How Can AI Improve Sleep? MIT Lab and Chinese Team Find a New Solution

Image: https://www.globalnewslines.com/uploads/2025/11/63a9c55247548c9822bb0026e52b0c1d.jpg

When fibers themselves possess sensing and thinking capabilities, technology can understand and improve human sleep in a more natural way.

On October 27, 2025, an MIT (Massachusetts Institute of Technology) lab held a closed-door demonstration event, releasing its latest research achievement: the "FiberCircuits" intelligent fiber platform. This platform was led by MIT tenured professor and academic head of the MIT lab, Director of the Sensor and Ambient Intelligence Research Group, Joseph Paradiso, and jointly developed with a research and industry team from China.

This innovative technology integrates sensing, computation, and feedback systems within the fiber itself, enabling flexible materials to possess autonomous intelligence. It opens new directions for applying artificial intelligence in sleep health research and also represents a significant achievement in Sino-US scientific collaboration in the field of intelligent soft materials.

This joint research is based on the core concept of "letting technology integrate into life as naturally as fabric." The research teams from both sides collaborated on multiple levels, including system architecture, algorithm design, and application validation. By embedding micro-sensors and AI chips into the fibers, they achieved an integrated design for signal acquisition, edge inference, and feedback mechanisms.

Paradiso pointed out that FiberCircuits marks the entry of human-computer interaction into the "weavable intelligence" stage - when fibers themselves possess sensing and thinking capabilities, technology can understand and improve human sleep in a more natural way.

"Embedding sensors and circuits into fiber-like objects allows them to be woven into items like clothing, which will elevate wearable computing to a whole new level," Paradiso said. "I anticipate this technology will spawn many applications. This is a potential game-changer that will bring forth many new possibilities."

Indeed, smart fibers are highly likely to become one of the next-generation forms of sensors, especially in scenarios involving contact with the human body, such as sleep monitoring, medical rehabilitation, and smart homes. However, they will not completely replace existing sensors but rather open a new path of "material as sensor," integrating sensing capabilities into the environment and everyday objects.

Image: https://www.globalnewslines.com/uploads/2025/11/695245e711588a02529c18b2a6b55166.jpg
Professor Paradiso

The Chinese partner team is from DeRucci Group, which played a key role in systems engineering and sleep scenario validation.

Leveraging China's research foundation in sleep science and home health, the researchers built experimental models and testing platforms targeting the sleep postures and respiratory characteristics of different populations. This provided substantial data support for the signal acquisition and algorithm training of FiberCircuits. The jointly developed experimental prototypes achieved staged results in comfort, durability, and response accuracy, laying the groundwork for future expansion into smart bedding and health monitoring research.

The core of FiberCircuits technology lies in enabling flexible materials to achieve the closed-loop function of "sensing-computation-feedback." The research team embedded integrated circuits as small as 0.9 mm into weavable flexible circuits, allowing the fibers to capture signals such as respiration, body temperature, galvanic skin response, and micro-movements in real-time, and perform local inference through optimized tinyML algorithms.

DeRucci Group CTO Chen Wenze told "Jazzyear" that the edge-side algorithms for FiberCircuits technology are primarily led by DeRucci. "We prefer to place most inference tasks, along with the data and functions capable of fitting our physiological characteristics, on the edge side. This reduces interference with users and protects user data privacy and security."

The system can also validate algorithms through feedback mechanisms like light, sound, or micro-stimulation, achieving non-invasive, low-power signal interaction. The Chinese team assisted in completing multi-dimensional signal calibration experiments related to sleep monitoring and participated in batch testing of fiber materials to ensure the stability and generalizability of model training.

It can be seen that the DeRucci team demonstrated three major advantages in the collaboration:

Engineering: Batch testing, material consistency verification.

Data: Extensive samples from different sleep populations.

Application Scenarios: Industry scenarios closer to consumer-grade implementation.

During the event, the joint research teams demonstrated multiple experimental validation prototypes for sleep science:

A mattress validation platform based on FiberCircuits, testing the sensitivity of low-density distributed sensor networks in posture recognition.

An eye mask prototype integrating Photoplethysmography (PPG) and Electromyography (EMG) modules, used to study the monitoring accuracy of respiratory rhythm and nocturnal muscle activity.

A pillow structure based on an Inertial Measurement Unit (IMU) and galvanic skin response sensors, used to analyze the dynamic relationship between pressure changes and environmental feedback.

All prototypes are for experimental purposes, aimed at validating the signal parsing capabilities and human-machine coupling effects of AI algorithms in different sleep scenarios.

MIT and DeRucci Group established sustainable technology exchange and talent interaction mechanisms during the collaboration. Both parties plan to further expand the sample population and experimental scenarios in the coming year, jointly promoting in-depth exploration of smart fiber technology in sleep research, rehabilitation assistance, and non-invasive health monitoring. The research team will also promote the joint release of some algorithms and testing protocols to foster collaboration and results sharing among global research institutions.

Professor Paradiso stated that FiberCircuits is not merely a combination of materials and AI, but an extension of a scientific philosophy - making computation and sensing part of life itself. He noted that the close cooperation in Sino-US scientific research has enabled this concept to be realized faster in engineering implementation and application validation. In the future, this research will continue to have an impact in the fields of health technology, textile intelligence, and human-machine integration, injecting new vitality into global technological innovation.

FiberCircuits demonstrates a new technological path: integrating sensing, computation, and feedback directly into the material itself, making intelligence no longer reliant on additional devices but becoming part of the environment. From sleep monitoring to medical assistance, and broader human-computer interaction scenarios, this material-level intelligence has the potential to reshape data acquisition methods and provide new interfaces for scientific and industrial collaboration.

Although it is far from mature, it has already demonstrated a possibility - intelligence can exist in our daily lives with lower intrusion and in a more natural way.

(Cover image and images in the text source: DeRucci Group)

AuthorWang Bo
Media Contact
Company Name: DeRUCCI
Contact Person: Jason
Email: Send Email [http://www.universalpressrelease.com/?pr=how-can-ai-improve-sleep-mit-lab-and-chinese-team-find-a-new-solution]
State: Dongguan
Country: China
Website: https://global.derucci.com

Legal Disclaimer: Information contained on this page is provided by an independent third-party content provider. GetNews makes no warranties or responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you are affiliated with this article or have any complaints or copyright issues related to this article and would like it to be removed, please contact retract@swscontact.com



This release was published on openPR.