Predictive analytics tools come in many forms but two aspiring teams of entrepreneurs at a StartUp Weekend for healthcare in Philadelphia see its usefulness for two applications — reducing biopsies and helping people assess the likelihood of getting a genetic disorder.
A group of prospective entrepreneurs and Penn Medicine research scientist put their heads together in a 48 hour session to sketch out a platform that would use an algorithm and ultrasound images to predict the risk of breast cancer. Team AlgoRhythm Diagnostics explained that when an ultrasound image to check for breast cancer is submitted for a second opinion, studies show that it increases the accuracy for determining whether the mass is malignant. Instead of a doctor, it is relying on an algorithm to make that calculation. It would assess each image and search for certain characteristics and also include other calculations like age. They had formed a collaboration over the weekend with Chandra Sehgal, a professor and director of ultrasound research at University of Pennsylvania.
The idea is to reduce the number of unnecessary biopsies, which would also bring down healthcare costs.
When Investment Rhymes with Canada
Canada has a proud history of achievement in the areas of science and technology, and the field of biomanufacturing and life sciences is no exception.
“It is a form of machine learning,” said Smeet Bhimani, who pitched the team’s idea. “Computer-aided detection is already being used for mammography. But taking it a step further and doing this for ultrasound would produce more accurate results because there are things you can pick up with ultrasound that you can’t pick up with mammograms.”
The team walked away with the top prize in the competition — an interview with DreamIt Health for participation in their Philadelphia or Baltimore program.
Another entrepreneur is interested in working with physicians to beta test an idea for a risk assessment tool. He’s developing a predictive analytics tool to review patient’s family history to determine their risk of developing a genetic disorder.
Josh Kleinjan of SEER Diagnostics explained that patients would access their electronic health records through a portal and allow the records to be shared with SEER, which would calculate the probability of users developing a genetic condition based on their family history. He noted that the goal would be for health insurers to pay for this service since it could lower costs down the road for members. He added that the company could advance the business case for interoperability since the need for doctors to share patient records would be pushed by patients.
Another team focused on tapping groups of physicians to source answers to patient questions. EtherHealth, already has an AngelList profile and is doing a pilot with a group of doctors from Jefferson University Hospital. Unlike similar physician query businesses like HealthTap, the company wants to provide a way for doctors to contribute their insight, experience and add to each other’s understanding of medical conditions, and collectively produce comprehensible answers to common health questions. The team also won crowd pleaser recognition in a public vote of the teams
SEER and EtherHealth snagged third and second place spots, giving them a chance to meet a DreamIt Ventures managing director to chat about their business.
