Connect with us

AI Research

Apple Researchers Create an AI Model That Uses Behavioural Data from Wearables to Predict Health Signals

Published

on


Apple researchers, in collaboration with the University of Southern California, have developed a new artificial intelligence (AI) model that tracks behavioural data over sensor signals. The new research builds on prior work by the Apple Heart and Movement Study (AHMS) and was aimed at understanding if behavioural data, such as sleep pattern and step count, can be a better determinant of a person’s health compared to traditional indices such as heart rate and blood oxygen level. As per the paper, the AI model performed surprisingly well, even if with some caveats.

New Apple Study Shows Benefits of Moving Beyond Traditional Health Data

The study, titled “Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions” was published in the pre-print journal arXiv and is yet to be peer reviewed. The researchers set out to develop an AI model, dubbed Wearable Behaviour Model (WBM), that relies on processed behavioural data from wearables such as how long a person sleeps and their REM cycles, daily steps taken and gait, and how their activity pattern changes over the week.

Traditionally, to predict or assess someone’s health, wearable health research has typically focused on raw sensor readings such as continuous heart rate monitoring, blood oxygen levels, and body temperature. The study believes that while this data can be useful at times, it also lacks the full context about the individual and can have inconsistencies.

Regardless, so far, behavioural data, which is also something most wearables process, has not been used in systems as a reliable indicator of a person’s health. There are two main reasons for it, according to the study. First, this data is much more voluminous compared to sensor data, and as a result, it can also be very noisy. Second, creating algorithms and systems that can collect and analyse this data and reliably make health predictions is very challenging.

This is where a large language model (LLM) comes in and solves the analysis problem. To solve the noise in data, researchers fed the model with structured and processed data. The data itself comes from more than 1,62,000 Apple Watch users who participated in the AHMS research, totalling more than 2.5 billion hours of wearable data.

Once trained, the AI model used 27 different behavioural metrics, which were grouped into categories such as activity, cardiovascular health, sleep, and mobility. It was then tested across 57 different health-related tasks, such as finding out if someone had a particular medical condition (diabetes or heart disease) and tracking temporary health changes (recovery from injury or infection). Compared to the baseline accuracy, researchers claimed that WMB outperformed in 39 out of 47 outcomes.

Comparison between performance of the WBM model the test model and the combination of both
Photo Credit: Apple

 

The findings from the model were then compared with another test model that was only fed raw heart data, also known as photoplethysmogram (PPG) data. Interestingly, when individually compared, there was no clear winner. However, when researchers combined the two models, the accuracy of prediction and health analysis was measured to be higher.

Researchers believe combining traditional sensor data with behavioural data could improve the accuracy in the prediction of health conditions. The study stated that behavioural data metrics are easier of interpret, align better with real-life health outcomes, and are less affected by technical errors.

Notably, the study also highlighted several key limitations. The data was taken from Apple Watch users in the US, and the broader global population was not represented in this. Additionally, due to the high price of wearable devices that accurately collect and store behavioural data, accessibility of preventive healthcare also becomes a challenge.



Source link

AI Research

Clarivate Unveils Enhanced 2025 G20 Research, Innovation Scorecard with Expanded Data, AI Insights

Published

on


Clarivate (NYSE:CLVT) is one of the cheap IT stocks hedge funds are buying. On July 9, Clarivate released its annual 2025 G20 Research and Innovation Scorecard. This scorecard was developed by experts at the Institute for Scientific Information/ISI at Clarivate and provides a data-driven overview of the research and innovation capabilities of G20 member nations.

The 2025 scorecard now incorporates data from the Emerging Sources Citation Index/ESCI, which is a part of the Web of Science Core Collection, to provide a more comprehensive view of global research. The scorecard has been refined to better emphasize collaboration and impact, reflecting South Africa’s Ubuntu philosophy, the G20 host for 2025.

Clarivate Unveils Enhanced 2025 G20 Research, Innovation Scorecard with Expanded Data, AI Insights

A state-of-the-art computer lab filled with engineers working on new analytics technologies.

Dynamic visualizations are included to showcase each member’s research performance within their economic context and academic priorities. New additions also include OECD field-level breakdowns, insights into open access, and research aligned with Sustainable Development Goals (SDGs), highlighting how G20 nations are collaborating to address global challenges.

Clarivate (NYSE:CLVT) is an information services provider in the Americas, the Middle East, Africa, Europe, and the Asia Pacific.

While we acknowledge the potential of CLVT as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you’re looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the best short-term AI stock.

READ NEXT: 30 Stocks That Should Double in 3 Years and 11 Hidden AI Stocks to Buy Right Now.

Disclosure: None. This article is originally published at Insider Monkey.



Source link

Continue Reading

AI Research

Louis Vuitton says UK customer data stolen in cyber-attack | Cybercrime

Published

on


Louis Vuitton has said the data of some UK customers has been stolen, as it became the latest retailer targeted by cyber hackers.

The retailer, the leading brand of the French luxury group LVMH, said an unauthorised third party had accessed its UK operation’s systems and obtained information such as names, contact details and purchase history.

The brand, which last week said its Korean operation had suffered a similar cyber-attack, told customers that no financial data such as bank details had been compromised.

“While we have no evidence that your data has been misused to date, phishing attempts, fraud attempts, or unauthorised use of your information may occur,” the email said.

The company said it had notified the relevant authorities, including the Information Commissioner’s Office.

The hack took place on 2 July, according to Bloomberg, which first reported the breach. It is the third breach of LVMH’s systems in the last three months.

As well as the two attacks on Louis Vuitton, LVMH’s second-largest fashion label, Christian Dior Couture, said in May that hackers had accessed some customer data.

On Thursday, four people were arrested as part of an investigation into cyber-attacks on Marks & Spencer, the Co-op and Harrods.

Those arrested were a 17-year-old British boy from the West Midlands, a 19-year-old Latvian man from the West Midlands, a 19-year-old British man from London and a 20-year-old British woman from Staffordshire.

M&S was the first retailer to be attacked, in April, in an incident that forced the closure of its online store for nearly seven weeks. The Co-op was attacked in the same month and forced to shut down parts of its IT system.

Harrods said on 1 May it had been targeted, and restricted internet access across its websites after attempts to gain unauthorised access to its systems.

The arrests came days after the M&S chair, Archie Norman, told MPs that two other large British companies had been affected by unreported cyber-attacks in recent months, as he gave details of the “traumatic” attack on the retailer.

Louis Vuitton has been approached for comment.



Source link

Continue Reading

AI Research

This Underrated Artificial Intelligence (AI) Stock Is Crushing the Market, and It Could Skyrocket Higher

Published

on


The tech-heavy Nasdaq Composite index has been in turnaround mode for the past three months — up an impressive 31% over that span after a rocky start to the year. That’s not surprising, as strong quarterly results from major technology companies seem to have boosted investor confidence in the sector.

Lumentum Holdings (LITE 1.07%) has been one of the beneficiaries of the tech stock rally. Shares of the company, which sells optical and photonic products that go into data centers and telecom networks to enable fast data transmission, have shot up by an impressive 75% in the past three months, putting them back within 10% of the peak they reached in January.

The good part is that Lumentum’s rebound wasn’t just a function of the broader market’s rally, but also because of the company’s healthy growth, which is being powered by its customers’ increasing investments in artificial intelligence (AI) infrastructure. Better still, this stock is likely to deliver more upside to investors.

Image Source: Getty Images

Lumentum’s products are playing a key role in AI data centers

Lumentum is witnessing terrific demand for its externally modulated lasers (EMLs), which are widely deployed in high-speed optical communications applications such as data centers and telecom networks. Their ability to transmit data over long distances at high speeds makes them ideal for deployment in AI data centers.

On Lumentum’s May earnings call, CEO Michael Hurlston noted:  “We set another record for EML chip set shipments this quarter and remain on track to more than double this business by the end of calendar 2025.” At the same time, the company is going to further expand its production capacity of EMLs in an effort to meet the end-market demand.

That’s the smart thing to do, considering that according to market research firm LightCounting, the optical transceiver market is expected to generate $10 billion in revenue in 2026 — as compared to $5 billion in 2024 — before rising to $20 billion in 2030. 

LITE Revenue (Quarterly) Chart

LITE Revenue (Quarterly) data by YCharts

The company’s revenue in its fiscal 2025 third quarter increased by 16% year over year to $425.2 million. Its adjusted earnings per share jumped by more than 500% — from $0.09 to $0.57 — on the back of an improvement in manufacturing utilization rates as well as the higher margins of its laser components deployed in AI servers.

Analysts’ consensus estimates are for a 20% increase in Lumentum’s revenue this year to $1.6 billion. Importantly, its top-line growth is expected to remain solid next year as well, thanks to the improving demand for AI-focused optical components.

LITE Revenue Estimates for Current Fiscal Year Chart

LITE Revenue Estimates for Current Fiscal Year data by YCharts.

Even better, that top-line growth is expected to filter down to the bottom line as well. The analysts are forecasting a 94% increase in earnings in the current fiscal year to $1.96 per share. 

LITE EPS Estimates for Current Fiscal Year Chart

LITE EPS Estimates for Current Fiscal Year data by YCharts.

That won’t be surprising when we take into account the margin gains that its laser components deployed in AI data centers are delivering.

Why this AI stock is set up for more upside

What’s worth noting is that Lumentum is trading at an incredibly attractive valuation despite the impressive growth that it has been clocking. It has a forward earnings multiple of 24, which is lower than the tech-laden Nasdaq-100 index’s average forward price-to-earnings ratio of 29.

Even assuming Lumentum continues to trade at a discount to the broader index after a couple of years, if it delivers the healthy earnings growth that analysts are anticipating, its stock could fly higher. Based on a P/E ratio of 24, its projected earnings of $5.21 per share in its fiscal 2027 point toward a stock price of $125. That would be 37% above the current level.

However, don’t be surprised to see this AI stock jumping higher than that, as the market could reward it with a premium multiple thanks to its strong earnings growth. That’s why even after its impressive rebound over the past three months, investors should consider buying Lumentum before it possibly flies even higher.



Source link

Continue Reading

Trending