A new daily living activity recording system has been developed for monitoring health conditions and living patterns, such as respiration, posture, activity/rest ratios and general activity level. The system employs a piezoelectric sensor, a dual axis accelerometer, two low-power active filters, a low-power 8-bit single chip microcomputer and a 128 MB compact flash memory. The piezoelectric sensor, whose electrical polarization voltage is produced by mechanical strain, detects body movements. Its high-frequency output components reflect body movements produced by walking and running activities, while the low frequency components are mainly respiratory. The dual axis accelerometer detects, from body X and Y tilt angles, whether the patient is standing, sitting or lying down (prone, supine, left side or right side). The detected respiratory, behavior and posture signals are stored by the compact flash memory. After recording, these data are downloaded to a desktop computer and analyzed.
We have developed an ultrasonic stride length measuring system for analyzing the human gait. All elements of the system are quite small and each fit into an appropriate package. An ultrasonic transmitter, a digital compass, a radio transmitter and a microcontroller are attached to the subjects heel on the right shoe and in the direction of the left shoe. Two ultrasonic receivers, a digital compass, a radio receiver, a microcontroller and a 1GB SD memory card are installed on the left shoe. The ultrasonic receivers are attached to the toe and heel in the direction of the right shoe. The walking direction is thus detected by the compass attached on the right and left shoes, respectively. The stride length is detected by the difference between the radio wave and ultrasonic propagation velocities. The stride length is corrected by the detected walking direction, and then the corrected stride length is stored in the SD memory card. When downloaded, the memory card gives the accurate stride length which then is used to characterize the subjects gait during daily activity.
A new communication support system has been developed for assisting elderly people communicating by telephone and e-mail. The system consists of a conventional tablet computer, a computer-telephony interface board, a microphone, a speaker, a telephone line and the Internet. The tablet liquid crystal screen (TLCS) is used as a display device and an entry device, instead of a keyboard. When an elderly person telephones or sends an email, they then choose a communication person from the registered support personnel pictures displayed on the TLCS. The computer dials or sends e-mail, which is directly handwritten on the TLCS. When their home system receives a telephone call or e-mail, the computer displays the calling communication person's picture and name on the TLCS. The elderly person can therefore easily recognize and verify the person. Since this newly-developed communication support system does not employ a keyboard, it is easily operated by elderly persons.
We have developed a new ultrasonic stride length measuring system for analyzing the human gait. An ultrasonic transmitter, a radio transmitter, a pressure sensor and microcontroller are attached to the subjects heel on the right shoe and in the direction of the left shoe. Two ultrasonic receivers, a radio receiver, a microcontroller and a 1GB SD memory card are installed on the left shoe. Ultrasonic receivers are attached to the toe and heel, in the direction of the right shoe. When the right foot contacts the ground, its heel-mounted ultrasonic and radio transmitters simultaneously transmit to the left shoe. However, radio propagation velocity is far faster than ultrasonic velocity. Therefore, the radio wave acts as a start signal to the radio receiver of the left shoe, indicating the start of ultrasound transmission from the right shoe. Upon receiving the start signal, the microcontroller timer starts to measure each ultrasound propagation time from the right shoe to the left shoe. Distance between right and left shoes is calculated with the time and ultrasound velocity and stored in the SD memory card. Stride length is calculated with a cosine function, by using the obtained distances and the distance between the toe and heel of the left shoe, by a conventional computer. The stride length can then be used for many characterizations of the subjects gait.
We have developed a telemedicine system to monitor a patients electrocardiogram (ECG) and heart sounds (PCG) during daily activity. The complete system, consisting of an ECG recorder, an accelerometer and a 2.4 GHz low power mobile phone, is mounted on three chest sensing electrodes. The accelerometer records the PCG produced by closing of the mitral and aortic valves (S1 and S2). The sampled ECG and PCG are stored in the system for two minutes and continuously updated. When a patient feels heart discomfort such as angina or an arrhythmia, he/she pushes the data transmission switch on the system. The ECG and PCG for the next two minutes are stored in the system, and then the system then sends the four minutes of stored data directly to a hospital server computer via the 1.9 GHz low power mobile phone. These data are stored on the server and then downloaded to the physicians Java configured mobile phone. The physician can then check the patients cardiac condition, regardless of patient or physician locations, and then take appropriate actions.
We have developed a drip infusion solution monitoring system for hospital and care facility use that is much more accurate than our previous reported system. The system consists of two electrodes and an acceleration sensor. The electrodes, which are wrapped around the infusion supply polyvinyl chloride (PVC) tube from the solution bag and the drip chamber, measure the growth and fall of each drop of infusion solution. The drip rate is detected from the fall of each drop. In addition, the acceleration sensor is attached to the outside of the drip chamber and detects the tilt angle of that chamber. The injected infusion solution amount is calculated by the infusion solution quantity per one drop and the drip rate. However, the quantity changes depend on the tilt angle of the drip chamber. The quantity of each drop is then corrected by the tilt angle of the drip chamber.
A non-invasive system has been developed to monitor cardiac vibrations, respiration and posture of inbed hospitalized patients and elderly people who need constant care. These physiological variables are recorded by four 2 x 28 cm piezoelectric film acceleration sensors, eight 2 x 2 cm small pressure sensors and a 5 x 100 cm long pressure sensor. The piezoelectric sensors, attached to the chest over the heart during bed sleep or rest, detect the movements produced by the heartbeat and respiration. The eight small pressure sensors are attached at various positions on the upper and lower body. A longer pressure sensor, to detect the patient leaving the bed, is attached to the side of the bed. These sensor outputs are digitized at a sampling rate of 200 Hz using a 12-bit A/D converter and stored on a personal computer. The computer detects cardiac vibrations and respiration from upper chest movements and posture from the pressures recorded by small and long pressure sensors.
A respiratory and activity recording system has been developed for monitoring health conditions and living patterns, such as activity/rest time periods and general activity level. The system employs a piezoelectric sensor, a low-power operational amplifier, a low-power 8-bit microcomputer and a 512 KB EEPROM. The piezoelectric sensor, whose electrical polarization voltage is produced by mechanical strain, records body movements produced by respiration, heart pulse, walking and running. The high and low frequency components from the recorded body movements are discriminated by high and low pass filters. The high frequency components reflect the body movements produced by cardiac vibrations, walking and running activities. The low frequency components are mainly generated by respiration. The degree of activity is obtained by summing the high frequency components for 1 second. The microcomputer stores the obtained low frequency components and activity to the EEPROM at 0.2 second and 1 second intervals, respectively. After recording, these stored data are downloaded to a desktop computer. The computer then detects behavior patterns from the activity data and respiration from the low frequency components.
We have developed a new infusion pathway monitoring system employing linear integrated circuits and a low-power 8-bit single chip microcomputer. The system is available for hospital and home use and it constantly monitors the intactness of the pathway. The sensor is an electro-conductive polymer electrode wrapped around the infusion polyvinyl chloride infusion tube. This records an AC (alternating current) voltage induced on the patient's body by electrostatic coupling from the normal 100 volt, 60 Hz AC power line wiring field in the patient's room. If the injection needle or infusion tube becomes detached, then the system detects changes in the induced AC voltage and alerts the nursing station, via the nurse call system or PHS (personal handy phone System).
A web-based "Home Helper" support system has been developed for improving scheduling and record keeping efficiency and for eliminating unnecessary travel. This support system consists of a wireless internet mobile phone for each "Home Helper" and a server at the main office. After each visit, the Home Helpers send their care reports via the mobile phone to the office server. This server computer then creates the "filings" automatically and in appropriate format for insurance and government use.
