The three-dimensional display technique has a wide range of medical applications. Pre-operative planning is one typical application: in orthopedic surgery, three-dimensional image processing has been used very successfully. We have employed this technique in pre-operative planning for orthopedic surgery, and have developed a simulation system for bone-drilling. Positive results were obtained by pre-operative rehearsal; when a region of interest is indicated by means of a mouse on the three-dimensional image displayed on the CRT, the corresponding region appears on the slice image which is displayed simultaneously. Consequently, the status of the bone-drilling is constantly monitored. In developing this system, we have placed emphasis on the quality of the reconstructed three-dimensional images, on fast processing, and on the easy operation of the surgical planning simulation.

View full abstract

This paper describes a hip joint surgical simulation system using a three-dimensional image (X-ray CT images). We developed this system in cooperation with surgeons, and we have already applied it to several clinical examples. According to a surgeon who used our system for actual surgical plannings, most of the hip joint surgeries operated at the present time can be simulated. Our system has useful functions in order to plan surgeries such as interference check in movement of a bone and display of the attached surface of bone. We show an actual surgical planning process performed by the surgeon with our system.

View full abstract

Fuzzy blood pressure control system has been developed to regulated blood pressure by using vaso-active drugs, which aims the medical treatment of the patient. Fuzzy logic is used to convert heuristic control rules as a physician's knowledge into an automatic control strategy. Usually, a physician decides infusion rate based on referring ideal pressure trajectory during control. To emulate such a physician's heuristic control, the controller refer an ideal pressure trajectory, for example, from the normal level to the target level. One of Inputs to the controller is a difference between real pressure and ideal pressure, where another is a rate of its pressure difference. An output from the controller is the change in drug infusion rate. Both of inputs and output are transformed into fuzzy sets, whose membership functions are predefined. Twenty one physician's control rules are converted into fuzzy control rules. The fuzzy control procedure consists of three steps: (1) computing both input values and finding their appropriate membership values (grade), (2) finding applied rules and applying fuzzy composition to infer weighted output fuzzy sets, (3) inferring the change in drug infusion rate by taking the center of mass of the membership function of output. Nine animal experiments are performed to evaluate the developed system by using rats with vasopressor drug, Angiotensin II Blood pressure was elevated from normal pressure (around 100mmHg) to 150mmHg. Satisfactory blood pressure control was realized on eight experiments and effectiveness of the controller was confirmed. The fuzzy control approach is especially useful for the closed-loop control in the medical fields, because the knowledge of expert physician's treatment is directly contained and represented in the control strategy. The system is expected to be applicable to the clinical cases.

View full abstract

A new type of the 3-dimensional model fabricating system was developed using a focused laser beam and liquid photopolymer. The system can easily fabricate solid models, without tools, directly from digital data received from medical CT scanners or CAD systems. Solid models are fabricated as follows. After a layer of liquid photopolymer is placed above the already solidified layers, a laser beam is scaned along horizontally sliced patterns on the liquid photopolymer. The parts where the light shone is solidified. And then the solidified layers agreed with the patterns are stacked. This process is repeated for each layer. This paper describes the fabrication of a smoothly natural solid model from an original CT image sequence which is taken with rough pitches. Interpolation between original neighboring sliced images which are composed of extracted contour data is performed by use of the third order spline curve. This method can be implemented using a rather small computer. This technique can play a major role in diagnosis procedure, personal prosthesis design, and surgical operation planning.

View full abstract

A personal computer based digital control system for driving the ventricular assist device (VAD) has been developed. Adaptive control, monitoring and analyzing functions are integrated into a software of the personal computer in this system. These functions can operate in parallel with one another because of the multi-interrupt processing used in the software. The system enables the operator to analyze the hemodynamic behavior of the circulatory system simultaneoutly with VAD pumping and to make on-line use of the result of the analysis in order to improve the driving condition of VAD.

View full abstract

A six channels telemetry system for recording masticatory force of complete denture wearers has been developed. This system is consisted of three blocks, the transmission block, the receiver block and the data storage and correction block. The transmission block is embedded in a lower complete denture. System has six load sensors and a standard resistor to correct sensor data. Data from six load sensors and standard resistor as well as standard zero are multiplexed using a division multiplex communication method. The sampling frequency is 666Hz. And the signal is transmitted by FM modulation. The carrier wave frequency is 64MHz. Received data are decoded using start-stop system. The wave form noise is reduced using a signal averaging technique at an integrator. The data are input into the computer with an A/D converter. In this system, the sensor data is corrected at the computer with the standard resistor data. This method could reduce the influence of battery power changes and thermal drift significantly. For the load of 10kg or less, the total linearity of the system remained at +/-4% of the calibrated value. The noise level is 300g.

View full abstract

Ultrasonic heating methods for hyperthermia have been developed because of several advantages i. e., its easiness in steering the applicator for the target, and its non-interference with other electronic equipment. Most of them make hot spots by focusing the ultrasonic beam, using acoustic lens or many transmitters. Therefore, these methods are based on analogous idea to the ray focusing, and higher frequency, i. e., from 500kHz to 5MHz, is used. However, there are some problems; hot spots are generated before the focus, and it is difficult to heat the region beyond gas and bones owing to the attenuation and scattering of the ultrasonic beam. Then, we propose a new method for heating the depths and local regions of the body. In this method, to heat the depths of the body low-frequency ultrasound is used since it has larger penetration depth and it is less scattered than higher-frequency ultrasound. To heat only tumours, hot spots are generated by synthesizing acoustic fields resulted from several incident waves. The heat generation and temperature distribution was analyzed using the models with properties similar to tissues for different values of parameters such as frequency, position and the number of sources. The results show that hot spots can be generated at the depths of the body, and that desirable temperature distribution can be obtaind by selecting optimal parameters and cooling condition.
To determine the temperature distribution more accurately it would be required to introduce the more complicated model of heat removal effect by blood flow and the experimental results. These problem are left for the future investigation.

View full abstract

A nonlinear digital filter system is proposed for automatic extraction of crackles which are discontinuous adventitious sounds in lung sounds. This system is composed of two nonlinear digital filters; one is a stationary-nonstationary separating filter, and the other is a width-discriminating filter. The former separates nonstationary signals from stationary ones in the lung sounds, using the prediction error to the input lung sound signal. If the prediction error is small enough, the lung sound is considered to be stationary, but if the error is large, a nonstationary signal is considered to occur and the nonstationary part is separated. The latter, the width-discriminating filter, performs signal extraction by considering the signal wave form of the crackles, simply realized by logical algebra. This filter extracts an impulsive signal, which is a small-width wave, and its succeeding waves; such wave form is typical of that of crackles. First, crackles are roughly separated by the stationary-nonstationary separating filter as nonstationary signals, and then are more precisely extracted from the nonstationay output by the width-discriminating filter. Some examples of processing actual lung sound data by this system show its high performance.

View full abstract