Automatic characterization of user errors in spirometry.

Spirometry plays a critical role in characterizing and improving outcomes related to chronic lung disease. However, patient error in performing the spirometry maneuver, such as from coughing or taking multiple breaths, can lead to clinically misleading results. As a result, spirometry must take place under the supervision of a trained specialist who can identify and correct patient errors. To reduce the need for specialists to coach patients during spirometry, we demonstrate the ability to automatically detect four common patient errors. Creating separate machine learning classifiers for each error based on features derived from spirometry data, we were able to successfully label errors on spirometry maneuvers with an F-score between 0.85 and 0.92. Our work is a step toward reducing the need for trained individuals to administer spirometry tests by demonstrating the ability to automatically detect specific errors and provide appropriate patient feedback. This will increase the availability of spirometry, especially in low resource and telemedicine contexts.

Towards fenceless boundaries for solar powered insect biobots.

Demonstration of remote navigation with instrumented insects, such as the Madagascar Hissing Cockroach, Gromphadorhina portentosa, has enabled the concept of biobotic agents for search and rescue missions and environmental monitoring applications. The biobots can form the nodes of a mobile sensor network to be established, for example, in unknown and dynamic environments after natural disasters to pinpoint surviving victims. We demonstrate here, for the first time, the concept of an invisible fence for insect biobots with an ultimate goal of keeping insect biobots within a certain distance of each other or a base station to ensure a reliable wireless network. For extended mission durations, this fenceless boundary would also be used to guide insects towards light sources for autonomous solar charging of their on-board batteries.

Solar powered wrist worn acquisition system for continuous photoplethysmogram monitoring.

We present a solar-powered, wireless, wrist-worn platform for continuous monitoring of physiological and environmental parameters during the activities of daily life. In this study, we demonstrate the capability to produce photoplethysmogram (PPG) signals using this platform. To adhere to a low power budget for solar-powering, a 574 nm green light source is used where the PPG from the radial artery would be obtained with minimal signal conditioning. The system incorporates two monocrystalline solar cells to charge the onboard 20 mAh lithium polymer battery. Bluetooth Low Energy (BLE) is used to tether the device to a smartphone that makes the phone an access point to a dedicated server for long term continuous storage of data. Two power management schemes have been proposed depending on the availability of solar energy. In low light situations, if the battery is low, the device obtains a 5-second PPG waveform every minute to consume an average power of 0.57 mW. In scenarios where the battery is at a sustainable voltage, the device is set to enter its normal 30 Hz acquisition mode, consuming around 13.7 mW. We also present our efforts towards improving the charge storage capacity of our on-board super-capacitor.

Kinect-based system for automated control of terrestrial insect biobots.

Centimeter scale mobile biobots offer unique advantages in uncertain environments. Our previous experimentation has demonstrated neural stimulation techniques in order to control the motion of Madagascar hissing cockroaches. These trials relied on stimulation by a human operator using a remote control. We have developed a Kinect-based system for computer operated automatic control of cockroaches. Using image processing techniques and a radio transmitter, this platform both detects the position of the roach biobot and sends stimulation commands to an implanted microcontroller-based receiver. The work presented here enables repeatable experimentation and allows precise quantification of the line following capabilities of the roach biobot. This system will help refine our model for the stimulation response of the insect and improve our ability to direct them in increasingly dynamic situations.