23andMe's new COVID-19 resource. Consumer genomics company 23andMe has released a new interactive tool that allows users to see how various health factors could impact the risk of COVID-19 hospitalization. The COVID-19 Severity Calculator pulls data from the company's COVID-19 study, which included 10,000 participants diagnosed with the disease and 750 who were hospitalized.

23andMe warned that the tool shouldn't be used as a predictor for individual risk, and doesn't take lifestyle or certain underlying health conditions into account. However, it does identify certain characteristics that were frequently tied to severe symptoms and hospitalization, chief among which were obesity, Type 2 diabetes and lack of exercise.

Another record quarter during quarantine. Apple's Q1 earnings report came with plenty of good news for the tech giant's investors. The company surpassed its projections with an all-time revenue record of $111.4 billion (up 21% year-over-year), with nearly two-thirds of that coming from international sales.

The increase was powered by growth across each of the company's product categories. This included 17% YoY growth for the iPhone and 30% YoY growth for the wearables, home and accessories business that houses the Apple Watch. The company's budding services division was also up 24% YoY, with the company's newly launched subscription workout offering receiving a special shout out during the call.

"We're very excited about the road ahead for these products. Look no further than the great potential of Fitness+, which pairs with Apple Watch to deliver real-time, on-screen fitness data alongside world-class workouts by the world's best trainers," Cook said during the investor call.

"There are new sessions added each week, and customers are loving the flexibility, challenge, and fun of these classes, as well as how the pairing with Apple Watch pushes you to achieve your fitness goals. This deep integration of hardware, software and services have always defined our approach here, and it has delivered an all-time quarterly services record of $15.8 billion."

Is it finally happening? A report published this week by Korea's ET News suggests that the next generation of smartwatches from Samsung and Apple will come equipped with tools to monitor the wearer's blood glucose level – a long-rumored feature for consumer wearables. The feature would be noninvasive, instead relying on optical sensors.

For Apple in particular, the report noted that the tech company "has already secured relevant patented technologies and is currently testing the feature’s reliability and stability before the commercialization."

Startup incubator seeking applicants. Healthworks, CareFirst BlueCross BlueShield's innovation and investment arm, has partnered with Maryland-based provider system LifeBridge Health to launch a new healthcare startup incubator.

Called 1501 Health, the incubator is "looking to find and engage startups developing cutting-edge technologies and solutions that have the potential to significantly impact the healthcare ecosystem, whether that’s healthcare consumers, payers, providers or other stakeholders,” Ricardo Johnson, SVP of Healthworks, said in the announcement.

The incubator will provide participating startups with as much as $100,000 while providing mentorship, education and networking opportunities – not to mention immediate strategic partnerships with potential customers. Designed as a virtual program in light of COVID-19, the incubator is accepting applicants from February 1 to February 22.

Otsuka taps smart device for adherence studies. Otsuka Pharmaceutical's Development & Commercialization arm has partnered with Spencer Health Solutions, the maker of an at-home medication adherence platform, to further evaluate and deploy direct-to-patient services through the device. In particular, the pharma company said it will be considering the device's medication adherence and patient engagement capabilities in upcoming traditional and decentralized clinical trials.

“Our commitment to the democratization of clinical trials, patient engagement and leveraging technology to improve diversity and inclusion in clinical research helps drive our success at Otsuka,” Dr. Christoph Koenen, EVP and chief medical officer at Otsuka Pharmaceutical Development & Commercialization, said in a statement.

“The spencer smart hub will be an asset in our efforts to improve treatment plans for people living with challenging health conditions.”

A secure way to sic AI on health records. Mount Sinai researchers have published their work on a machine learning model that uses EHR records to predict COVID-19 patients' outcomes with reduced privacy risks.

Specifically, the team evaluated a federated learning approach that trains an algorithm that acts across multiple data storage sites, a method that better preserves patients' privacy by avoiding local aggregation of clinical data.

To do so, they collected EHR data from five different hospitals within the Mount Sinai Health System, and found that their federated models generally outperformed local data-dependent models when predicting seven-day mortality among the hospitalized COVID-19 patients.

"Machine learning models in health care often require diverse and large-scale data to be robust and translatable outside the patient population they were trained on," Benjamin Glicksberg, assistant professor of genetics and genomic sciences at the Icahn School of Medicine at Mount Sinai and the study's corresponding author, said in a statement.

"Federated learning is gaining traction within the biomedical space as a way for models to learn from many sources without exposing any sensitive patient data. In our work, we demonstrate that this strategy can be particularly useful in situations like COVID-19."