An arteriovenous (AV) fistula is commonly constructed in patients who need dialysis treatment. Daily monitoring of the AV fistula is crucially important to maintain its correct function. We developed a new technique using a safe and simple instrument for noninvasive monitoring of the AV fistula. Using this technique, we can observe and analyze the blood vessel condition using transillumination imaging with near-infrared (NIR) light. For an AV fistula in the forearm, we illuminate NIR light from both sides of the arm. Then the transillumination image is captured using a camera placed above the forearm. An imaging system was developed to validate the proposed technique. It was tested with model phantoms that simulate the blood vessels of the human adult forearm. Severe blurring of the image caused by strong scattering in the interstitial tissue can be eliminated by deconvolution with a point spread function of the blur. This process clarified the blood vessel image and facilitated measurement of its inner diameter. After confirming the accuracy of the system, we attempted measurement of localized stenosis. Results show that we can detect stenotic portions as small as 2 mm in diameter and 2 mm in length. The detectability of non-coaxial stenosis was also confirmed. These results verified the feasibility and applicability of the proposed technique for non-invasive daily monitoring of AV fistulae.

Hiesho is the condition of having a cold sensation in one's hands or feet. This is a well-known health problem for young Asian females. However, the definition of Hiesho is still controversial. In this study, we aimed to develop a quantitative and non-invasive approach to determine Hiesho. Sixty-three young females participated in this research. Temperature difference (ΔT) between the forehead and foot sole was utilized to define Hiesho or non-Hiesho condition, and the result was crosschecked with that of a self-reported questionnaire. Central systolic blood pressure and augmentation index were measured to evaluate subjects' physiological indicators. The results of the questionnaire showed that 49% of young females (31 of 63 people) reported Hiesho. There was a significant difference in ΔT between non-Hiesho and Hiesho (1.85℃ and 5.55℃, respectively, p < 0.01). After cross-checking with the self-reported questionnaire, ΔT of 3.64℃ demonstrated acceptable reliability and accuracy for defining Hiesho. Central systolic blood pressure and augmentation index were not different between Hiesho and non-Hiesho. In conclusion, young females with Hiesho had drastically different temperatures at the forehead and foot sole. The temperature difference between the forehead and foot could be used as a quantitative and objective parameter for defining Hiesho.

To prevent aspiration pneumonia, a system for non-invasive and quantitative evaluation of the swallowing function is required. Therefore, we have previously proposed a method of using ultrasound videos to establish evaluation indicators of the swallowing function. The proposed method aims to automatically estimate the velocities of the esophageal wall region and the bolus region in the ultrasound video. In this method, estimation of the bolus region comprises two steps: estimating the esophageal region through which the bolus flows and extracting only the frame in which the bolus passes through the esophageal region. However, the step of extracting the frame in which the bolus passes is still performed manually. Therefore, to automate this step, the purpose of this study was to automatically determine the frame in which the bolus flowed into the screen. This method was tested five times on five healthy adult male subjects by recording a cervical ultrasound video while swallowing a bolus of water. We identified the different elements of the esophageal region in the video by first identifying the esophageal wall region with the maximally stable extremal regions (MSER). Then, we used the luminance histogram of each frame to establish the graph of the histogram similarity. This, in turn, was used to detect a change in the observed region, thus indicating the inflow of the bolus. Moreover, we could distinguish the change caused by the inflow from the change caused by the elevation of the esophageal wall using the velocity results obtained by optical flow estimation in the anterior esophageal wall region. Our results showed that in most cases, the proposed method was successful in recognizing the inflow of the bolus and distinguishing it from the elevation of the esophageal wall. Furthermore, an accuracy sufficient for estimation of the velocity of the bolus was achieved.

The rhythm of vocalizing a written language depends on a merge process that combines meaningless linguistic units into a meaningful lexical unit, word, or Bunsetsu in Japanese. However, in most previous studies, written language was presented to the participants in lexical units (word-by-word) with explicit inter-word (or inter-Bunsetsu) marks or spacing. Therefore, it has been difficult to conduct psychophysical assessment of the participants' own speed in segmenting meaningful units from unstructured written language when reading. Here, we hypothesized that the spontaneous reading speed of Japanese readers reflects their own punctuation process, even when sentences are written without punctuation marks or spaces. To test this hypothesis, we developed a new “self-paced sequential letterstring reading task,” which visually presents sentences letter-by-letter. The task required participants to push a button to proceed to the next letter at their own pace, hence allowing evaluation of the reaction time (RT) to individual letters. We found that the average RT decreased parametrically as the position of the letter approached the end of a Bunsetsu. Moreover, the RT increased drastically at the last letter completing the Bunsetsu. Participants were not shown any punctuation marks and not instructed to explicitly recognize the punctuations during reading. Therefore, these effects strongly suggest that the implicit and spontaneous punctuation is the origin of the rhythm in reading. These results show that spontaneous punctuation of letterstring affects the reading speed. The task we have developed is a promising tool for revealing the temporal dynamics of natural reading, which opens a way to shape the fluency of script-to-speech human interfaces.

Watching television (TV) is a popular leisure activity for older people and its usage is almost constant in the rhythm of their daily life. A telemonitoring system of television-operating-state for older people living alone has been developed, and an older female living alone has been using this system for 10 years. On her working days, the average TV usage was 192 min per day in the first year, increasing year by year to 329 min per day 10 years later. TV usage temporarily increased when some TV programs took up a social topic of interest to her.

To facilitate capturing the characteristic variations of blood pressure (BP) such as BP surges during the nocturnal period, the in-bed continuous daily measurement of BP may be useful. In this study, we proposed and evaluated a method for cuffless continuous estimation of relative mean arterial pressure (MAP) using capacitive ballistocardiogram (cBCG) and electrocardiogram (cECG) measured using an unrestrained and noncontact method. We adapted a well-known equation for calculating MAP, which is equal to the product of cardiac output and peripheral vascular resistance. We then derived an estimation formula for the relative MAP using the J–K amplitude from the cBCG, heart rate from the cECG, and pulse beat arrival time calculated from the cBCG and cECG. To determine the coefficients for the estimation formula, we measured the MAP of a subject with a commercial device and used the least squares method. To obtain input data for the estimation formula, the cBCG from the heel of the right leg and cECG from the back of the subject were measured simultaneously with capacitively coupled electrodes placed under a bed sheet. The total length of the input data was 80 s for each Valsalva test (VT), and the Valsalva maneuver was used to increase BP during measurement. The data for each VT was separated into a training segment (Tr) and a test segment (Te). To evaluate the proposed estimation method, the following indices were calculated for each VT in 7 subjects: (1) correlation coefficient (CC) between estimated and reference MAP values, (2) confidence interval (CI), and (3) root mean square error (RMSE). For the Tr, average CC was 0.93 ± 0.06, average CI was 2.96 ± 1.29 mmHg, and average RMSE was 0.75 ± 0.33 mmHg. Furthermore, average RMSE for the Te was 2.49 ± 2.22 mmHg. These results indicate that the continuous cuffless method proposed in this study can be used for estimating relative MAP over a short time period. As the subjects in this study were all men in their early twenties, further validation in diverse subjects is required for broad application of the proposed method.

Knee osteoarthritis (KOA) is a common joint disease of the lower limbs. Its progression reduces the patients' quality of life. Varus thrust (VT), one of the abnormal gait patterns of KOA patients, is considered as an effective index for assessing KOA. Hence, several studies have assessed VT using various measurement methods. Since VT is the momentary lateral knee motion that increases knee varus angle and moment, optical motion capture system is widely used. However, optical motion capture system has some disadvantages in clinical usage, such as high cost, requirement of technical skills, and time-consuming attachment process. Recently, inertial measurement units (IMU) have emerged as a measurement system instead of optical motion capture systems. IMU-based method is regarded as more suitable than optical motion capture systems for VT assessment because of its simplicity. This study aimed to assess the gait of KOA patients using IMUs and to quantitatively evaluate VT based on disease progression. For this purpose, we recruited 7 healthy participants and 15 KOA patients. Subsequently, their knees were classified into 3 progression groups: 14 healthy (grade 0) knees, 9 grade 3 knees, and 14 grade 4 knees. As a gait test, all the participants wore IMUs positioned at the trunk, both thighs, and both shanks, and traversed a 10-m walkway. VT was considered as the first peak value of the mediolateral acceleration and the varus–valgus angular velocity data, which were collected via the IMUs at both shanks. Thereafter, these acquired data were compared and evaluated among the three progression groups. The results indicate that both peak values were significantly greater in the KOA patients than in the healthy subjects. Moreover, there was a significant positive correlation between the two peak values. Thus, this study is expected to contribute toward early detection of KOA.

Brain-computer interfaces (BCIs) recently have been focusing on combining various BCI modalities to form different combinations of hybrid BCIs. These paradigms are designed to elicit more than one brain potential in the form of BCI features. This research is being carried out with the objective of increasing classification accuracy and information transfer rate (ITR) based on measurement of brain potentials. This study proposed a novel hybrid BCI elicitation and measurement technique combining steady-state visually evoked potential (SSVEP) and P300 potentials to increase the ITR. The hybrid BCI also increased the number of target options compared to SSVEP paradigm for a set number of presumed frequencies of flickering. One of the hybrid BCIs used distinct colours along with distinct flickering frequencies for targets, with an aim to increase the accuracy of classification and reduction of system uncertainty parameter known as false activation rate (FAR). The results of a study in 10 volunteers established that the novel SSVEP-P300 hybrid BCI with distinct colours for target frequencies had average parameters as follows: classification accuracy of 90.76%, ITR of 81.10 bits/min and FAR of 2.99%. A comparative study of the two novel paradigms with SSVEP and P300 paradigms in the same environment was conducted. The results of the comparative study concluded that the hybrid BCI with distinct colours for various target frequencies yielded the best results and hence can be considered as a viable paradigm option for the development of an assistive device.

Visually guided saccadic eye movement has been considered a promising screening tool for cognitive function because of its simple and objective nature. However, its application to young children, especially those with developmental disorders, is limited due to the lack of sustained attention required to complete the measurement using the traditional electrophysiological protocol. We have previously reported that saccades can be reliably evaluated in typically developing young children using an eye tracker, which allows non-contact measurement of eye movement with a sufficiently short preparation time. Using the eye tracker system combined with an in-house developed analysis software, we investigated the changes in saccadic behavior between typically developing children (n = 30) and children with developmental disorders (attention-deficit/hyperactivity disorder [ADHD] and autism spectrum disorder [ASD], n = 27) at ages ranging from 4.8 to13.2 years. Four saccade responses were measured, consisting of eye movement to the instantaneously shifted visual target either to the right or left (step condition), to the visual target that appeared 200 ms after turning off the fixation point (gap condition), to the visual target while the fixation point remained (overlap condition), and in the direction opposite to the visual target (anti-saccade condition). Statistically significant deficits were found in children with developmental disorders compared to typically developing children, with decreased number of correct saccades in the step condition in children with ASD and reduced peak saccadic velocity in the overlap condition in children with ADHD. Interestingly, when saccadic parameters were further evaluated with regard to the direction of eye movement (rightward and leftward), a significant decrease in peak velocity in children with developmental disorders compared to typically developing children was only confirmed in saccades to the right side in the overlap condition. In addition, right-lateralized deficits in peak velocity and initial accuracy were also found in the step, overlap, and anti-saccade conditions in children with developmental disorders. These lateralized oculomotor responses may represent the affected cortical function in children with developmental disorders, suggesting a possible role of non-contact saccadic examination as an assessment tool for visual cognitive function, especially in young children.

Oral respiration causes constriction of the upper airway in the retropalatal and retroglossal regions, thereby increasing the risk of sleep disorder. One of the best methods to detect early signs of obstructive sleep apnea syndrome (OSAS) is daily monitoring of the respiration rate and mode of respiration during sleep. The vital signs are measured by a noncontact method in order to avoid burdening the subject and to allow differentiation between the various respiratory modes. In this study, we developed a system to measure the respiration rate and mode using far- and near-infrared cameras, and assessed the effectiveness of the proposed system and algorithm. A near-infrared camera detected the positions of the nostril and mouth, which are the pathways of expired and inspired air, respectively; while the far-infrared camera measured temperature changes in the nostril and mouth to derive the respiration rate and mode for detecting apnea. We enrolled 10 participants and measured their respiration rates using the aforementioned system under three states: nasal respiration, oral respiration, and apnea. The root-mean-square error for the respiration rate was 0.27 bpm, indicating that the system measured respiration without error in 92% of the trials. There was no error in discriminating between nasal and oral respiration. Additionally, this system detected apnea quite satisfactorily. The results of the experiment confirm that the system we developed effectively measures respiration in a noncontact manner.

Surgeons insert approximately 3000 hair implants in a single hair restoration surgery, and the needle shape critically affects surgical performance including surgical time, survival rate, and the surgeon's fatigue. There are several experimental studies on the relationship between needle shape and performance; however, there are no studies on hair implant needles, which are different from other medical needles as they have a comparatively short bevel. This study compared several hair implant needles to determine the relationship between needle shape and insertion force. Eight hair implant needles with the same diameter were collected, and their geometrical data were visually measured. The insertion force and total work of each needle were measured using custom-made force measurement equipment; the results were statistically compared, and their correlation was analyzed. Moreover, a qualitative comparison was made. The needle insertion force ranged from 1.30 to 2.97 N, and the total work ranged from 8.15 to 10.97 mJ. The primary bevel and acute angles showed a moderate positive correlation with the total work. A longer point length, smaller primary bevel, and acute angle decrease the total work of hair implant needles. As per the results of the qualitative comparison, we found that the surgeons can identify the differences in needle insertion force among various hair implant needles, and their grading was consistent with the grading according to the total work of needle insertion.

Several kinds of biometric authentications have been used as countermeasures against identity spoofing. Recently, some approaches have utilized time-series biosignals for biometrics, and one of the approaches is photoplethysmogram (PPG)-based authentication. PPG sensing has the advantage of fewer restriction of measurement sites than other time-series physiological signals. Moreover, it can connect the authentication and the healthcare applications seamlessly with one sensor. However, identity spoofing against PPG-based authentication may occur by exploiting this advantage. To develop a PPG-based authentication system with countermeasures, we propose the feasibility of a presentation attack against PPG-based authentication. The attack stealthily records PPGs on non-genuine measurement sites, and transmits the signals to the authentication device, thereby utilizing the advantage of PPG sensing in which signals can be recorded on various sites of a subject's body. We conducted an experiment to investigate the feasibility of the attack. We developed a PPG-based authentication system comprising a PPG sensing system including PPG sensors for multiple measurement sites, and an authentication algorithm based on an existing PPG-based identification algorithm. We recorded PPGs on three measurement sites on the subjects' bodies using the developed system. Then, we investigated the feasibility of the attack by inputting the feature values extracted from the PPGs recorded on non-genuine measurement sites to the classifier generated by the values from the PPGs recorded on genuine measurement sites. The results indicate that the attack can occur within a short time without any mapping under an ideal condition. Therefore, countermeasures such as liveness detection and utilization of unique information of measurement sites are required against the attack.

In traditional Chinese pulse diagnosis, the pulses of bilateral radial arteries are palpated by three fingers. In this study, a device that can objectively perform pulse diagnosis using multiple pressure sensors with a cuff was developed. The device has three pressure sensors in series and a cuff to apply external pressure load. Pulse waves were measured using the non-invasive pulse diagnostic device. The relationship of the data obtained from the proposed system with the diagnosis by the traditional method was examined. The study was conducted with the approval of the Ethics Committee in Tohoku University. Twelve adult male subjects were studied. Data of pulse waveforms recorded from the pulse diagnosis device, electrocardiogram, and continuous blood pressure measured at the fingertip were stored and analyzed. The pulse waves of the subjects were categorized as middle pulse, floating pulse, or sunken pulse according to pulse diagnosis in traditional Chinese medicine. Pulse–waveform data were obtained to measure the pulse–pressure changes, which showed different patterns of the rise time at the three measurement sites. Envelope analysis of the pulse waveform from the middle sensor revealed that the peak value of the pulse pressure increased during calculation stress compared to the resting state. It was possible to measure the pulse waves even when a higher external pressure was applied. Objective pulse-wave measurement using a method similar to that in traditional pulse diagnosis was realized, and pulse waveforms were measured using the proposed pulse diagnosis system.

Physical analgesia has been used for the treatment of sports injuries. We previously developed a special patch (bumpy patch, BP) that adds pressure to the pain area, demonstrating its analgesic effects in various sports injuries. In this study, we further verified acute and chronic analgesic effects of the BP in 28 subjects (14–44 years of age, 15 males and 13 females) with various pain sites and histories of pain (knee, shoulder, low back, elbow and other joints; 6 months to 3 days). First, the trigger area with muscle induration or tenderness was identified in each subject by palpation, and each subject evaluated the degree of pain using the numerical rating scale (NRS). Then the BPs were applied to completely cover the pain area. The BP treatment was repeated at 2 to 3-day interval until all subjects reported no pain (NRS = 0), and the differences between pre- and post-treatment NRS values were compared for each treatment as the measure of the analgesic effect. Significant analgesic effects were observed in the first and second BP treatments (P = 0.000774 and 0.00149, respectively). The third BP treatment also reduced the pain intensity, but the difference was not statistically significant (P = 0.3734). Some subjects reported mobilization of the pain area after the BP treatment, which was discussed in relation to the potential analgesic mechanism of the BP. These results indicate that the BP has acute and chronic analgesic effects and can be used for the treatment of sports injuries.

The creation of laser surgical systems with feedback, which allows performance of high-precision low-trauma operations, is the current trend of modern surgery. CO₂ lasers with pulse-periodic pumping which generate radiation at a wavelength of 10.6 µm and modulated at a frequency of 5–20 kHz are widely used in medical practice. This paper reports the possibility of creating feedback based on the autodyne effect that occurs in such surgical CO₂ lasers during laser dissection/evaporation of biotissues. The algorithm for extracting the information component (Doppler signal) of the autodyne signal for such CO₂ lasers has been developed. We showed that application of this algorithm permits extraction of the Doppler component spectrum in the autodyne signal that occurs when dissecting biotissues. Doppler signals were obtained when dissecting pig tissues in vitro, with a signal-to-noise ratio in the range of 5–15. The results obtained can be used in the development of smart laser surgical systems with feedback.

Segmenting blood vessels is an important step in a wide variety of tasks in medical image analysis. Patch-based convolutional neural networks (CNNs) are often used for vascular detection, but the impact of patch size and choice of CNN architecture have not been addressed in detail in previous studies. In this study, we aim to investigate the impact of patch size and CNN architecture on the accuracy of vascular detection from contract enhanced computed tomography (CT). We targeted the renal arteries as the primary focus of detection. We conducted experiments using contrast enhanced abdominal CT data of 30 cases. For the experiments, arteries in pre-defined regions of interest were manually labeled to build a dataset of input CT images and ground truth labels. We repeated the experiments with four patch sizes and two patch-based 3D CNN architectures (U-Net-like model and a simple sequential model) and evaluated the differences. Moreover, a Hessian-based line enhancing method was included in the evaluation to compare this non-deep learning method with the CNNs. The experimental results showed that patch size had a significant impact on detection accuracy. U-Net-like model showed peak accuracy at a certain patch size, unlike the sequential model that plateaued at large patch sizes. Although both CNNs outperformed Hessian-based line enhancement by a large margin, Hessian-based line enhancement achieved good recall when enhancing vessel structures not included in the CNN training. Our experiments show that different network architectures have different characteristics regarding their response to various patch sizes and vessel structures unseen during training.

We developed a virtual reality (VR)-based gait training system for inpatients aiming to improve their gait performance in a simulated home environment and to reduce the risk of falls after discharge. The proposed system simulates a home environment in a head-mounted display projection, in which the user can walk freely. The system provides combined visual, auditory, and tactile feedback when the user collides with an object in the simulated environment, aiming to encourage the user to modify his or her gait pattern. We report the training effect of the proposed system in three hospitalized patients who had stroke and lower limb paralysis. Changes in the balance function (Berg Balance Scale: BBS) and mobility function (Timed Up and Go test: TUG) were evaluated in all patients, and change in the fear of falling (Modified Falls Efficacy Scale: MFES) was additionally assessed in two patients (A and B). Patients A and B performed five sessions of gait training in the VR home environment every day for 5 consecutive days. Patient A who had moderate motor paralysis (Brunnstrom stage V) and moderately impaired gait function showed a remarkable improvement in MFES score, with minor improvement in TUG score and comparable BBS score after the intervention. Patient B who had severe motor paralysis and highly impaired gait function (Brunnstrom stage IV) showed remarkable improvement in TUG and BBS scores while there was no change in MFES score. Patient C who had severe motor impairment (Brunnstrom stage III) and a history of falls during a short visit at home underwent 20 minutes of gait training daily for 3 weeks in the VR-TUG room with a corridor having varying widths. The combined conventional and VR-based intervention diminished fall events during the TUG test in the patient's home, which remained after the same duration of conventional gait training prior to the combined intervention. These results suggest the potential of VR-based gait training in assisting patients to adapt to the environment, improve balance and gait function, and reduce the fear of falling.

High-frequency oscillation (HFO) is an important electrophysiological biomarker for estimating the epileptogenic zone in patients with epilepsy, but its clinical use is limited due to the high false-positive detection rate associated with conventional auto-detection methods based on one-dimensional spectral energy features. The purpose of this study was to apply a convolutional neural network (CNN)-based classifier to the candidate signals detected using conventional methods, and to extract HFOs more accurately and automatically. We adopted an image-based CNN because HFOs exhibit a localized power distribution in both time and frequency, which is utilized for the visual inspection of HFOs. To reduce the number of training datasets required for one patient, we employed transfer learning of an existing natural image classifier or the CNN HFO-classifier of another patient. We applied the proposed methods to the electrocorticography data of two patients with focal epilepsy who underwent pre-surgical examination. When the natural image discriminator AlexNet was transfer-learned to the HFO classifier, an accuracy of 93.0 ± 0.997% was achieved using 3000 training datasets. The false discovery rate (FDR) of HFO was 78.0% at the completion of the conventional method, which was significantly improved to 19.0 ± 4.42% after applying the CNN-based HFO classifier. When the HFO classifier trained with one patient was further relearned using the training datasets of another patient, the accuracy of determining HFOs in the latter patient was consistently above 91.0% (maximum 93.3 ± 0.967%) with the incorporation of 200 or more training datasets. These results suggest that the proposed method may provide an accurate, automatic, and personalized HFO classifier while liberating neurologists from the time-consuming manual detection of HFO signals for diagnosis.