slide27Market Dynamics - Radiologist
Radiologist – “With IMR I can speak intelligently with a surgeon.” Today, providing surgical plans based on 2.5D views – incomplete – below threshold in many cases. Need:
to step up to IMR and “own” the surgical planning game. Result – Bridging the Radiology Surgery Gap - BRSG
CARS - 6-26-20
27
Slide28Interoperability Driven Enterprise
Enterprise has been incomplete due to the lack of adequate processes in Interoperability.Interoperability has been incomplete due to the lack of “bridging protocols” between different departments, e.g. Radiology and Surgery.Interactive Mixed Reality is an approach to Bridge the Radiology-Surgery Gap and demonstrate a process for creating “bridging protocols.”
CARS - 6-26-20
28
Slide29AI Constraints – Leonard D’Avolio, CEO CYFT
BD/ML is not a sentient being – not able to perceive & feel things.
AI must be embedded into the complex workflows of healthcare, to significantly create value.
Tech is the enabler; people and process improve care
CARS - 6-26-20
29
Slide30AI – Ray Kurzweil – 12/22/19
Ray started thinking about AI approximately 60 years ago.25 years later, Multi Layer Neural Nets (MLNN) came into being for reinforcing Neural Nets (too many errors). MLNN still wasn’t adequate – need for 10 to the 14th computations per sec.
Neural Folding in conjunction with Machine Based Learning did a much better job of meeting the needs.Today, AI is still not adequate for Language Based Transactions, e.g. Visual Language.
In Bridging the Radiology-Surgery Gap the IMR solution bridges 2D/2.5D with 3D.HUMAN LEVEL AI IS ABOUT A DECADE AWAY
CARS - 6-26-20
30
Slide31Optimizing Intuition, Interoperability & Patient Outcomes
(1) Accuracy of Head Tracking – critical for looking around objects(2) Latency – critical for real-time movements when interacting with anatomical objects
(3) Difficulty of Rendering Blacks – impact on diagnostic contrast(4) Accuracy of Head Mounted Camera – leads to errors in visualization & interpretation(5) Ultrasound Guidance – requires 3D Echo in open 3D space
(6) Co-Registration – more accurate in open 3D space(7) Ability to Reach for and Interact with Objects –
CARS - 6-26-20
31
Ref: Elliott Siegel - 2020
Slide32Radiomics – Dr. Paul Chang
Requirement for Improved PRECISION, QUALITY & IMPACT of Radiological Services.Morphology to Functional, Physiologic, Molecular Imaging.Qualitative to Quantitative Imaging.
Generic Characterization to More Specific and Actionable PhenoTypic Characterization.Isolated Image Interpretation to Multidisciplinary Synthesis and Relevant Actionable Information.
Need to Improve Existing Suboptimal Human Machine Workflow Collaboration to Make Precision Radiology Feasible.
CARS - 6-26-20
32
Ref: Paul Chang - 2020
Slide33ContactThank you!
ron@echopixeltech.com CARS - 6-26-20
33