Search

EP-4738382-A1 - HANDS-FREE MEDICAL DEVICE CONTROL

EP4738382A1EP 4738382 A1EP4738382 A1EP 4738382A1EP-4738382-A1

Abstract

The present disclosure provides concepts for detecting a hands-free input from a user for controlling a medical device. Specifically, an occurrence of an alert triggered by the medical device is identified, and the hands-free input is detected from the user. The detection of the hands-free input is based in part on one or more characteristics of the alert. Thus, more accurate and context-aware detection of hands-free inputs in medical environments is achieved by leveraging characteristics (e.g., timing, origin, urgency, etc.) of an alert generated by the medical device, improving the reliability and efficiency of medical device control in medical environments.

Inventors

  • BULUT, MURTAZA
  • VAN LIESHOUT, RON MARTINUS LAURENTIUS
  • HUIJBREGTS, LAURENTIA JOHANNA
  • VAN DER HEIDE, Esther Marjan
  • NIKOLAYEV, Aleksander

Assignees

  • Koninklijke Philips N.V.

Dates

Publication Date
20260506
Application Date
20241105

Claims (15)

  1. A method (100) for detecting a hands-free input from a user for controlling a medical device, the method comprising: identifying (110) an occurrence of an alert triggered by the medical device; and detecting (120) the hands-free input from the user, wherein detection of the hands-free input is based in part on one or more characteristics of the alert.
  2. The method of claim 1, wherein detecting (120) the hands-free input from the user comprises detecting (124) an action of the user based on the characteristics of the alert.
  3. The method of claim 2, wherein detecting (124) an action of the user is initiated responsive to the alert being triggered.
  4. The method of claim 3, wherein initiating detection (124) of the action from the user is delayed for a predetermined length of time, and optionally wherein the predetermined length of time is based on at least one of an expected time to review the alert by the user, an expected time for a user to observe the medical device, a length of an output associated with the alert, and/or a user preference.
  5. The method of any of claims 2-4, wherein detecting (124) the action of the user is performed during a detection window, wherein the length of the detection window is based on the characteristics of the alert.
  6. The method of any of claims 2-5, wherein detecting (124) the action of the user is performed using one or more selected sensors operated according to one or more operation parameters.
  7. The method of any of claims 1-6, further comprising detecting (122) the location of the user relative to the medical device, and wherein detection (120) of the hands-free input is further based on the detected location of the user.
  8. The method of any of claims 1-7, wherein detecting (120) the hands-free input from the user comprises processing (128), with an action recognition algorithm, a detected action of the user to determine the hands-free input, wherein the processing is based on the characteristics of the alert.
  9. The method of claim 8, wherein processing (128) the detected action of the user based on the characteristics of the alert comprises: modifying (126) the action recognition algorithm based on the characteristics of the alert; and processing, with the modified action recognition algorithm, the detected action of the user.
  10. The method of claim 9, wherein modifying (126) the action recognition algorithm comprises at least one of increasing or decreasing a sensitivity of the action recognition algorithm to user actions, and/or altering a mapping between detected user actions and hands-free inputs by the action recognition algorithm.
  11. The method of any of claims 8-10, wherein processing (128) the detected action of the user based on the characteristics of the alert comprises: filtering the detected action based on the characteristics of the alert; and processing, with the action recognition algorithm, the filtered detected action of the user.
  12. The method of any of claims 8-11, wherein processing (128), with the action recognition algorithm, the detected action of the user to determine the hands-free input is further based on one or more characteristics of the user.
  13. The method of any of claims 1-12, wherein the one or more characteristics of the alert include an alert timing, an alert origin, an alert urgency, an alert trigger, an alert location, an alert source, an alert output medium characteristic, an expected user action associated with the alert, and an alert circumstance.
  14. A computer program comprising computer program code means adapted, when said computer program is run on a computer, to implement the method of any of claims 1-13.
  15. A controller for a medical device (202), the method comprising: a monitoring system (210) configured to identify an occurrence of an alert triggered by the medical device; and a hands-free interface (220) configured to detect a hands-free input for controlling the subject-monitor from a user (204), wherein detection of the hands-free input by the hands-free interface is based in part on one or more characteristics of the alert.

Description

FIELD OF INVENTION The present invention relates to hands-free control of medical devices, and more particularly to systems and methods for detecting hands-free input from a user for controlling a medical device. BACKGROUND Patient monitoring systems are widely used in healthcare settings to continuously track vital signs and alert medical staff to potential issues. These systems typically generate alerts (e.g., alarms, notifications, communication requests, etc.) when measured parameters fall outside of preset ranges. However, the high volume of alerts produced by modem monitoring equipment has led to challenges for healthcare providers. Indeed, a wide range of medical devices beyond just patient monitors now utilize alerts to attract the attention of medical staff. Studies have shown that up to 85-99% of alerts may not require clinical intervention, contributing to alert fatigue among medical staff. Responding to and managing alerts consumes a significant portion of clinicians' time and attention. Current alert management typically requires manual interaction with the medical device to silence, acknowledge, or adjust alert settings. This process can be disruptive to workflow and patient care, especially in busy clinical environments. Additionally, the need for direct physical contact with medical devices may increase the risk of contamination in sensitive healthcare settings. It has been appreciated that an improved means for interacting with medical devices is needed that overcomes one or more of these problems. SUMMARY OF INVENTION The invention is defined by the claims. According to an aspect of the present disclosure, a method for detecting a hands-free input from a user for controlling a medical device is provided. The method comprises identifying an occurrence of an alert triggered by the medical device, and detecting the hands-free input from the user, wherein detection of the hands-free input is based in part on one or more characteristics of the alert. The present disclosure provides concepts for detecting a hands-free input from a user for controlling a medical device. Specifically, an occurrence of an alert triggered by the medical device is identified, and the hands-free input is detected from the user. The detection of the hands-free input is based in part on one or more characteristics of the alert. Thus, more accurate and context-aware detection of hands-free inputs in medical environments is achieved by leveraging characteristics (e.g., timing, origin, urgency, etc.) of an alert generated by the medical device, improving the reliability and efficiency of medical device control in medical environments. Proposed concepts thus aim to provide schemes, solutions, concepts, designs, methods, and systems pertaining to hands-free control of medical devices (e.g., subject monitors), particularly in the context of alert management. In particular, embodiments aim to provide improved methods and systems for detecting hands-free input from users to control medical devices based on alert characteristics. The invention presents a novel approach to hands-free medical device control that leverages the context related to alert events. By identifying the occurrence of an alert triggered by a medical device and detecting hands-free input from a user based in part on characteristics of the alert, the system can more accurately interpret user intentions and actions. This context-aware approach to hands-free input detection can significantly improve the reliability and efficiency of medical device control in potentially noisy or complex healthcare settings. A key advantage of the proposed method is its ability to adapt to different alert scenarios. The detection of user actions can be initiated in response to an alert, with optional delays based on factors like expected review time or alert audio length. This flexibility allows the system to align with natural user behaviors and workflows. Furthermore, the detection window and sensor parameters can be dynamically adjusted based on alert characteristics, optimizing the system's performance for each specific situation. In addition, the interpretation of the user actions (e.g., gestures, speech, etc.) and translation of user actions into the hands-free input for control of the medical device, may be adjusted based on the alert characteristics to further improve accuracy of control. By considering a wide range of alert attributes (e.g., timing, urgency, source), the proposed method offers a comprehensive and nuanced approach to hands-free medical device control. This holistic consideration of context has the potential to significantly enhance the user experience, improve efficiency, and reduce errors in critical healthcare environments. In some embodiments, the method may further comprise detecting an action of the user based on the characteristics of the alert. Basing the detection of user actions on alert characteristics can lead to more accurate, reliable and eff