MAROKO133 Hot ai: Wearable devices could monitor pregnancy abnormalities by tracking physi

📌 MAROKO133 Update ai: Wearable devices could monitor pregnancy abnormalities by t

A simple Apple watch or fitness tracker could potentially fill a crucial gap in health care by helping women track their pregnancies, and isn’t it funny that many of us might already own one?

More than two million women of childbearing age in the United States live in “maternal care deserts,” or areas with limited access to obstetric care.

According to a press release, the new study, published in Lancet eBioMedicine, couldn’t have come at a more opportune time.

Recent data shows that 24% of women in the U.S. own a smartwatch, and a 2020 survey found women were more likely to own a fitness tracker than men. That’s 41.2 million people.

Pregnancy complications such as miscarriage and preterm birth put mothers and children at risk, making solutions necessary to find.

And it turns out that the one that might make the most significant impact already exists on many women’s wrists, so Apple watches and fitness trackers could revolutionize preventative care.

“Wearable devices offer a unique opportunity to develop innovative solutions that address the high number of adverse pregnancy outcomes in the U.S.”

But how?

Monitoring pregnancy via heart rate

Researchers from Scripps Research collected data from 5,600 participants, including 108 women who provided more detailed health information three months before their pregnancy through six months after delivery. They even monitored their sleep and activities.

The data didn’t lie — scientists identified physiological patterns that aligned with the fluctuation of key pregnancy hormones such as estrogen, progesterone, and human chorionic gonadotropin (hCG). Critical to healthy pregnancy outcomes, these hormones provide insights into how a pregnancy progresses.

“The heart rate data was particularly compelling,” per the press release.

Early during the pregnancy, “the heart rate dropped around weeks five to nine and then steadily increased until eight to nine weeks before delivery, reaching peaks up to 9.4 beats per minute above pre-pregnancy levels. After birth, the heart rate dropped below baseline levels before stabilizing around six months postpartum.”

To validate all this, “the Scripps team compared the patterns the wearable sensor detected with published hormone-level data from previous pregnancy studies, creating detailed models that predicted heart rate changes based on expected hormonal fluctuations throughout pregnancy,” the press release continued.

The research is still in its beginning stages. However, it demonstrates that good news might be on the horizon — that wearables might assist in prenatal care, especially if women are living in these “maternal care deserts.”

Gotta track those hormones ladies, as you’re power walking

“Hormones play a key role in pregnancy outcomes,” explained co-senior author Tolúwalàṣẹ Àjàyí.

“Discovering the association between heart rate and hormone changes could unlock new ways to predict the beginning of pregnancy or identify signs of adverse outcomes such as gestational diabetes or preeclampsia.”

In a small number of cases, pregnancies that ended in miscarriage or stillbirth —and in the U.S., that’s about 1 in 4 pregnancies, according to Yale- heart rate patterns differed from healthy pregnancies. But, as it goes in science, researchers have to conduct tests on larger groups of women.

All the same, some of us have access to technology that might find a new role to play in our lives. By transforming consumer devices into medical monitoring tools, gaps could be bridged in health care by providing continuous oversight. Early detection is typically preferred.

Now, researchers will expand their analysis to consider demographics, socioeconomics, and geography. “Our goal is to determine whether this approach could eventually contribute to more personalized pregnancy care,” stated senior author Giulia Milan in a press release.

The press release concludes that future studies must investigate whether these physiological changes captured by wearables could support clinical decision-making and patient care — meaning, would they help eliminate a step? Can they?

It looks promising for the moment, but the extent to which this data can be reliable is yet to be determined.

Read the study in The Lancet.

🔗 Sumber: interestingengineering.com

📌 MAROKO133 Hot ai: ByteDance Introduces Astra: A Dual-Model Architecture for Auto

The increasing integration of robots across various sectors, from industrial manufacturing to daily life, highlights a growing need for advanced navigation systems. However, contemporary robot navigation systems face significant challenges in diverse and complex indoor environments, exposing the limitations of traditional approaches. Addressing the fundamental questions of “Where am I?”, “Where am I going?”, and “How do I get there?”, ByteDance has developed Astra, an innovative dual-model architecture designed to overcome these traditional navigation bottlenecks and enable general-purpose mobile robots.

Traditional navigation systems typically consist of multiple, smaller, and often rule-based modules to handle the core challenges of target localization, self-localization, and path planning. Target localization involves understanding natural language or image cues to pinpoint a destination on a map. Self-localization requires a robot to determine its precise position within a map, especially challenging in repetitive environments like warehouses where traditional methods often rely on artificial landmarks (e.g., QR codes). Path planning further divides into global planning for rough route generation and local planning for real-time obstacle avoidance and reaching intermediate waypoints.

While foundation models have shown promise in integrating smaller models to tackle broader tasks, the optimal number of models and their effective integration for comprehensive navigation remained an open question.

ByteDance’s Astra, detailed in their paper “Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning” (website: https://astra-mobility.github.io/), addresses these limitations. Following the System 1/System 2 paradigm, Astra features two primary sub-models: Astra-Global and Astra-Local. Astra-Global handles low-frequency tasks like target and self-localization, while Astra-Local manages high-frequency tasks such as local path planning and odometry estimation. This architecture promises to revolutionize how robots navigate complex indoor spaces.

Astra-Global: The Intelligent Brain for Global Localization

Astra-Global serves as the intelligent core of the Astra architecture, responsible for critical low-frequency tasks: self-localization and target localization. It functions as a Multimodal Large Language Model (MLLM), adept at processing both visual and linguistic inputs to achieve precise global positioning within a map. Its strength lies in utilizing a hybrid topological-semantic graph as contextual input, allowing the model to accurately locate positions based on query images or text prompts.

The construction of this robust localization system begins with offline mapping. The research team developed an offline method to build a hybrid topological-semantic graph G=(V,E,L):

• V (Nodes): Keyframes, obtained by temporal downsampling of input video and SfM-estimated 6-Degrees-of-Freedom (DoF) camera poses, act as nodes encoding camera poses and landmark references.
• E (Edges): Undirected edges establish connectivity based on relative node poses, crucial for global path planning.
• L (Landmarks): Semantic landmark information is extracted by Astra-Global from visual data at each node, enriching the map’s semantic understanding. These landmarks store semantic attributes and are connected to multiple nodes via co-visibility relationships.

In practical localization, Astra-Global’s self-localization and target localization capabilities leverage a coarse-to-fine two-stage process for visual-language localization. The coarse stage analyzes input images and localization prompts, detects landmarks, establishes correspondence with a pre-built landmark map, and filters candidates based on visual consistency. The fine stage then uses the query image and coarse output to sample reference map nodes from the offline map, comparing their visual and positional information to directly output the predicted pose.

For language-based target localization, the model interprets natural language instructions, identifies relevant landmarks using their functional descriptions within the map, and then leverages landmark-to-node association mechanisms to locate relevant nodes, retrieving target images and 6-DoF poses.

To empower Astra-Global with robust localization abilities, the team employed a meticulous training methodology. Using Qwen2.5-VL as the backbone, they combined Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). SFT involved diverse datasets for various tasks, including coarse and fine localization, co-visibility detection, and motion trend estimation. In the GRPO phase, a rule-based reward function (including format, landmark extraction, map matching, and extra landmark rewards) was used to train for visual-language localization. Experiments showed GRPO significantly improved Astra-Global’s zero-shot generalization, achieving 99.9% localization accuracy in unseen home environments, surpassing SFT-only methods.

Astra-Local: The Intelligent Assistant for Local Planning

Astra-Local acts as the intelligent assistant for Astra’s high-frequency tasks, a multi-task network capable of efficiently generating local paths and accurately estimating odometry from sensor data. Its architecture comprises three core components: a 4D spatio-temporal encoder, a planning head, and an odometry head.

The 4D spatio-temporal encoder replaces traditional mobile stack perception and prediction modules. It begins with a 3D spatial encoder that processes N omnidirectional images through a Vision Transformer (ViT) and Lift-Splat-Shoot to convert 2D image features into 3D voxel features. This 3D encoder is trained using self-supervised learning via 3D volumetric differentiable neural rendering. The 4D spatio-temporal encoder then builds upon the 3D encoder, taking past voxel features and future timestamps as input to predict future voxel features through ResNet and DiT modules, providing current and future environmental representations for planning and odometry.

The planning head, based on pre-trained 4D features, robot speed, and task information, generates executable trajectories using Transformer-based flow matching. To prevent collisions, the planning head incorporates a masked ESDF loss (Euclidean Signed Distance Field). This loss calculates the ESDF of a 3D occupancy map and applies a 2D ground truth trajectory mask, significantly reducing collision rates. Experiments demonstrate its superior performance in collision rate and overall score on out-of-distribution (OOD) datasets compared to other methods.

The odometry head predicts the robot’s relative pose using current and past 4D features and additional sensor data (e.g., IMU, wheel data). It trains a Transformer model to fuse information from different sensors. Each sensor modality is processed by a specific tokenizer, combined with modality embeddings and temporal positional embeddi…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com