Technology

Breakthrough Medical Technology

Imaging3 has developed a breakthrough medical technology (patented) that will produce 3D medical diagnostic images in Single 3D SafeScan and Continuous 3D Scan mode.

Single 3D SafeScan

In the late 19th century, when xray technology was newly discovered, there was a debate over whether or not fluoroscopic moving images would replace radiographic still images. Due to the single dose of radiation that one-time photos required compared to the ongoing dose that motion pictures needed freeze frames won the controversy and became the standard of the industry. In the early 21st century the two concepts no longer need to be mutually exclusive. Motion pictures in 2D from every angle can be captured near instantly and viewed individually or transformed into 3D images that can be moved and manipulated. In Single 3D SafeScan mode as the generator/scanner orbits the xray shutter is opened briefly then promptly closed to collect 2D images from all angles. The set of images is then turned into a 3D model.
The conversion from 2D images to a 3D model occurs in a fraction of a second on a super computer. The Azure cloud will be used to host transform software just as soon as Microsoft releases the promised N-Series virtual machine product. Cloud hosting allows use of software on a periodic subscription basis not local copy which can easily be pirated and ensures that the hardware and software available to customers is up-to-date and modern. The 3D models generated by N-series virtual machines in the cloud, once downloaded, can be viewed and manipulated with roll/pitch/yaw commands and sliced on regular machines using regular video cards ranging in price from $500 to $5000. No need to purchase a fancy $50,000 super computer that will become obsolete within two years.
If the 3D scanner technology is coupled with a 3D printer to create a 3D photocopier combination the conversion from 2D images to 3D printout will be an option. This opens up the possibility of an entirely new debate contest. Whether or not physical 3D printouts will replace virtual 3D models viewed and manipulated on computer screens?
The concept of the Single 3D SafeScan originates from the way the Tropospheric Emission Spectrometer (TES) instrument operates on the Aura Earth Observing Satellite. Shutters on the TES instrument open and close for several seconds at a time in order to perform data acquisition from far above. The engineer who developed the DVIS onboard and ground data system and optimized the image processing and transform algorithms suggested to Dane Medley that the shutter on the DVIS xray generator functions the same way as the shutter on the spacecraft and could be opened then closed for several seconds at a time in order to acquire the data. If the subject does not change or move then the extremely low dose of radiation involved with this approach would be preferable and beneficial. Dane came up with the name SafeScan and this product then came about.

Continuous 3D Scan

The DVIS spins at an extremely high velocity, similar to a washing machine as it dries clothes only much faster. Think of it like a LIDAR system on top of an autonomous vehicle which rotates super fast. LIDAR like the DVIS has a light generator, but it emits lasers, not xrays, in the outward direction. Instead of performing data acquisition in the outward direction like the LIDAR the DVIS performs data acquisition in the inward direction. The DVIS software is based on Earth Observing Satellite (EOS) technology principles with three key differences. 1) EOS instruments study above surface features and do not penetrate the object being observed (the earth). 2) EOS orbits last approximately 100 minutes in contrast to DVIS or LIDAR orbits which take place in less than a second. 3) It takes JPL/NASA hours to process the data. For example MISR data products are available 3 hours after data acquisition. DVIS level 3 (visualization data) must be available within a fraction of a second. Xray generators and scanners mounted on opposite sides of a gantry allow for data acquisition in the inward direction. Frame rates of up to 60 frames per second are possible using COTS (common of the shelf) medical scanner equipment. This setup allows for 3D refreshes which occur as the 3D model is being manipulated by roll, pitch, yaw, zoom, and slice navigation commands.
The 2D to 3D transform with refresh using obsolete 2009 Tesla NVidia cuda technology takes place in 0.25 seconds in compact mode and 0.9 seconds in full resolution mode. No doubt with modern up-to-date hardware and software this speed could be improved. When Dean Janes, the inventor of real time 3D imaging, was forced to leave the company in early 2014 he admitted his mistake. He always expected the computer hardware and software to be the bottleneck. It turned out that computer processing can do its job faster than the data acquisition.

Radiation

There are two things that Imaging3 does better than any other manufacturer on this planet earth. Nobody produces a 3D model faster than Imaging3. Not even close! Nobody produces a 3D model with less radiation than Imaging3. Not even close! The catch phrase: "Choose Imaging3 because radiation is bad for you" holds true and awaits FDA submit and approval so that consumers can make a choice. Just how little radiation is needed to generate a 3D model?
The International Atomic Energy Agency (IAEA) in Vienna claims that fluoroscopy doses range between 0.08 and 0.15 mSv per minute. See section #12 link here. The mid range value 0.115 mSv per minute divided by 60 seconds is 0.002 mSv per second. This is a rough approximation and the amount can be lower or higher depending on subject matter. For example to view a piece of corn requires a different dose than a human skull.
The Radiological Society of North America (RSNA) and American College of Radiology (ACR) sponsored website claims that a dental xray exposure is 0.005 mSv and is equivalent to one day of natural background radiation. See table link here.
Since xrays penetrate an object only 180 degrees is needed to produce a 3D image. The image at 0-179 degrees is the same exact image as 180-359 degrees for example. With only a single xray generator/scanner combination in order to perform a data acquisiton of 180 degrees with 1 degree granularity at 30 frames per second requires 6 seconds + 1 second for padding = 7 seconds total. The Single 3D SafeScan images on this website were all created with the DVIS version 1 proof of concept prototype which consists of 1 xray generator/scanner pair. Using several combinations of synchronized xray generator/scanner pairs spinning much faster and acquiring images at 60 frames per second can reduce the radiation tally even further. Referencing the IAEA fluoroscopy dose of 0.002 mSv per second as a guide and assuming that an xray shutter stays open for 7 seconds exposes an object to 0.014 mSv of radiation.
In summary the DVIS version 1 proof of concept prototype exposes an object to roughly 0.014 mSv in other words 3 dental xrays or about 3 days of natural background radiation in order to produce the quality of 3D image one sees in the Single 3D SafeScan galleries on the yellow sidebar menu.

Safe 3D Images with Radiation Concept

One can move a finger through a candle rapidly and not get burned. The damage and pain only takes place if the finger is kept in place for a longer period of time. The processing speed and algorithms developed for real-time imaging can be applied to a one time scan to produce a 3D image of the human body with as little harm as possible.

Safety

Conventional fluoroscopy patient exposure time is usually measured in minutes. Single 3D SafeScan time is measured in seconds. The current model requires 6 seconds of exposure to acquire 180 images. An onboard computer runs an independent service which monitors the xray. If at any time for any reason the xray stays on for longer than 7 seconds the monitor turns the xray off. This xray monitor runs continuously on the onboard computer and stays active for as long as the system is on.
The Los Angeles County Health Department visited Imaging3. They brought their instruments to measure radation and any other possible effects emitted by the DVIS. They recommended that visitors stay at least six feet away from the xray source, in other words three feet away from the DVIS since the gantry is three feet in diameter. This was prescribed in an abundance of caution since a small dose of occasional scatter radiation should reach no further than this far. No other mitigation measures were suggested or requested. Special reflective striped caution tape on the floor indicates this perimeter. Visitors during demos do not approach closer than this distance to the unit.

FDA Concerns

There was an issue with image quality and fire hazard during the early submission process. The image quality of the DVIS is no better or worse or different than a standard conventional C-Arm image for a very simple to understand reason: the DVIS has a COTS (common of the shelf) C-Arm x-ray generator and scanner mounted on it. A special consultant familiar with medical protocol will be employed to prepare the images submitted to the FDA in the requested DICOM format. The hardware certification should address any fire concerns. In addition adequate and appropriate documentation regarding usage will be provided. The C-Arm documentation warning regarding leaving a towel or rag on the unit by a nurse or doctor will not apply to the DVIS version which is being considered to submit to the FDA because the Single Scan form will have several monitors, hardware and software, to ensure that the xray does not stay on longer than 7 seconds.
Part of the submission process to the FDA is CriTech certification of the software. The firm also prepares documentation and design and technical diagrams. Imaging3 is taking a piecemeal approach because the Single Scan and Continuous Scan products use common shared components. Below is a diagram of what was covered during the CriTech testing process. If/after the Single Scan version passes FDA approval the Continuous Scan version will soon follow and use most of the same components. The key difference is the output: 180 2D images created by the Single Scan or a steady stream of 2D images created by the Continous Scan and broadcast over the network.
The 2D to 3D transform and visualization/manipulation portion is not part of the scope of this product because it occurs on a machine or machines on a dedicated intranet network or N-series virtual machines on the Azure cloud network. Once the core hardware device and software to drive it is approved by the FDA money will be spent to have CriTech certify this portion of the software and it will be submitted to the FDA for clearance.

The Right Stuff

During the cold war Burbank was the home of Lockheed's skunk works. This small shop produced top of the line fighter jets for the US military. All that remains of this relic are the Lockheed airplane figures which decorate the local shopping center. The new Burbank skunk works consists of companies like E-Solar and Imaging3. These companies' technologies revolve around a branch of physics which involves trigonometry and light energy. E-Solar transforms light energy from two dimensions (a plane) into light energy in zero dimensions (a point). Their Crescent Dunes project in Nevada powers 100,000 homes 24x7. Their SandStone project in the works will generate 1 gigawatt, as much as a nuclear plant, and power 1,000,000 homes. Imaging3 transforms light energy measurements from two dimensions (a plane) into light energy measurements in three dimensions (a volume). At Disney down the street the Hyperion team models how light scatters in complex settings in real life in three dimensions.

Frequently Asked Questions

  • How does the DVIS differ from a CT Scan? The CT was designed and developed in the 1970s and 80s. It transforms light energy measurements in one dimension (1D = a line) into light energy measurements in three dimensions (3D = a volume). This line is actually a narrow beam of light. The CT uses conventional xray radiation. A CT rotates around an object and moves foward one line at a time like a screw thread. The DVIS transforms light energy measurements in two dimensions (2D = a plane) into light energy measurements in three dimensions (3D = a volume). This two dimensional plane is actually a photo. The photo is acquired by means of a special video camera known as a scanner which operates at 30 or 60 frames per second. The DVIS uses fluoroscopy xray radiation. This much more efficient transform, DVIS 2D to 3D vs CT 1D to 3D, is made possible with the help of a massively parallel super computer which was not available when CT was coming of age.
  • How does the Single 3D SafeScan product differ from the Continuous 3D Scan product? The hardware component of the SafeScan product produces 180 two dimensional images. The hardware component of the Continouous Scan product broadcasts a steady constant stream of 2D images over a dedicated internal network channel. The software component for the two products uses the same common shared 2D to 3D transform algorithm. Most other aspects of software component of the two products contrast and the details are still being worked out and await FDA approval of the hardware component.
  • What happens if a patient moves while the Single 3D SafeScan is acquiring data? The current version of this product requires that a patient stay still for 6 seconds. The follow on version of this product will do the job in 1 second. The follow on version of the product will also expose the patient to the equivalent of only 1 dental xray of radiation. As the DVIS orbits at a higher velocity and the scanner operates at a higher frames per second rate the consequences of fidgeting diminish and the patient is exposed to a lower amount of radiation.
  • Does the fast spinning of the DVIS make the pictures blurry? Light travels at a speed of 300,000 kilometers per second. The lateral motion of the DVIS is negligible compared to this speed. Also keep in mind that relative to the xray generator the scanner does not move as the DVIS rotates.
  • If adding xray generator/scanner combinations makes for better image quality why not just add more? The Imaging3 patent covers adding more xray generator/scanner pairs. The problem with adding too many more combinations is that light is both a particle and a wave. Using an analogy it would be like an intersection with green lights for all cars. Particle collisions and crashes occur which scatters the light. The sources of light need to be synchronized like traffic signals on our street corners. Disney's Hyperion project harnesses and controls the chaos of light scattering in a virtual environment for the purpose of motion picture animation. In real life light scattering due to photons crossing paths is a serious problem that is not feasible to overcome.
  • Will running in Continuous 3D Scan mode expose the patient to too much radiation? First off, the first version of the Continuous Scan product will perform two 3D refreshes per second. This rate requires no more and no less radiation exposure than a typical conventional C-Arm which is used today. Secondly, 3D viewing capability might allow an operator to perform a procedure much quicker than using the current limited C-Arm 2D view. Thirdly, Continous Scan can be used for applications other than medicine. It is a 3D vision system that can be used to guide end effectors for robotics. If it is used to harvest, process, or prepare food for example the task could be finished in several seconds.
  • Have you looked into using Virtual Reality (VR) to display and manipulate 3D models? Yes, we have considered VR and came to the conclusion that it is great for viewing but deficit at navigation. For now Roll/Pitch/Yaw and Zoom buttons are used for demonstrating DVIS capabilities until a better way to navigate 3D models appears on the market.
  • Which technology is on the Imaging3 wish list? 16 bit per color monitors. The human eye cannot discern more than 8 bits = 256 per RGBO (Red, Green, Blue, Opaque) color. The xray world operates using 16 bit grayscale values, not 8 bit RGBO values. What ends up happening is radiology applications destroy or dumb down half the image quality in order to fit in the 8 bit per RGBO color space of commercially available monitors. Expensive 10 bit per color monitors are available but the data stills needs to be reduced to fit.
  • What are the limits of the DVIS? If the goal is to use inexpensive common off the shelf (COTS) components then 60 frames per second scanners are the bottleneck to acquiring data. Special custom video cameras that capture visible light are capable of acquiring images at a rate of 20,000 frames per second. If cost is no object then such a camera can be converted to work with fluoroscopy xrays and mounted on a DVIS. The deciding factor to the limits of the DVIS with this setup would be the rate of spin rather than frames per second rate. The new paradigm is that the amount of radiation exposure needed to produce a single manipulatable 3D model depends solely on the period of the orbit. For example if the DVIS rotates at 3 revolutions per second this would allow for Single 3D Scan to do its job with the equivalent of 2 hours of exposure to natural background radiation that we all encounter in every day life and the Continuous 3D Scan to perform twelve 3D refreshes per second. How fast can the DVIS spin? For every orbit per second increase the amount of radiation needed to produce a 3D model decreases. At a ridiculously high rate the amount would be equivalent to one second per second, in other words no radiation. Thus the DVIS, in theory, may be able to produce 3D models with no xray radiation.