SUMMARY
This technology enables on-body AI data analysis of data from skin-like wearable electronics for achieving human-integrated/mimetic intelligent systems.
The Unmet Need: Data processing for precision medicine
Precision medicine, the future landscape of healthcare, provides personalized diagnosis and treatment to individuals by taking into account the underlying differences in people’s genes, ages, health histories, and living environments. On the technological level, this futuristic vision for healthcare requires the development of two major capabilities:
-
The effective and continuous acquisition of multi-modal health data during long-term daily activities outside of clinics, for which wearable electronics emerge as the ideal solution
-
The high-throughput and intelligent analysis of such datasets for extracting underlying personalized health patterns, which is becoming one of the main application directions of artificial intelligence (AI)
While many electrical components have been designed to incorporate increased stretchability for wearable technology, computing units for implementing AI data processing also need to have skin-like mechanical properties. However, there is a major hardware gap in realizing these stretchable computing units.
Neuromorphic computing, a new computing paradigm that mimics brain operation, has recently been introduced as a more suitable platform for AI. It offers lower system complexity, lower energy consumption and faster speed. However, there have not been any reports of stretchable neuromorphic devices that possess the performance needed for implementing AI algorithms.
The Proposed Solution: Flexible devices that enable on-body data processing
Sihong Wang has developed the first intrinsically stretchable neuromorphic device that provides all desired computational and mechanical characteristics:
-
High stretchability (100% strain)
-
A large number (>800) of memory states
-
Quasi-linear/symmetric weight update
-
Excellent switching endurance (>108)
-
Low variation in weight update
-
Good state retention (>104 s)
The further integration of this device into a neuromorphic array has demonstrated the ideal implementation of the basic computation in ANN algorithms—vector-matrix multiplication (VMM)—under stretching to 100% strain.
Researchers have successfully realized different types of neural-network simulations on a large-scale array built from the stretchable neuromorphic devices. The testing demonstrates the promise and the possible pathway for realizing skin-like, on-body AI computation.
FIGURE
ADVANTAGES
ADVANTAGES
-
High stretchability of 100% strain over 100 repeated cycles
-
Can be combined with VMM for realizing AI-based classification of electrocardiograms under different stretching states
-
Provides a large number (>800) of memory states
-
Highly linear/symmetric weight update with low switching variations
-
Excellent switching endurance (>10^8)
-
Good state retention (>10^4 s)
APPLICATIONS
-
Human-integrated/mimetic intelligent systems
-
Wearable healthcare devices
-
Soft robotics
-
Stretchable and flexible circuit components
PUBLICATIONS
Shilei Dai, Yahao Dai, Zixuan Zhao, Fangfang Xia, Yang Li, Youdi Liu, Ping Cheng, Joseph Strzalka, Songsong Li, Nan Li, Qi Su, Shinya Wai, Wei Liu, Cheng Zhang, Ruoyu Zhao, J. Joshua Yang, Rick Stevens, Jie Xu, Jia Huang, Sihong Wang. Intrinsically stretchable neuromorphic devices for on-body processing of health data with artificial intelligence, Matter, Volume 5, Issue 10, 2022, Pages 3375-3390, ISSN 2590-2385
February 26, 2024
Proof of concept
Patent Pending
Licensing
Sihong Wang