Zepp Health Corp v. University Of Connecticut

1. Case Identification

• Case #: IPR2025-00523
• Patent #: 10,278,647
• Petitioner(s): Zepp Health Corporation
• Patent Owner(s): University of Connecticut
• Challenged Claims: 1-75

2. Patent Overview

• Title: Heart Rate Monitoring and Oxygen Saturation Estimation
• Brief Description: The ’647 patent discloses methods and apparatus for monitoring heart rate and estimating oxygen saturation, particularly during intensive physical activity. The technology involves using time-varying spectral analysis on a heart-related signal (e.g., a photoplethysmography or PPG signal) and a motion signal to identify and suppress motion artifacts, thereby improving the accuracy of the physiological measurements.

3. Grounds for Unpatentability

Ground 1: Obviousness over Lai - Claims 1-75 are obvious over Lai.

• Prior Art Relied Upon: Lai (Po-Hsiang Lai et al., Lightweight Wrist Photoplethysmography for Heavy Exercise: Motion Robust Heart Rate Monitoring Algorithm, Healthcare Technology Letters (Feb. 2015)).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Lai, a journal article published before the ’647 patent’s priority date, discloses or renders obvious every limitation of the challenged claims. Lai described a method for mitigating motion artifacts in wrist-worn PPG sensors by processing both PPG and accelerometer signals. Petitioner asserted that Lai’s system performs the key functions of independent claims 1, 35, and 69. Specifically, Lai’s use of Short-Time Fourier Transform (STFT) on both the PPG and motion signals constituted the claimed "time-varying spectral analysis." The process of filtering these signals before analysis met the "pre-processing" limitation, and the resulting spectral plots were equivalent to the claimed "time-frequency spectrum" (TFS). Petitioner contended that Lai’s method of analyzing spectral peaks to distinguish cardiac components from motion-induced noise inherently performs a "classification of the movement." The overall process of processing noisy raw signals to produce a robust heart rate estimate was argued to be equivalent to the claimed "reconstructing a representation of the heart-related signal."
  • Prior Art Mapping (Dependent Claims): Petitioner systematically argued that Lai also rendered all dependent claims obvious. These arguments asserted that the additional limitations recited in the dependent claims were either directly taught by Lai or represented routine, predictable implementations that a person of ordinary skill in the art (POSITA) would have found obvious. This included well-known signal processing techniques like filtering and down-sampling (claims 4-5, 38-39), the use of fallback mechanisms such as relying on prior values when a signal is lost (claims 8C, 12), analyzing signals within a standard physiological heart rate range of 0.5-3 Hz (claims 9, 11), and segmenting signals into time windows for analysis (claims 26-27). Further, Petitioner argued that implementing Lai's system in real-time or co-locating the PPG and motion sensors would have been obvious design choices for a wearable device.
  • Motivation to Combine: As a single-reference ground, the argument focused on a POSITA’s motivation to apply Lai’s teachings to solve the well-known problem of motion artifacts corrupting PPG signals in wearable devices. Petitioner argued that Lai itself provides the complete framework and that any minor variations required to meet specific claim limitations would have been obvious modifications based on established principles in the field of physiological signal processing. A POSITA would have recognized Lai’s techniques as a direct solution for improving the reliability of wearable heart rate monitors.
  • Expectation of Success: Petitioner asserted that a POSITA would have had a high expectation of success. The methods described in Lai, such as using accelerometers to track motion and applying standard spectral analysis techniques like STFT, were well-established and known to produce predictable results in mitigating noise from physiological signals.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under either 35 U.S.C. §314(a) (considering Fintiv factors) or §325(d) would be inappropriate.
• Fintiv Factors: Petitioner contended the Fintiv factors strongly favor institution. It offered a Sotera stipulation, committing not to pursue in the parallel district court litigation any grounds raised or that reasonably could have been raised in the IPR. Petitioner also argued that the parallel litigation is in a very early stage with minimal investment and that the scheduled trial date of February 2026 is distant and likely to be delayed, meaning a Final Written Decision (FWD) from the Board would issue well before trial.
• §325(d) Factors: Petitioner argued that denial under §325(d) is unwarranted because the primary prior art reference, Lai, presents different teachings and was not before the Examiner during the original prosecution of the ’647 patent.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-75 of the ’647 patent as unpatentable under 35 U.S.C. §103.