Also: PainChek’s app picks up European and Australian regulatory clearances; Digital health access as a social determinant of health.

AI isn’t ready for COVID-19 prime time. A systematic review of published this week in nature machine intelligence warns that new models using machine learning to review chest radiographs and chest computed tomographies for COVID-19 have major methodical deficiencies or underlying biases.

Among the 62 published or pre-print papers outlining these approaches, the authors wrote that not a single one was of potential clinical use. Many were hampered by low-quality data, and in particular the high likelihood of duplicated images across different sources that result in so-called “Frankenstein datasets.” All the proposed models also suffered some degree of bias, they wrote, such as including samples from nonrepresentative populations.

“Despite the huge efforts of researchers to develop machine learning models for COVID-19 diagnosis and prognosis, we found methodological flaws and many biases throughout the literature, leading to highly optimistic reported performance,” the reviewers wrote.

“Higher-quality datasets, manuscripts with sufficient documentation to be reproducible and external validation are required to increase the likelihood of models being taken forward and integrated into future clinical trials to establish independent technical and clinical validation, as well as cost-effectiveness.”

Easy med refills for psychiatric patients. Today marked the launch of New York-based Minded, a digital service that helps those taking psychiatric medications renew, adjust, refill and order delivery of their prescriptions.

The startup, which has raised more than $5 million from investors, aims to cut down the burden and cost of regular visits to a traditional provider for assessment and prescription renewal.

Through its app-based platform, users can instead complete a five-minute online assessment regarding their mental health and a 10-minute video consultation. If appropriate, they can either have their prescription filled at a local pharmacy or delivered to their home for free.

The subscription service costs $30 per month plus $5 for each medication, and includes 24/7 access to the company’s care team and other long-term medication management support.

“Once I found what worked for me, I did not want to go to the doctor every 90 days to pay $300 for a five-minute appointment. I wanted to take the frustrating, time-consuming, and expensive process of renewing my prescription and make it magically simple,” David Ronick, Minded cofounder and CEO, said in a statement. “We’re tackling the critical issues of access and affordability facing millions of Americans.”

Regulatory wins for pain measurement app. PainChek, the maker of a pain assessment and monitoring app for smartphones, announced this week that it’s received a CE Mark and a Therapeutic Goods Administration clearance for its Universal Pain Assessment Solution.

Designed for caretakers and others providers, the tool helps assess pain severity among those who cannot adequately describe it, or otherwise document quantified pain levels for those who can self-report. With these, the company said that it’d be rolling out the app in the U.K. and Australia next month, and then moving onto the rest of Europe and other international markets.

“PainChek can become a single, simple and rapid point-of-care solution for healthcare professionals in assessing and documenting pain across all their patients, in a broad range of settings including the larger home care and hospital care markets,” CEO Philip Daffas said in a statement. “Based on initial market feedback, we expect this novel solution will be well received by our existing users and attract a wider global audience.”

Not everyone has a smartphone. A comment letter published today in NPJ Digital Medicine makes the case that access to digital tools, and subsequently mobile health technologies, is increasingly important for healthcare stakeholders to view as another social determinant of health (SDOH).

Economic access, Internet connectivity and general tech literacy are becoming core issues as care delivery is digitized and novel tools are built using software or devices, they wrote. As such, they recommended that health systems adopt “a digital-inclusion-informed strategy regarding mobile health” that not only takes access into account, but works to assess and support patients as they learn digital skills.

“Mobile health technologies hold significant promise to increase the efficiency of care and improve health outcomes. Yet, we must be cognizant of their potential to increase health disparities,” they wrote.

The post Machine learning analyses of lung imaging for COVID-19 falls short, Minded launches to streamline psychiatric med refills and more digital health news briefs appeared first on Artificial Intelligence.

The challenge to analyze earthquake signals with optimum precision grows along with the amount of available seismic data. At the Karlsruhe Institute of Technology (KIT), researchers have deployed a neural network to determine the arrival-time of seismic waves and thus precisely locate the epicenter of the earthquake. In their report in the Seismological Research Lettersjournal, they point out that Artificial Intelligence is able to evaluate the data with the same precision as an experienced seismologist.

For precisely locating an earthquake event, it is critical to determine the exact arrival-time of the majority of seismic waves at the seismometer station (the so-called phase arrival). Without this knowledge, further accurate seismological evaluations are not possible. Such evaluations can be very useful in predicting aftershocks that sometimes cause more serious damage than the initial main earthquake. By precisely locating the epicenter, even physical processes occurring deep inside the Earth can better be distinguished, and this, in turn, allows inference about the structure of the Earth’s interior. “Our results show that Artificial Intelligence can significantly improve earthquake analysis—not only with the support of large data volumes, but also if only a limited dataset is available,” explains Professor Andreas Rietbrock from the Geophysical Institute (GPI) at KIT.

The evaluation of the recorded seismograms, which is called phase picking, helps determine the arrival-times of the individual phases. Traditionally, this is a manual procedure. The precision in manual phase picking may be affected by the subjectivity of the seismologist in charge. Most notably, however, a manual evaluation meanwhile requires unacceptable time and staff resources, due to the growing amount of seismic data and the higher density of the seismometer networks. Automated evaluation has become necessary in order to leverage all available data quickly. Indeed, the phase picking algorithms developed so far are not able to deliver the precision achieved with manual picking by an experienced seismologist—due to the extreme complexity of the formation and propagation of earthquakes, with many physical processes acting on the seismic wave field.

The post Artificial intelligence improves seismic analyses appeared first on Artificial Intelligence.