The Automated Device for Asthma Management
The wearable Automated Device for Asthma Management or ADAM by Health Care Originals, as shown to AppleInsider at CES can count coughs as well as measure respiration, wheeze and heart rate. To undermine effective symptoms monitoring and medication which would lead to poor health condition, a device has been proposed which could continuously monitor asthma symptoms especially coughing which is the most common symptom in asthma patients.
The identified symptoms can then be stored and retrieved later by the patient or the physician for reference, indicating the level of asthma control. It is a unique iOS connected medical device which is focused in providing the users with a solution in managing their asthma condition.Measuring vital signs together with other biological reading, the ADAM is capable of predicting as well as logging asthma attack and also recommends behaviour changes.
In other words it will be featuring inhaler detection with alerts and alert forwards together with symptom tracking as well as trending treatment plans. It is proposed to be launched during the second quarter this year, though the price has not been disclosed yet. ADAM users could also have the facility of receiving medication reminders and the device has potential of HIPAA compliant data storage.
Focused on Preventing Complication with Asthma
The device has been focused on preventing complications with asthma which could tend to get serious if not detected in time. Timely treatment could save a lot of time and trouble to the victim. The ADAM sensor has been designed with its own iOS connected application and HealthKit support is also on its way for the smart wearable accessory.
The ADAM comprises of a consumer electronic mobile platform that runs a custom application which acquires an audio signal from external user worn microphone which is connected to the device analog to digital converter or a microphone input.
To determine the absence or presence of cough sounds, the signal is processed where symptoms are tallied and raw audio waveforms are recorded, making easy access for later review by a physician. Symptoms detection algorithm depends on standard speech recognition as well as machine learning paradigms, consisting of an audio feature extraction step with a Hidden Markov Model - HMM, based Viterbi decoder which has been trained on a huge database with audio examples on various subject. Performance of recognizer is shown in terms of sensitivity as well as the rate of false alarm determined in a cross validation test.
Personal Asthma Monitoring Device
Mobile technology has made much progress so much so that the hardware needs of personal asthma monitoring devices have been created in existing consumer electronics platforms. Due to the broad install base together with software development kit –SDK, Apple’s iOS device have been the chosen platform, specifically, the 4G iPod.
The present approach is based on proven technology and method in the field of speech recognition, keyword as well as key audio effect detection, content based audio analysis and machine learning, like the other audio systems for cough monitoring which have been reported in literature.
Here the cough detection algorithm presented, processes the audio date in two ways where in the initial stage, the incoming stream of digital audio samples tend to get rearranged in fixed length frames from which set of audio features are computed and the sequence of feature vectors are then passed to HMM Viterbi decoder.