Donuts ARE delicious! But you didn’t need TPOH or Dr. David Shaywitz to tell you that. There is also a downside to donuts, and we all know what that is. Weight gain. And with weight gain and obesity come risks for diseases, most notably diabetes. So can a tech startup stop diabetes with an algorithm and remote log-in? And could this be a new horizon in health care?

Imagine a treatment for Type 2 diabetes that requires neither surgery, medication nor calorie restriction, but rather relies on adherence to a low-carbohydrate, high-fat diet, tracked by regular finger-stick checks of blood chemistry and guided remotely by a team of physicians, coaches and algorithms.

That’s the premise of Virta Health, a San Francisco-based digital health company formed in 2014 and launched officially today, with $37M in the bank from investors including Dr. Robert Kocher of Venrock. The kickoff follows Tuesday’s publication of the results of an uncontrolled clinical study of several hundred patients in Indiana, who will be treated and followed for two years; the just-published data – an interim report of sorts – represent the first 10 weeks of study, sponsored by Virta.

Shaywitz, an endocrinologist and chief medical officer of a genomics data management company, DNAnexus, reviewed the study’s preliminary results, and they are impressive. The number of patients whose diabetic control level reached a healthy range nearly tripled after 10 weeks using the new program and a majority of patients reduced or eliminated use of one or more diabetes medications.

Can the smart use of technology really make people healthier? After all, “disease is messy” and “people are complicated.”

For Shaywitz, the answer seems hard to quantify. Low-carb diets have been a fad for ages, and an uncontrolled clinical study of 10-weeks duration is hardly a harbinger of things to come.

On a personal level, can you imagine a life essentially bereft of birthday cake, pizza, French fries and blueberry pie? After all, the trouble with most diets isn’t that they don’t work – it’s just that they aren’t sustainable. When embarked on for weight loss, just about all fail in the long term; the statistics are staggeringly depressing.

Proponents argue that low-carb diets are sustainable because hunger is attenuated, and weight loss occurs even though calorie counting isn’t generally party of the drill. Perhaps in the context of Type 2 diabetes, patients who respond to low-carbohydrate eating (and at least in the short term, many seem to) might be especially motivated to remain with the program, and enjoy and extend the benefits they achieved.

Moreso, Virta’s use of technology is dependent on people staying with the program, which, after all, amounts to a “d-i-e-t,” and plenty of companies, like Weight Watchers, have been tracking people’s diets online for years.

Virta’s CEO Sami Inkinen, himself a diabetes patient (and not as a result of eating donuts), is no doubt a successful businessman, having sold the real estate website Trulia to Zillow for $3.5 billion in 2014. He is now onto his next venture, turning research on a low-carb lifestyle into an interactive program.

The true benefit of startup innovations like Virta may be in the way they operate, the impact these programs have on patient care, alternatives to traditional treatments, and overall health care cost-savings.

Virta presumably makes money based on number of diabetic patients who reduce their medication load or can drop them entirely. In both cases, profit is based on saving payors (principally employers) the incremental costs of caring for diabetic patients–a difference of nearly $8,000/year on average in the commercially insured population, according to one (somewhat dated) estimate.