Abstract: Described herein are systems and methods for training first and second neural network models. A system comprises a memory comprising instruction data representing a set of instructions and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to set a weight in the second model based on a corresponding weight in the first model, train the second model on a first dataset, wherein the training comprises updating the weight in the second model and adjust the corresponding weight in the first model based on the updated weight in the second model.

Abstract: A method, system and computer-program product for identifying neural network inputs for a neural network that may have been incorrectly processed by the neural network. A set of activation values (of a subset of neurons of a single layer) associated with a neural network input is obtained. A neural network output associated with the neural network input is also obtained. A determination is made as to whether a first and second neural network input share similar sets of activation values, but dissimilar neural network outputs or vice versa. In this way a prediction can be made as to whether one of the first and second neural network inputs has been incorrectly processed by the neural network.

Abstract: According to an aspect, there is provided a method of operating a non-invasive blood pressure, NIBP, monitor to measure the blood pressure of a subject, the NIBP monitor comprising a cuff, a pressure sensor for measuring the pressure in the cuff and for outputting a pressure signal representing the pressure in the cuff and a physiological parameter sensor, the method comprising obtaining a first measurement of a physiological parameter for the subject during inflation of the cuff, the first measurement being obtained from the pressure signal; obtaining a second measurement of the physiological parameter for the subject during inflation of the cuff, the second measurement being obtained from the physiological parameter sensor; comparing the first measurement and the second measurement; and estimating the reliability of a blood pressure measurement obtained by the NIBP monitor during inflation of the cuff based on the result of the step of comparing.

Abstract: The present invention relates to a device, system and method for monitoring a subject. To obtain increased performance at reasonable cost and complexity, the device comprises an input interface (81) configured to obtain image data acquired over time by an imaging unit from a field of view containing at least part of a subject, a monitoring signal determination unit (82) configured to determine a monitoring signal from the obtained image data of the subject, a processing unit (83) configured to evaluate the image data to determine the spatial configuration of the subject with respect to the imaging unit and to generate a feedback signal for providing feedback about the determined spatial configuration, wherein the feedback signal is configured to control a projection unit (9) to indicate through illumination of the field of view if the spatial configuration can be maintained or shall be changed, and an output interface configured to output the feedback signal.

Abstract: There is provided a method and apparatus for determining at least one of a position and an orientation of a wearable device on a subject. At least one physiological characteristic signal is acquired from a subject (402). The acquired at least one physiological characteristic signal from the subject has one or more characteristics indicative of at least one of a position and an orientation of a wearable device on the subject. The at least one of the position and orientation of the wearable device on the subject is determined based on the one or more characteristics indicative of the at least one of the position and orientation of the wearable device on the subject (404).

Abstract: There is provided a computer-implemented method (200) and system for training a model. A first user input is received to annotate a first parameter in a portion of data (202). The first model is used to predict an annotation for at least one other parameter of the portion of data based on the received first user input for the first parameter (204). The annotated first parameter, the predicted annotation of the at least one other parameter and the portion of data are used as training data to train a second model (206).

Abstract: There is provided an apparatus (12) for use with a wearable cuff (14) in determining blood pressure and/or pulse rate. The wearable cuff (14) is inflatable to pressurize a measurement site of a subject (18). The apparatus (12) is configured to acquire a pressure signal indicative of pressure in the wearable cuff (14) during inflation of the wearable cuff (14) to pressurize the measurement site of the subject (18) and analyze the acquired pressure signal to detect an artifact corresponding to a decrease in pressure during inflation of the wearable cuff (14) due to cuff slippage. The apparatus (12) is also configured to determine the blood pressure and/or pulse rate of the subject (18) by analyzing oscillations in the acquired pressure signal, wherein determining comprises compensating for the detected artifact.

Abstract: There is provided a restraint management apparatus. The restraint management apparatus comprises a processing unit arranged to: receive one or more types of sensor data; determine a status of a subject based on the received sensor data; determine, based on the determined subject status, a restraint parameter for a restraint device configured to restrain a body part of the subject; and output a signal based on the determined restraint parameter.

Abstract: The present invention relates to an apparatus (18) for providing a control signal for a blood pressure measurement device, comprising: an input interface (24) for obtaining a health state parameter being indicative of a health state of a patient (12); a processing unit (28) for determining one or more operation settings of a blood pressure measurement device (14) based on the health state parameter, said one or more operation settings including a parameter that can be adjusted at the blood pressure measurement device (14) when conducting a blood pressure measurement with the device and that affects a precision of said blood pressure measurement and a patient comfort resulting from said blood pressure measurement; and a control interface (30) for providing a control signal for a blood pressure measurement device (14) to perform a blood pressure measurement based on said one or more operation settings. The present invention further relates to a corresponding method.

Abstract: A method, system and computer-program product for identifying neural network inputs for a neural network that may have been incorrectly processed by the neural network. A set of activation values (of a subset of neurons of a single layer) associated with a neural network input is obtained. A neural network output associated with the neural network input is also obtained. A determination is made as to whether a first and second neural network input share similar sets of activation values, but dissimilar neural network outputs or vice versa. In this way a prediction can be made as to whether one of the first and second neural network inputs has been incorrectly processed by the neural network.

Abstract: There is provided a context detection apparatus for use with a medical monitoring device arranged to measure a physiological characteristic of a subject. The context detection apparatus comprises at least one context sensor comprising a camera; and a processing unit. The processing unit is arranged to receive image data acquired by the camera from the at least one context sensor; analyze the received image data to generate a plurality of different types of context information; wherein each type of context information comprises information which relates to a factor capable of influencing measured values of the physiological characteristic and which is not measurable by the medical monitoring device; and output a signal to the medical monitoring device based on the generated context information.

Abstract: Method of operating a non-invasive blood pressure, NIBP, monitor to measure the blood pressure of a subject, the NIBP monitor comprising a cuff, a pressure sensor for measuring the pressure in the cuff and for outputting a pressure signal representing the pressure in the cuff and a physiological parameter sensor, the method comprising obtaining a first measurement of pulse rate or heart rate for the subject during inflation of the cuff, the first measurement being obtained from the pressure signal; obtaining a second measurement of the pulse rate or heart rate for the subject during inflation of the cuff, the second measurement being obtained from the physiological parameter sensor; comparing the first measurement and the second measurement; and estimating the reliability of a blood pressure measurement obtained by the NIBP monitor during inflation of the cuff based on the result of the step of comparing.

Abstract: According to an aspect, there is provided a method of operating a non-invasive blood pressure, NIBP, monitor to measure the blood pressure of a subject, the NIBP monitor comprising a cuff, a pressure sensor for measuring the pressure in the cuff and for outputting a pressure signal representing the pressure in the cuff and a physiological parameter sensor, the method comprising obtaining a first measurement of a physiological parameter for the subject during inflation of the cuff, the first measurement being obtained from the pressure signal; obtaining a second measurement of the physiological parameter for the subject during inflation of the cuff, the second measurement being obtained from the physiological parameter sensor; comparing the first measurement and the second measurement; and estimating the reliability of a blood pressure measurement obtained by the NIBP monitor during inflation of the cuff based on the result of the step of comparing.

Abstract: A method, and corresponding system (300), for monitoring patients based on activity state and posture. Activity state and/or posture of a patient are measured. Further, one or more vital signs of the patient are measured according to a schedule. Based on the measured activity state and/or posture of the patient and the measured one or more vital signs, the schedule is adjusted and/or patient deterioration is monitored for.

Abstract: Described herein are systems and methods for training first and second neural network models. A system comprises a memory comprising instruction data representing a set of instructions and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to set a weight in the second model based on a corresponding weight in the first model, train the second model on a first dataset, wherein the training comprises updating the weight in the second model and adjust the corresponding weight in the first model based on the updated weight in the second model.

Abstract: There is provided a method and apparatus for determining at least one of a position and an orientation of a wearable device on a subject. At least one physiological characteristic signal is acquired from a subject (402). The acquired at least one physiological characteristic signal from the subject has one or more characteristics indicative of at least one of a position and an orientation of a wearable device on the subject. The at least one of the position and orientation of the wearable device on the subject is determined based on the one or more characteristics indicative of the at least one of the position and orientation of the wearable device on the subject (404).

Abstract: A method includes generating at least first and second histograms respectively for at least first and second sets of vital sign measurements using at least first and second predetermined bins. The first and second sets of the vital sign measurements each include at least two measurements acquired at different times, and the first and second vital sign are different vital signs. The method further includes generating a first score distribution for the first vital sign by mapping each bin of the first predetermined bins to a corresponding predetermined score. The method further includes generating a second score distribution for the second vital sign by mapping each bin of the second predetermined bins to a corresponding predetermined score. The method further includes generating a compound score distribution for the first and second vital signs based on the first and second score distributions, the compound score distribution indicates a patient's health state.

Abstract: The invention relates to a fever detection apparatus for detecting fever of a living being. A heart rate providing unit (3) provides a heart rate value and a peripheral physiological value providing unit (3) provides a peripheral physiological value, wherein a heart rate characteristics determination unit (4) determines heart rate characteristics from the heart rate value and a peripheral characteristics determination unit (5) determines peripheral characteristics from the peripheral physiological value. A fever detection unit (6) detects fever depending on the heart rate characteristics and the peripheral characteristics. Since the heart rate characteristics and the peripheral characteristics are influenced by fever, the fever detection apparatus can be used to reliably and preferentially unobtrusively detect fever.

Abstract: A method of calibrating a monitoring device to be attached to a user is provided. Prior to attachment of the device, the device is aligned with respect to the user such that the measurement reference frame of the device is substantially aligned with a reference frame of the user. A first measurement of the orientation of the device with respect to a world reference frame is obtained. After attachment of the device, a second measurement of the orientation of the device with respect to a world reference frame is obtained. A transformation matrix is determined for use in transforming subsequent measurements obtained by the device into the reference frame of the user. The matrix is calculated using the first and second measurements and information on the amount of rotation of the device relative to the user about a vertical axis in the world reference frame between the first and second measurements being taken.

Abstract: The present invention relates to an apparatus (18) for providing a control signal for a blood pressure measurement device, comprising: an input interface (24) for obtaining a health state parameter being indicative of a health state of a patient (12); a processing unit (28) for determining one or more operation settings of a blood pressure measurement device (14) based on the health state parameter, said one or more operation settings including a parameter that can be adjusted at the blood pressure measurement device (14) when conducting a blood pressure measurement with the device and that affects a precision of said blood pressure measurement and a patient comfort resulting from said blood pressure measurement; and a control interface (30) for providing a control signal for a blood pressure measurement device (14) to perform a blood pressure measurement based on said one or more operation settings. The present invention further relates to a corresponding method.