ARTICLE SUMMARY:
The first application among many possible for the start-up’s edge computing paradigm is a new screening method for COVID-19 infection. Excerpted from our February 16 profile in MedTech Strategist.
There are myriad applications for NeuTigers Inc.’s patented computational methodology, which miniaturizes neural networks so they can work in wearables and other point-of-care devices. COVID screening is one of them, but CMO Gregory Nicola, MD, notes that NeuTigers’ advances will really pay off in the early identification of disease exacerbations that might otherwise land patients in the hospital. This is the case with sickle cell anemia, where the company has already entered into a research collaboration, and congestive heart failure, the next slated indication.
Among medical artificial intelligence applications, the company is unique in offering an “edge computing” paradigm that makes sense of physiologic data gathered from sensors. AI applications based on neural networks generally operate with huge amounts of data and consume large amounts of energy, and that’s why the current model is to send data to the cloud for analysis and to return an answer to the clinician. That model isn’t ideal, since it introduces delays in screening or diagnosis and the potential for data security issues.
NeuTigers was founded in 2018, with initial funding from its management team, which is led by co-founder and CEO Adel Laoui, PhD, an executive who’s led several mobile health and medical machine-learning start-ups. In addition to his role as NeuTigers’ chief medical officer, Nicola is a practicing neuroradiologist and holds leadership positions at several health systems. Co-founder and inventor Jha, who holds 20 patents, is a Princeton professor of electrical and computer engineering and an expert in IoT and smart healthcare.
Nicola says Jha’s neural networks “are the latest and greatest because once they are built, they don’t need cloud access.” This advance is due to the architecture of the neural nets, which is created by a method called “grow and prune.”
First, Nicola explains, the network is built up around the appropriate datasets to learn from them. “Then you dig into the neural network to determine which decision trees really added no value to the final inference, and you prune those out.”
There are other unique architectural aspects, notes Nicola, for example, how the neural networks are layered. In addition, Jha and his team have developed algorithms that allow for training the networks with smaller amounts of data.
By the time COVID-19 was in full swing in early 2020, it became clear that the company’s core platform, called StarDeep, could be useful at identifying the disease by its physiological signature and thus serve as a first-line screening tool.
The COVID screener, in providing a “yes” or “no” judgment to support decision-making, is an obvious and useful go-to-market application, but, Nicola notes, “The paradigm in healthcare is rarely black and white, and screening mechanisms often cause too many complications.” Above all, he says, “healthcare needs some mechanism for monitoring chronic disease to make sure that the patient is not deteriorating. That’s where this technology is really going to have an impact.” NeuTigers’ advances will yield the greatest clinical and financial value if they can help patients with diseases prone to exacerbations identify those trends early enough to get treatment at home, thereby avoiding expensive episodes of hospitalization.
Excerpted from “NeuTigers Brings Edge Computing to COVID Detection,” MedTech Strategist, February 16, 2021.
Trial MyStrategist.com and unlock 7-days of exclusive subscriber-only access to the medical device industry's most trusted strategic publications: MedTech Strategist & Market Pathways. For more information on our demographics and current readership click here.