About our project
Reconcilio is the first technology to provide quality assurance in cancer treatment. Our software offers automated consensus contouring of tumor images, which increases the safety, accuracy and efficiency of radiation treatments.
Essentially, Reconcilio creates an outline around the tumor that is the “best-fit” or average of the other outlines, ensuring that all the tumor is included and that the least amount of healthy tissue is included.
Killing cancer cells via radiation therapy is a common treatment method. The first key step requires the clinician (normally an oncologist) to analyze images of the tumor and to then outline the tumor and its surrounding tissue which will then undergo radiation treatments.
These markings or outlines, called contours, are crucial for efficient treatment. The more accurate the contours, the higher the success rate of killing cancer cells in a shorter time. Inaccurate contouring can lead to partial tumor radiation and damage to the surrounding healthy tissues.
 |
| Figure 1. Inaccurate Contouring |
Currently, there is no quality assurance for contouring (or outlining the edges of tumors). Cancer patients have to rely on the knowledge and contouring skills of their doctor. Unfortunately, it is well documented that clinicians, given the same image, generally produce different contours (outlines). Depending on the tumor type, size and location, clinical studies have shown that clinicians’ contours deviate considerably from each other ranging from 5% up to a frightening 60% difference!
 |
| Figure 2. Inter-observer Variability. Different clinical experts contour (outline) the same image differently. |
This long-standing problem is called “inter-observer variability” and results in suboptimal patient care and a more costly health care system.
How it Works
So, what is the solution? Ideally, we would have several doctors contour (outline) the tumor images of one patient. From this, a “consensus contour” that reflects the collective knowledge of all the doctors is created and then serves as a benchmark for accurate measurements. This approach can drastically reduce the variability and assure greater accuracy in treatment.
But no healthcare system can afford a multi-doctor approach, which is why we designed Reconcillio.
Our software generates multiple contours (blue) based on Knowledge Maps (Atlases) of many experts who use the software to build a consensus contour (red).
 |
| Figure 3. Our software can automatically generate consensus contours (red) |
Taking this a step further, our intention is to put this technology on the “cloud” in order to extend the software to a large number of doctors and create atlases for many cancer cases. With the press of a button, any doctor will be able to request a “consensus” to get assurance that his/her contours are correct.
Clinical Proof
We have already developed the core engine of our technology, designed and developed a contouring software package for prostate cancer, validated it with help of seven oncologists, and submitted it to the FDA. We have received FDA Clearance, 510(k), for this first product. A validation report has also been published in The Red Journal, one of the most prominent oncology journals. This laid the foundation for us to have a FDA “predicate” for our main product “Reconcillio”, which we want to validate in this project.
How We Help Patients
Quality assurance based on consensus contouring has multiple benefits for cancer patients:
- Shorter treatment time (because tumors are marked more accurately)
- Reduced side effects (because adjacent healthy tissue can be spared more confidently)
- Higher likelihood of successful radiation therapy
How We Help Doctors
Automated quality assurance reduces the probability of human error and can increase both patient satisfaction and service provider ranking. Automated quality assurance maximizes efficiency, allowing doctors to spend more time on high-quality patient care and research. The use of this software can also be leveraged when negotiating malpractice insurance premiums as consultations have the potential to increase efficiency and quality and thus, minimize risk.
How We Help Hospitals, Institutions, and/ or Medical Facilities
Cancer incidence is on the rise and predicted to increase by 22% by 2020. Conversely, the number of oncologists will only increase by 2% (Benjamin et al., 2010). This is a problem for both patients and healthcare providers. Auto QA will enable more efficient treatments and shorter patient care — this can reduce provider costs, increase profit, and quality of care measurements.
Benjamin et al., The Future of Radiation Oncology in the United States From 2010 to 2020: Will Supply Keep Pace With Demand? American Society of Clinical Oncology, October 18, 2010
We have already been in contact with multiple hospitals and research institutions such as Sunnybrook Hospital (Toronto, ON), Princess Margaret Hospital (Toronto, ON), London Health Sciences Centre (London, ON), National Cancer Institute (Bethesda, MD), Masonic Cancer Centre (Minneapolis, MN), & the Windsor Regional Cancer Centre (Windsor, ON).
Many clinical experts have expressed their willingness to support our cloud consensus project and we would love for you to join them and us in augmenting the quality of treatment and care for cancer patients everywhere.
How We Help Our Partners
There are many medical applications that could benefit from consensus contouring, either for quality assurance or for fast batch processing. Partnering companies can integrate the technology into their existing solutions to offer more value to hospitals.
How We Will Spend Your Contributions
We are asking for 50,000 in this first round of funding to leverage as evidence of our product’s credibility, traction, and success when presenting it to various VCs for funding. The actual amount needed to complete all our goals is $5 million, but we believe that your and MedStartr’s support will be crucial in reaching that goal.
With your funding, we can move towards extending this service to doctors everywhere. Specifically, the money will be used to customize the technology for cloud services and for the resources needed to launch a cloud portal for the validation process.
Who We Are
Hamid Tizhoosh (CEO) has 16 years of experience in medical image analysis: LinkedIn
Sam Visaisouk (President) has an extensive background in commercialization: LinkedIn
Andrew Wong (CFO) is an expert in financial matters: LinkedIn
Farzad Khalvati (CTO) has developed the first prototype of this technology: LinkedIn
How to Find Us
www.segasist.com
twitter.com/segasist
-
Have a question?
If the info above does not help, you can ask the project creator directly.
-
-
){kind=link}
){kind=link}