In order to solve the problem of infection at the exit of the catheter, a anti-infective system that the irradiates ultrasound to titanium dioxide (TiO₂) particles bonded on the cuff and produces the reactive oxygen species (ROS), has been studied (Tominaga et al., 2019). In this study, conditions for maximizing ROS production while ensuring safety during percutaneous ultrasonic irradiation were investigated. By using a planar transducer (500 kHz, 45 mm), continuous wave of ultrasound was irradiated to a suspension consisting of diluted methylene blue (MB) solution and TiO₂ particles. The ROS production was evaluated by the absorbance of the MB solution and the safety was evaluated by the temperature rise on the transducer surface. As the result, as the above conditions, the ultrasound intensity of 1 W, the TiO₂ particles of 10 mg, and the total irradiation time of 15 minutes were determined.

We have developed a virtual reality (VR) system for home gait training of stroke patients, which provides visual, auditory, and tactile feedback upon collision with objects such as walls and furniture. The current study recruited young healthy individuals to test the effect of multimodal feedback on the number of collisions with objects, walking speed, and the number of steps in the gait test in the virtual home environment. Four times of consecutive gait training with multimodal feedback and those without feedback both resulted in moderate and non-significant improvement of the gait parameters. This is possibly due to the high gait function in the current participants and the limited statistical power due to the small number of participants. In the future, we would further increase the number of subjects and examine the effectiveness of VR gait training with multisensory feedback.

PDT (photodynamic therapy) is the cancer treatment. But PDT is not adapted to advanced cancers. 405nm is the peak absorption wavelength of Laserphyrin® and it does not penetrate into tissues deeply interfered with hemoglobin absorption and scattering attenuation in tissues. The next peak absorption wavelength 664nm is used for PDT now. Since the absorption rate at 405nm of Laserphyrin® is very high, the effect of two-photon excitation at 810nm can be expected. In this study, we investigate a feasibility of two-photon excitation PDT using near infrared light and compared at 810nm and 1328nm. We have developed Nd:YAG SHG pumped KTP-OPO that oscillate at both wavelengths by replacing the output coupler and adjusting phase matching angle with high peak power. In this paper, we report KTP-OPO configuration and optical properties at these wavelengths. We have accomplished 200kW peak power 20ns pulse width 50Hz repetition rate at both wavelengths.

Laparoscopic surgery is rapidly replacing the conventional open surgery because of its less invasiveness, improvement of postoperative quality of life, less pain and better cosmetic result. However, the current surgical instruments may cause internal interference with other instruments during the operation and blind spot may occur due to its mechanical structure. To solve these problems, we have been developing a multi-joint manipulator using smart materials called magneto-rheological gel (MRG) for the use of laparoscopic surgery.In this study, based on the results of magnetic circuit simulation, we devised a new joint structure using a permanent magnet and an electromagnet in order to reduce the amount of joule heat generated in the coil part. From the results obtained by the bending test, it was confirmed that this structure, i.e., so called "magnetic bias structure", was highly effective for reducing the energy consumption of the magnetic coils.

During urethral lithotripsy, a flexible ureteroscope is inserted into the kidney through urethra and bladder. The calculus inside the kidney is crushed by laser transmitted through an optical fiber which is inserted through channel of the ureteroscope. Because of the complex and elaborate structure in the kidney, sometimes it is hard for an ureteroscope which has a fixed optical fiber to reach the targeted calculus. In order to solve this problem, an actuator device was fabricated which has the ability to control the laser irradiation position precisely in the view of ureteroscope by bending the optical fiber with SMA (Shape Memory Alloy) wires. In this study, the structure of the device was improved to enable bending in any direction by slacken SMA wires. The new performances of it such as bending angle and response time were measured to confirm the functions.

Infrared thermography has been rarely used in medical situation. Infrared camera attached with smartphone are recently used in checking the construction area and power distribution. On the other hand, in medical situation, it has been proved that thermographic images obtained with a smartphone thermal camera have high concordance with the method such as ultrasound flow meter and computed tomography angiography. We carried out perforator detections on 11 patients admitted to the Department of Plastic and Aesthetic Surgery of Yokohama Minami Kyosai Hospital and Kitasato University School of Medicine from Apr.2018 to Dec.2019. 11 flaps were designed on buttock, calf and abdominal regions by using iphone and FLIR one, in 6 cases, 4 cases and 1 case respectively. There were 30 hot spots and, in 24 hot spot areas, ultrasound flows were detected too. Those results make it an useful method for mapping perforators in preoperative flap designing, comparable with CTA.

Reduction of the moldability of the postoperative breast shape during breast reconstruction has been cited as an issue. We have developed a system that obtains three-dimensional shape difference information before mastectomy and during reconstruction, and maps the local shape difference information to the patient's skin superficial. The system evaluation by a plastic surgeon using this system revealed that it was desirable to be able to grasp shape information more quickly. Therefore, the purpose of this study was to develop a system that calculates the shape difference information in real time based on the left and right breast shapes. As a system evaluation, the shape difference during reconstruction was reproduced using a mannequin simulating a patient, and measurement and projection experiments were performed. It is now possible to project the shape difference information onto the skin surface without the process of pasting the shape difference texture onto the three-dimensional shape data.

We conducted electrical train stimulation of the retina before and after transcranial direct current stimulation (tDCS) to examine the effects of tDCS on retinal prostheses. In anesthetized rats, the dorsal surface of the right visual cortex (VC) was exposed and covered with a conductive gel. For the tDCS, anodal electrical current was applied for 10 min through the VC via the conductive gel. A train of 100 biphasic electrical pulses (0.5 ms, 20 Hz) was applied to the left retina before and after tDCS application, and the electrical evoked potentials (EEPs) from the right VC were recorded. EPP amplitude was defined as the difference between the first negative response peak after each electrical stimulation and its difference with the subsequent first positive peak. EEP amplitude significantly decreased after tDCS compared with before tDCS. Decrease in EEP suggests that tDCS increases the current stimulation thresholds to the retina.

An artificial hip joint is a medical implant which used for total hip arthroplasty. There are several issues for the artificial hip joint. For example, dislocation is one of crucial issues. We proposed a new artificial hip joint with the structure for preventing dislocation. Dislocation dose not easily occur due to the artificial femoral head covered with the acetabular cup over the half sphere line. In this study, the pull-out forces were estimated using finite element analysis. A finite element model was constructed by software (ANSYS 17.2). For analytical condition, the acetabular cup was restricted and the femoral head was pulled out from the acetabular cup with various pull-out directions (0, 30, 60, 90 degrees).In the case of 90 degrees, the pull-out force was lower than that in case of other pull-out directions. From this result, we obtained a design guideline for the artificial hip joint.

We have developed a new magnetic fluid seal for an miniature axial flow blood pump that can be placed by catheterization. The new magnetic fluid seal consists of a ring-shaped neodymium magnet, iron ring and magnetic fluid. The size of the magnetic fluid seal is an outer diameter of 4 mm and length of 3 mm. The electromagnetic field analysis shows that the magnetic fluid seal has an estimated pressure-resistant of 432 mmHg. We examined its durability by an in-vitro experiment. The magnetic fluid seal has been sealing the motor from water against a water pressure of 110 mmHg at a motor rotational speed of 8000 rpm for over 120 days, and the durability test is going on. From the results, the new magnetic flow seal is a promising non-contact seal to accomplish a catheter-type miniature axial flow blood pump for long-term usage.

The main purpose of this study is to develop performance evaluation techniques that can evaluate the hydrodynamic characteristics of pediatric pulmonary heart valves. In this study, aperture ratios for artificial valves were evaluated using image processing. An ePTFE conduit with the bicuspid valve (14 mm) was fixed to a mock circulation system and was evaluated under hydrodynamic conditions found in pediatric pulmonary circulation (HR = 120 bpm, systolic ratio = 35 %, pulmonary artery pressure = 20-7 mmHg, flow rate = 700 mL/min). A biological valve (19 mm, Magna, Carpentier-Edwards) was used as a control valve. The valve leaflet movement was captured from the downstream side and the aperture ratios were calculated by image processing. As results, maximum aperture ratios of the bicuspid valve and the biological valve were 38.9±0.4 % and 31.6±0.7 %, respectively. This study suggested that it is possible to evaluate the performance of pediatric pulmonary heart valves using image processing.

This paper describes prediction method of thrombus formation at the wall on orifice flows by using particle method (DPD) with aggregation of platelets and finite difference method (FDM) for transport process of concentration. By the hybrid model of the particle method and FDM, the prediction of thrombus formation rate was compared with the experimental data conducted in the previous investigation. In this method, the total number of particles in the DPD can be decreased by using novel technique of virtual particle method. As a result, the effects of aggregation between the platelet in the CFD are not large comparing with the data in the previous experiments. It was also found that there is a possibility to improve the accuracy by modifying the model of virtual particle method.

Hemolysis is one of the major blood trauma in extracorporeal circulation. The occlusion is a parameter to eliminate deteriorative blood trauma as well as to improve assisted circulation efficiency during cardiopulmonary bypass (CPB). A new occlusion simulation model was developed to evaluate the occlusion by the thickness of the flow in the tubing compressed with the CPB roller and raceway. The red-coloured densimetry images were used to calculate the thickness of occluded flow channel. In this study, we examined two types of occlusion measurements and compared the thickness of the flow in the model: a) the coloured densimetry, and b) a 3D measurement of the casted silicone. As a result, the cross-section depths showed a linear relation with the thickness lower than 120 microns. Therefore, the colour densimetry occlusion measurement model might be useful for the quantitative evaluation for CPB occlusion by the 3D reconstruction of the squeezed flow layers.

Artificial lung is a medical device for gas exchange composed of hollow fibers (HF). However, the HF membrane caused a plasma leakage attributed to pores on the membrane. To solve this problem, we suggest that the combination of homogeneous membrane and cross-linked 2-Methacryloyloxyethylphosphorylcholine (MPC) polymer. But there is no report in a paper on artificial lungs as the effect of the coating on gas permeability is unknown.The purpose of this study is to achieve both oxygen permeability and blood compatibility when a silicone surface is coated to a crosslinked phospholipid polymer. First, the effect of the MPC coated of silicone on oxygen permeability was evaluated. For the oxygen permeability test, an electrochemical method used. Sylgard184, one of the silicones, was confirmed to have an oxygen permeability coefficient of 358 barrer.

Silicone elastomer (SE) materials are commonly used for biomedical devices. However, unwanted reactions including immune cell adhesion and blood clotting occur easily on the unmodified SE surface because the hydrophobic property induces non-specific protein adsorption. Previously, the coated of phosphorycholine-based crosslinking-type polymers (cross-MPC copolymer) on SE can reduce protein adsorption successfully. In this study, the SE plates and tubes were coated with the cross-MPC copolymer by alkaline treatment as a pre-treatment. The coated-polymer membrane on SE were evaluated with fluorescence of Rhodamine 6G, hydrophilicity evaluation, and protein adsorption. For the hemocompatibility evaluation of polymer-coated SE, the polymer coated SE tubes and SE tubes were circulating with pig blood for 1 hour at 37 Celsius degree in fluidic condition (shear stress = 2782 mPa) with the loop-system. As the results, the copolymer was coated successfully on SE with alkaline treatment to reduce the protein adsorption and prevent the blood clotting.

In this study, recirculation ratio based on venous pressure was evaluated.

Hemodialysis was performed at blood flow rate: 200 mL/min, venous pressure thresholds: 250 mmHg (high) and 100 mmHg (low) with a 15-, 16- or 17-G needle using urea-added bovine blood. The recirculation ratio evaluated from the venous pressure was compared with that calculated by the urea dilution method (stop flow-stop pump technique).

In the blood return phase, an initial sharp drop in venous pressure was observed with a subsequent gradual pressure decrease. This initial sharp drop was considered due to recirculation (suction) of purified blood. This trend was prominent with a 17-G needle. Next, we evaluated the recirculation ratio based on the initial decrease of venous pressure and the urea dilution method. Both of the recirculation ratios showed similar values with a similar dependency on the needle gauge.

The aim of this work was to study usefulness of needle-free injection as a device for regenerative treatment of the kidney. In this experiment, we evaluated the penetration and diffusion of the injection solution by injecting and observing the fluorescent paint into the pig kidney with needle-free injection. In addition, we injected the solution containing the contrast agent into the kidney, and observed by CT and MRI. The results showed that drugs can be injected directly into the kidney by needle-free injection and differences in the penetration and diffusion of the solution into the kidney due to the difference in the spring force of the needle-free injection. Additionally, it was also confirmed that the influence of the structure inside the kidney. Thus, we concluded that these characteristics of needle-free injection have created expectations for new organ treatment possibilities.

The purpose of this study is to examine the accuracy of sleep scores with a single channel EEG signal between two trained technicians and, our developed automatic scoring. Polysomnography (PSG) is the gold standard for sleep assessment in medical and sleep scientific fields. Recently, it is expected a small number of the EEG signal can estimate accurate sleep scores for screening. In the next stage of sleep study, quantitative evaluation of the accuracy for sleep scoring should be verified for reduction between the technicians or automatic scoring. The single EEG channel (C4-A1) from each PSG data set of ten young male subjects were scored. Our results shows the corresponding ratio 83.1% between the technicians and the automatic algorithm has almost near the accuracy are presented. The fact supports that single EEG channel recording may have the sufficient ability to estimate sleep scores for screening sleep assessment.

Melanoma is a type of skin cancer. The diagnosis of melanoma depends on the subjectivity of a doctor. Therefore, a quantitative diagnosis method is required. We have developed a melanoma diagnosis system using a hyperspectral imager (HSI). However, HSI tends to be larger and more expensive than a commercially available digital camera. By selecting some wavelengths suitable for melanoma diagnosis from among the wavelengths measured by the HSI, we think it will result in simplification and cost reduction of HSI. A spectral angle map for the three wavelengths is calculated from data measured by HSI. Then, entropy and some image features are calculated from a spectral angle map. Analysis of a diagnostic performance from the entropy and image features showed that a diagnostic rate of 77.8% was achieved with a combination of 500 nm, 620 nm, and 710 nm.

Melanoma is a type of superficial tumor and should be treated in an early stage. Early-stage melanoma is difficult to diagnose because it looks like a benign lesion. However, melanoma is still subjectively diagnosed by a dermatologist. Therefore, there is strong need for development of a quantitative diagnostic method. We are developing an automatic melanoma diagnostic system using convolutional neural network and hyperspectral imager. Hyperspectral imager acquires position and wavelength information simultaneously. 3D patches (16x16x201) extracted from hyperspectral data (560x680x210) is input to our newly proposed network. The total number of hyperspectral data was 202, including 88 melanomas and 114 non-melanomas. Our dataset was evaluated by 5-fold cross-validation and calculated sensitivity, specificity and accuracy. In this paper, the result of training and validation using patch data is reported.

Melanoma is a type of superficial tumor. Melanoma is known to have a poor prognosis, as it accounts for 80% of skin cancer-related deaths. On the other hand, early melanoma is likely to be completely cured. Therefore, early detection of melanoma is essential. Even now, melanoma is visually diagnosed by a doctor, and thus a more accurate and quantitative diagnosis method is required. We develop a melanoma diagnostic system that combines a device called hyperspectral imager and deep learning. The data measured by the device must be reduced to a format that can be trained with deep learning. In this study, principal component analysis is adopted as a method of reduction. In this paper, the results of melanoma discrimination by deep learning trained with principal component score images are reported.

The number of rheumatoid arthritis (RA) patients is about 700,000 in Japan. The modified total sharp score (mTSS) calculated from hand X-ray image is a standard diagnosis method of RA progression but can be time-consuming for physicians. So, improving the diagnostic quality of RA patients requires a computer-aided diagnostic (CAD) system. We have previously proposed a CAD system, which detects finger joint positions using support vector machine, and estimates mTSS using support vector regression. This study improves the finger joint detection accuracy by introducing statistical shape model, which statistically models individual variety of spatial relationship among finger joints and fingertips. And, improving the mTSS estimation accuracy, we introduce a data cleansing process based on an image clustering method. Experimental results on radiographic images of the hands of 90 RA patients showed that the finger joints were detected with an accuracy of 94.5%.And the accuracy of the mTSS was also improved.

This study proposes an automatic puncture robot into artificial blood vessel with features of automatic selection and 3-dimensional position estimation of blood vessel.Selection of blood vessel is achieved by using the length, linearity and thickness of blood vessel image. Three-dimensional position of blood vessel is estimated from a 3D linear equation which is obtained from an intersection of two 3D planes through 2D image of blood vessel.Puncture robot is constructed using one slide mechanism,four RC servo motors and one linear actuator.The robot has a 4-degree-of-freedom arm to control the position and direction of needle.Linear actuator advances the needle straight to the artificial blood vessel.Artificial skin with blood vessel is manufactured so that the camera image of it is similar to the human skin.Experimental results show that the automatic puncture is confirmed because reverse blood is observed by pulling the syringe.

In artificial dialysis, an arteriovenous fistula (AVF) is used often. Since stenosis and occlusion in the AVF are common in repeated treatment, the frequent check of the AVF is indispensable. We have proposed an optical transillumination imaging technique to visualize the AVF, and confirmed its possibility to detect the stenosis by evaluating the inner diameter of the AVF. In this confirmation, the direction of the blood vessel was assumed to be parallel to the longitudinal axis of the forearm. This is not always the case in the human arm. Therefore, we constructed a model phantom which simulates AVF in an adult forearm and analyzed the effect of the AVF angle on the measurement accuracy. The results show that the measurement error increases with the angle. However, the error in the measured AVF diameter was within 0.4 mm in clinical conditions. This suggests the feasibility of our technique in clinical application.

Atherosclerosis is one of the major risk factors of cardiovascular disease. Carotid ultrasonography is widely used to diagnose the atherosclerosis. The authors are examining the applicability of a ultrasonic doppler blood velocimeter as a simple screening device. However, the relationship between the atherosclerosis and blood flow velocity waveforms (BFVW) has not been fully elucidated. In this study, we investigated whether BFVW in common carotid artery was related to the degree of arteriosclerosis in the patients(N=83). The intima-media thickness and plaque index obtained by ultrasonography were used as the degree of atherosclerosis. Two types of BFVW in common carotid artery were determined by the k-means clustering from results of principal component analysis of measured values of ultrasonography. As results of them, there was a significant difference between BFVW types in each index of atherosclerosis. These results suggest that the degree of atherosclerosis can be evaluated by BFVW.

In some diseases, serum protein concentration ratios can cause characteristic abnormalities. In other words, a specific disease state can be estimated from the concentration ratio of proteins fractionated by electrophoresis. Proteins can be visualized by staining. However, stain solution is harmful to the human body. Therefore, in this research, we are studying a visualization method using terahertz (THz) waves that are harmless to the human body and highly absorbable to proteins. It has been reported that THz waves can be used to visualize "proteins of human serum" fractionated by cellulose acetate membrane electrophoresis. However, the contrast of the visualized image was low and the location of the protein was unclear. In this paper, we examined the processing conditions required to prepare a sample for electrophoresis to acquire THz transmission images. In addition, we examined a method for extracting protein features from THz transmission images of human serum.

The purpose of this study was the fabrication of the near-Infrared camera system with the thermal stimulation system using Peltier device for evaluating the risk of a heat shock while bathing. In the fabrication of the system, Peltier device was attached to a heat sink with a small electric fan. In the biometric measurement, the thermal stimulation system was attached to a hand of subject and blood vessel of the finger was taken a picture with a near infrared camera and a near infrared light source as a transmissive image during thermal stimulation. In the results, the vessel expanded or constricted depending on the heat stimulation. But, the vascular response was different by subjects. This result suggests the possibility of the quantitative evaluation of the risk of heat shock and the detailed study will need.

Sweating have a central function in body temperature control. But the abnormal secretion of sweat causes the sweat stain to document and the equipment break and decreases the incentive to work. Although there are various approach such as surgery and medication, these treatment method have problems such as side effect, invasiveness and the compensated sudation. We focus on the psychological therapy without side effect and invasiveness and set as an ultimate goal to fabricate the biofeedback treatment system. In this study, the humidity sensor was attached to a hand via self-produced jig and temporal change of humidity was measured. In the results, the humidity increase with time and the gradient of humidity change was different by subjects. This result suggests that the humidity would be available as a sweat indicator.

The quantitative detection of the magnetic nanoparticles (MNPs) in sentinel lymph nodes (SLNs) can help to diagnose cancer metastasis for breast cancer patients because the SLN (i.e., 1st SLN) containing the largest amount of the MNPs has the metastasis with the highest probability. In this study, we have optimized the frequency of alternating current (AC) magnetic fields under the application of direct current (DC) magnetic fields to achieve the highly-sensitive detection, and demonstrated the intraoperative detection of the MNPs accumulated into the SLNs to reduce the injection amount. Considering the noise signal and hysteresis loss, AC fields with 2-5 kHz was effective for the detection. The clinical trials with breast cancer patients, for the MNP detection using a developed device, revealed that the MNP amount of 1st SLN was ~40 ug. We will pursue the further detailed magnetic characteristics of the device and the MNPs for biomedical applications.

In stage 1 5-year survival rate of breast cancer is more than 90%. So, early detection is important for breast cancer, and it is effective to have regular examination for breast cancer.But, inspect accuracy depends on inspector skills because it is difficult to keep the probe vertical. And, the size and position of cancer is unclear.The aim of our study is to keep the probe by the auxiliary equipment.Here we describe operation verification of auxiliary equipment. In the guidelines of breast cancer screening, it needs to find minimum 5mm object on ultrasound image. The goal of this study was to detect object of minimum 5mm. Next, we consider image acquisition method. When auxiliary equipment scans, the phantom was deformed by the probe. We suggest new scanning method. It can prevent to deform the phantom and to obtain ultrasound image. We describe the result of operation verification.

Molecular-targeted anticancer drugs elicit less toxicity than conventional reagents. Yet, patients often suffer from severe adverse effects. The reason is "fixed dosage" administration of the drug to all the patients regardless of their body size and complications; because of this strategy, the plasma concentration seems to occasionally exceed the therapeutic window. Although frequent measurement of the drug level at a clinical site is a solution, currently available methods are unsuitable. To overcome this shortcoming, in this study we developed a procedure with an electrochemical sensor composed of conductive diamond, which yields more stable reactions than conventional materials. When guinea-pig plasma mixed with imatinib, lenvatinib, or pazopanib were tested, the sensor detected a clinically relevant concentration. Time and sample amount necessary for each series of the measurement was <1 min and 100 μL, respectively. This method may enable therapeutic drug monitoring of molecular-targeted drugs and accelerate tailored medicine for cancer.

The vein valve disorder causes the varicose and pulmonary enbolism.It is reported that about 9% Japanese people have varicose veins of leg which needs surgical treatment.Therefore, we diagnosed functional characterization of the vein using ultrasound non invasively. And, it is good for prevention of varicose.The venous pressure changes by changing the height of the leg by Bernoulli's theorem, and changes the vein cross section.After we lifted leg and, recorded a cross section change of the vein for 30 seconds by ultrasound echogram.We made a filter for detecting the veinous vessel walls of the great saphenous vein of the recorded echogram and conducted mask processing. Thereby, we measured a vein cross section change semiautomatically.We make indexes to evaluate venous characteristics by using these indexes.Then we approximate expression will examine the relationship between there indexes and the risk factor of varix.

We quantitatively evaluated the risk of skin photosensitivity after talaporfin sodium injection in photodynamic therapy by fluorescence measurement and in silico pharmacokinetics model. We performed a clinical study of drug fluorescence measurement in skin tissue inside the upper arm using a transdermal drug fluorescence measurement system. Then we established a model of pharmacokinetics in the skin tissue. A pharmacokinetic model consisting of three compartments of plasma-interstitial space-cells was constructed, and the drug concentration in skin tissue was calculated by defining the volume contribution of each compartment in skin tissue. The rate constant between compartments was optimized by the conjugate gradient method to minimize the difference between the calculated drug concentration in skin tissue and the fluorescence measurement data of clinical research, and the change in drug concentration in human skin tissue was estimated.

Purpose: We investigated the effect of wearable ECG textile (Techno sensor ER, Teijin Frontier) on the uses of chest radiograph.Methods: We obtained 10 X-ray images of chest phantom (Kyoto Kagaku) with and without Techno sensor ER. Each image were averaged, and the image without Techno sensor ER was subtracted from the image with Techno sensor ER. Pixel values of electrode on lung, cardiac and mediastinum on subtracted image were measured. Effect on diagnosis and visibility of Techno sensor ER were evaluated using a 5-point scale (1 = bad effect for diagnosis, 5 = not visible) by 12 medical doctors on chest radiograph with Techno sensor ER.Result: The pixel values of Techno sensor ER were very low. The Techno sensor ER on lung was barely recognizable and them on mediastinum and cardiac were not recognizable.Conclusion: Techno sensor ER does not have a negative effect on the chest radiograph.

In the previous study, the method of extracting brain regions from the MR brain image was proposed. That method uses Snake to detect the position of the scalp and remove scalp and skull automatically. In this study, we propose the method for automatic placement of initial contour of snake. At first, the MR brain image is divided into several small areas. Each area could be classified into the area with no object, and the area with brain parenchyma, and Composite area by considering the average and the standard deviation of histogram of the image. The position of the contour of snake can be initialized automatically by place it in the area with no object. As a future, we will apply the proposing method of placing the initial contour of Snake to several MR head images and estimate the effect by comparing with the conventional manual methods.

Interstitial pneumonia (IP) is a noninfectious inflammatory disease with progressive fibrosis in the lung and is frequently associated with lung cancer. Early detection of lung cancer with IP is difficult. Deep learning (DL) excels in the ability to extract useful features. We aim to extract image features associated with lung cancer from CT images of IP using DL. As a basic study, the purpose of this study is to extract the features by DL, assuming that there are invisible differences between the non-cancerous lung region of the carcinogenic lung and the non-carcinogenic lung region.Image ROIs with either carcinogenic lung or non-carcinogenic lung were created from CT images, and DL was learned to classify them. An accuracy of 57% was obtained. And the accuracy for the training data was about 80%, indicating overfitting. However, lung cancer often occurs bilaterally, the possibility of both lungs had the same features cannot deny.

The aim of this study is to investigate whether fatigue state affect brain activities in healthy people or not. Fifteen healthy subjects were divided into fatigue and normal groups according to the profile of mood states (POMS) fatigue sub-scale. EEG was measured under speech task as acute stress stimulus and analyzed with a discrete Fourier transform (DFT) to obtain power spectra density in the theta, alpha, and beta frequency bands. Analysis of variance (ANOVA) was performed for each frequency band on the between-subjects factor "group" and within-subject factors "epoch" and "cluster". The results showed that the brain activity in the alpha frequency band of the fatigue group was significantly lower in pre-stimulus state and 120-180 second after speech task than those of the normal group. These results suggest that fatigue state may affect the performance of sustained brain information processing in healthy people.

The working memory is necessary important work for a person in living. Previous study shown the relation between the quantity of experiment of playing musical instruments and working memory. But it was not clear which musical skills relate working memory. We measured the relation between the ability of musical skills and working memory. As a result, we could find the strong relation between some skills (sense of absolute pitch, relative pitch hearing, and solfege) and working memory. In our future study, we will increase the number of subjects to make clear the relation between musical skills and working memory more accurately.

So far, we investigated the relationship between sleepiness and amplitude of amplitude-modulation following response (AMFR) to find the best tone for waking up a drowsy driver. In this study, we recorded AMFRs evoked by amplitude modulated (AM) tones. In this study, we used AM tone whose carrier frequencies (FC) were 250, 500, 1000, 4000, and 8000 Hz. Sleepiness level was divided into three states, low sleepiness, high sleepiness, and sleep states, by using a reaction time of vigilance task. We found that AMFR evoked by AM tones whose frequency was above 1000 Hz AM tone in high sleepiness and sleep states were small that in low sleepiness, but AMFR evoked by 250 and 500 Hz AM tone did not decrease during high drowsiness state. These results suggest that the low-frequency tone is suitable for drowsy drivers to wake up.

In recent years, it is known that the pupillary light reflex occurs to the bilateral pupil when light stimulation is applied only to either pupil, but there remains the argument about the light reflex when distinct light stimulation presented to the left and right pupils. In this study, we observe the pupil responses when visual stimuli of 1.2 Hz are presented to the left eye and 1.5 Hz to the right eye for 10 participants. As a result of the frequency analysis of the pupil fluctuation, it is clarified that the peaks of the spectrum are detected in the same two frequency bands as the visual stimulation, and the time variations of the bilateral pupil are observed as the superposition. This result suggests the possibility that a pupillary light reflex treats the light stimulus to the left and right pupils as equals.

Visually induced motion sickness (VIMS) is a phenomenon that causessymptoms similar to motion sickness that occur during video viewing.Various explanations have been proposed for the cause of VIMS, but theexact reason is still an open problem. We focused on characteristicsof eye movement during video viewing as one of the causes of videosickness, and measured eye movements during viewing to videos withdifferent spatial frequencies. The degree of sickness was measuredusing SSQ (Simulator Sickness Questionnaire). As a result of eyemovement analysis, subjects with a high SSQ score tended to have a bigeye movements. On the other hand, eye movement distance was not largein subjects with low SSQ score. The phenomenon was remarkable in videothat seemed to have a high spatial frequency. In the future, we planto increase the number of subjects and conduct more precise experiments.

Elucidation of muscle activity is important in clinical fields such as diagnosis of sarcopenia.Muscular activity is regulated by types and number of motor units(MU). The action potential generated by MUs are measured as electromyography(EMG).The multi-channel surface EMG(sEMG) is used to analyze the conduction of muscle potential. It is recorded as the interference electric potential generated by MUs. In previous study, we proposed a method for searching all conducting waves appearing quantitatively in the multi-channel sEMG by calculating the similarity of the waveforms. It has possibility to estimate the mobilization of MUs by examining conduction waves. In this study, we investigated the feasibility of the motion analysis using conduction waves by examining the relationship between the acquired conduction waves and the sEMG with different muscle states and motion conditions using this method.By analyzing conduction waves, we could confirm the change of sEMG due to the movement state.

We adopted diffuse correlation spectroscopy (DCS) to measure blood flow velocity of vastus lateralis during cycling exercise. Participants performed a cycling exercise whose load was increased by 40 W every 2 min from 40 W to 160 W (active condition). They also performed passive pedaling exercise to verify motion artifact (passive condition). A 10 s interval was inserted following each exercise period. The average values of blood flow velocity during exercise and interval periods were compared among different exercise loads. The blood flow velocity during the exercise period was larger during active condition than passive condition. The blood flow velocity during the interval period increased with the preceding exercise load in active condition but not in passive condition, suggesting the load-dependent increase in the blood flow with active cycling exercise. Our results suggest a potential of DCS to measure active muscle blood flow during cycling exercise.

Imaging-photoplethysmography (iPPG) is a non-contact technique to monitor blood volume pulse (BVP) via detecting tiny color changes in skin caused by blood circulation using a CCD camera. The most basic approach is to extract BVP from only green channel, however such method can be easily affected by ambient light noise and motion artifacts. In this study, we focus on improving the robustness of extracting the BVP from CCD camera, thereby accurately measuring beat-to-beat intervals for calculating heart rate variability (HRV) indices. We proposed a method to reconstruct BVP waveform using all RGB channels to optimize a linear function to improve the signal-to-noise ratio based on the cardiac physiological properties. The proposed method was evaluated on HRV indices with 4 healthy subjects comparing among the conventional green channel alone and the ground truth ECG. The results show that the proposed method is more effective for HRV measurement than the conventional method.

Optical mapping method is a powerful tool for observing the cardiac excitation wave propagation. In this method, generally the measurement of action potential is performed on the assumption that the same pixel is continuously reflected on the same myocardium, but cardiac muscle must contract with depolarization. In general measurement, the measurement of myocardial movement was stopped using BDM or Cyto-D, one of the muscle contraction inhibitors, to suppress pulsation artifacts. However, BDM has been shown to affect cardiac electrophysiology such as shortening action potential duration and slow conduction velocity.In this study, we developed a system that visualization cardiac excitation propagation without BDM using offline image processing with Non-rigid registration and affine transform.

Judgment of sleep stage is not easy because it requires measurement of multiple biological signals and expert knowledge. The purpose of this study was estimating the sleep stage using an Electrocardiogram (ECG) that can be easily measured. In this study, we investigated whether Detrended Fluctuation Analysis (DFA) was effective as a parameter for estimating sleep stage. We applied the result of DFA as a parameter for estimation. In addition to the results of the DFA, sleep stage was estimated using a low-pass filter, the trend direction and a decision-tree. As the result, 63.7% accuracy was achieved. In this time, in order to verify whether DFA is an effective parameter for estimating sleep stage, sleep stage was estimated using only the results of DFA. In the future, we aim to further improve the accuracy by combining with various parameters.

Distance to a major reflection site, effective reflection distance (ERD), is associated with aging and augmentation index (AIx). However, it remains unclear whether and how ERD varies and associates with AIx when hemodynamics is acutely perturbed. In 13 anesthetized dogs, we measured arterial pressure, aortic flow, while altering the hemodynamic condition over wide ranges by administering a variety of cardiovascular drugs. Using measured data, we determined ERD based on an arterial model comprising a tube with Wind-Kessel (ERDtl), a physiologically valid model. We also determined ERD based on wave separation (ERDwsa) and pressure-based analyses (ERDaw). ERDtl was shortened significantly in response to vasodilator or inotrope administration, and was significantly and negatively associated with AIx in multiple regression analysis. ERDwsa or ERDaw did not necessarily correlate with ERDtl in terms of drug responses or association with AIx. Present results should be kept in mind in analysing ERD and AIx.

We proposed the new adaptive digital filter for high accuracy detection of the R-peak in electrocardiogram (ECG) derived from head. The proposed adaptive filter is used to estimate a noise waveform in difference between ECG during rest and ECG with noise. The estimated noise waveform is updated by 50% of noise derived from the latest beat and 50% of the noise derived from accumulated beat. The updated noise is devoted to ECG for detection of R-peak in next beat. In 2 healthy male participants, stationary state and 40, 50 W loads exercise using cycle ergometer for 30 seconds was performed, respectively. The results indicate that R-peak detection ratio for filtered ECG using proposed filter was increased (95.1% and 80.9% from 88.3% and 78.7%). In conclusion, these findings suggested that the proposed adaptive filter might be practical signal processing for detection of R-peak detect in ECG derived from head.

We analyzed the heart rate (autonomic nervous system) response to sleep apnea events. We compared the response between sleep apnea syndrome (SAS) patients with Parkinson's disease (PD) patients and SAS patients without PD. The increasing amplitude was significantly smaller in SAS patients with PD comparing with that in the case of SAS patients without PD. The increasing latency was significantly longer in SAS patients with PD compared with that in the case of SAS patients without PD. We applied logistic discriminant analysis using the heart rate response indices to discriminate the two patient groups. The result showed relatively high accuracy for discrimination. These results imply that autonomic nervous system does not response rapidly and efficiently to sleep apnea events in PD patients due to the sympathetic degeneration.

Sleep disordered breathing (SDB) has been known to reveal cyclic changes in heart rate in association with apnea/hypopnea, so called cyclic variation of the heart rate (CVHR). We have developed an algorithm for detecting such CVHR from 24-hour Holter electrocardiogram. However, it is difficult to distinguish the change in heart rate whose magnitude is too small to detect as a CVHR from noise. In this study, we focused on the amplitude ratio of the fundamental frequency to the lower and higher frequency area in RR interval trend. We evaluated the new algorithm using 70 data of the Apnea-ECG Database (PhysioNet) and found that (1) the correlation coefficient between AHI and CVHR frequency improved from 0.71 to 0.82, (2) sensitivity and specificity to detect CVHR improved from 86% to 91% and 86% to 93%, respectively. These results suggest that the new algorithm is useful to detect severity of SDB by CVHR.