AI-Enabled Surgical Planning Software
VasoGnosis has created a diagnostic and surgical planning software platform.
Using Artificial Intelligence, its software provides automatic detection of cerebrovascular diseases. It also enables doctors to determine the optimal course of action for treating patients.
VasoGnosis has submitted a provisional patent for its platform, and has raised funding from professional investors including Health Wildcatters and GPG Ventures. Its team of advisors includes the former Chief Financial Officer of GE Healthcare and an assistant professor at the Stanford School of Medicine.
It’s seeking capital to bring its software to market later this year through pilot programs at two hospitals.
Cerebrovascular diseases are disorders in the brain that are typically caused by bleeding. One of the most common are brain aneurysms.
A brain aneurysm is a weak, bulging area in an artery in the brain. Because its walls may be weak and thin, an aneurysm is at continuous risk of rupturing. If it ruptures, blood spills into the space between the skull and the brain, and a stroke occurs.
Right now, more than 150 million people worldwide are living with a brain aneurysm. Ruptured brain aneurysms are fatal in roughly 50% of cases. There are nearly 500,000 deaths globally each year caused by these aneurysms. And of those that survive, 66% suffer permanent neurological defects.
In the medical industry, the main problems are associated with diagnosing and treating patients with these aneurysms.
Once diagnosed, patients are placed on a watch list. They have to submit to head scans on a regular basis. And each time, a radiologist has to review all previous scans to perform a risk assessment.
If at any time the patient is deemed high risk, they’re offered the option of surgery. This presents another problem:
The diagnosis and selection process of the surgical device — there are more than 32 FDA-cleared devices on the market — used to treat the aneurysm is flawed.
For one, it’s based on the surgeon’s personal preference, not the patient’s anatomy. For another, the incorrect tool often leads to poor patient outcomes. 25% of surgeries are unsuccessful and lead to more than $65,000 in additional charges for the patient in the year following the operation.
VasoGnosis’ software-as-a-service has been designed to solve these problems. More specifically, it helps neurosurgeons and neuroradiologists make timely, data-driven decisions.
Here’s how its software works:
First, AI algorithms are trained to extract vasculature information from medical images to detect abnormalities in a patient’s scan. This detection can happen in less than 5 minutes.
Next, if an anomaly is found, the algorithm identifies the disease and shows its location. The location is revealed using three-dimensional measurements and is presented in a customizable radiology report.
Finally, the software performs a detailed analysis that determines the necessity of surgery. If a surgeon verifies that a procedure is required, VasoGnosis’ surgical planning tool enables them to rehearse the operation via a blood flow simulator. This enables them to select the proper tool and evaluate patient outcomes.
This system is unlike anything on the market. One competitor, Sim&Cure, provides surgical planning for brain aneurysms, but doesn’t have detection capabilities or AI evaluation tools. Another competitor, HeartFlow, has similar software, but focuses on cardiovascular diseases. In 2018, HeartFlow was valued at $1.3 billion.
VasoGnosis’ business model is a combination of subscription plans and pay-per-use.
More specifically, the company charges hospitals an annual fee for unlimited use of its detection software. This fee is $50,000, on average. In addition, it charges a $2,000 fee for each surgical planning case it requires.
VasoGnosis released a beta version of its software in May 2020. It will soon begin two pilot programs at the Medical College of Wisconsin and the UMass Memorial hospital system.
In the meantime, the company has been accepted into the NVIDIA Inception program. This accelerator focuses on guiding medical technology startups. One alum, Subtle Medical, graduated in 2017 and was named to CB Insights’ 2020 “AI 100 List of Most Innovative AI Startups.”
Prior to starting VasoGnosis, Ali was a postdoc researcher at the neurosurgery department of Medical College of Wisconsin. He specialized in performing research on brain aneurysms.
Before his time in Wisconsin, he was a research assistant at the University of Iceland. He began his career by starting a pair of IT-related companies.
Ali earned a Master’s degree in Mechanical Engineering from the University of Iceland and a Ph.D. in Biomechanics from the University of Wisconsin.
Sundeep is a clinical and research expert in interventional neuroradiology and endovascular neurosurgery. He is focused on advancing the diagnosis and treatment of complex cerebrovascular diseases.
After completing his training at the University of Southern California, Yale, and University of Iowa, he served as a faculty member at several schools including Columbia, Cornell, and NYU.
Notably, he spent 16 years at the SUNY Downstate Medical Center, and then spent 10 years in the radiology and neurosurgery department at Lincoln Medical and Mental Health Center. During this time, he performed several thousand neuromuscular interventions for brain aneurysms.
In addition to his clinical experience, Sundeep has an extensive record of peer-reviewed publications and international presentations on neuroimaging, stroke intervention, and aneurysm therapy. He also has spent time completing biomedical diagnostic tool development and testing.
Alexa is an engineer with experience in software, hardware, fluid dynamics, and rapid physical prototyping.
Before starting VasoGnosis, she spent 10 years coding and completing data analysis. This came after she received her degree in mechanical engineering from the University of Wisconsin.
Pedram has an extensive background in engineering and computer science. Throughout his career, he’s won awards for his machine learning skills and knowledge on multi-sensor data fusion.
Most recently, he served as head of the machine learning group at Helmholtz-Zentrum Dresden-Rossendorf (HZDR), a Germany-based research company. Prior to that, he was co-chairman of the IEEE Image Analysis and Data Fusion Committee.
Earlier in his career, Pedram was an associate editor for Remote Sensing, a technology company, and began as a research scientist at the German Aerospace Center.
He earned a Ph.D. in Electrical and Computer Engineering from the University of Iceland.
A VC firm investing in early-stage companies. Portfolio includes Savara Pharmaceuticals, Brainspace, and Seimler Scientific.
An accelerator focused on improving healthcare by supporting medical and healthcare-related startups.