ECE’s Jeremy Gummeson Collaborates with Four Other UMass Inventors to Patent a Loose-fitting “Smart” Textile

Content

Recently, there has been a boom in textile articles with the ability to monitor physiological conditions, such as heart rate or respiration. These so-called “smart textiles” often use flexible electronic components that are integrated with one or more textile layers to form wearable, tight-fitting garments. The challenge, however, is designing smart systems that integrate flexible components into comfortable, loosely worn garments that people enjoy wearing. Now, Assistant Professor Jeremy Gummeson of our Electrical and Computer Engineering (ECE) Department and four other inventors from UMass Amherst have patented a smart textile that can not only be fashioned into relaxed and cozy garments but can also measure biological signals such as the respiration, heartbeat, and posture of a wearer. See patent description: https://patents.justia.com/patent/12064262.

Gummeson collaborated on the patent with four other researchers from UMass Amherst: Professor Deepak Ganesan of the College of Information and Computer Sciences and an adjunct in ECE; Professor Trisha Andrew of the Chemistry Department and an adjunct in the Chemical Engineering Department; and former graduate students Ali Kiaghadi (now at Apple) and S. Zohreh Homayounfar (now with DuPont), who studied under Ganesan and Andrew.

The revolutionary aspect of this newly patented smart textile is that it can be tailored into loose-fitting articles of clothing, such as pajama shirts, rather than needing the tight fit present in the watches, bracelets, or form-fitting shirts that typify current smart-textile clothing. Instead, the new patent describes “a novel, distributed, multi-modal, textile-based sensor system that can be integrated with loosely worn clothing, such as pajamas, to measure physiological signals.” 

The patented new system relies exclusively on textile-based elements in sensed regions of the body. According to the patent, “While looseness may appear to present a problem, even when a garment is considered ‘loose,’ there are several parts of the garment that are pressed against the body due to posture and/or contact with external surfaces. By carefully observing the different locations where a textile garment is naturally pressured…a myriad of sensing opportunities can be leveraged to measure vital signals, such as cardiac or respiratory properties.”

As the patent explains, “Currently existing technologies for sensing respiratory and cardiac signals all rely on tightly worn bands or electrodes that are placed at specific locations on the skin. For example, a FITBIT fitness tracker or an APPLE WATCH on the wrist is robust to a small amount of motion noise, but this is considerably less movement than what is encountered with a loose-fitting garment. Similarly, many of the ECG-sensing shirts on the market need a tight fit at several locations on the body to obtain a cardiac signal.” 

In contrast, the garment systems and methods in this new patent “enable physiological sensing with a wearable garment at the other end of the spectrum in terms of looseness; e.g., an extremely loose, daily-use textile like a pajama that is designed with comfort in mind.”

The new patent addresses the challenges created for physiological sensing in loose-fitting clothing by employing unique approaches. For locations in which there are moderate to large amounts of pressure, as the patent explains, “A novel, textile-based, pressure sensor can be used to leverage resistive and capacitive changes to measure pressure changes, such as those that result from respiration and heart beats.” 

Alternatively, “For locations where there is a tiny amount of pressure, but where the textile is dynamic, a triboelectric, textile-based sensor can be used to leverage small amounts of compression to extract the dynamics of the textile.”

As the patent summarizes the two approaches, “These sensors can be combined in a loose-fitting, textile-based garment, and their signals can be fused using a combination of signal processing and machine learning to enable holistic, textile-based sensing of physiological variables without sacrificing comfort. The system described herein combines the novel, textile-based, pressure sensor and the textile-based, triboelectric sensor and fuses signals from a distributed set of sensors to extract ballistic signals from multiple locations.”

As Gummeson sums up the work of his own ECE lab as related to the new patent, “My multidisciplinary research looks at how we can co-design computer hardware and software to enable small-form-factor, energy-efficient, mobile computing systems that seamlessly provide insights into human health and behavior.” (November 2024)