JKJun’iChi Kotoku

Jun’iChi Kotoku

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Jun’iChi Kotoku has a total of 121 co-authors. Here are the top co-authors by number of shared publications:

  • Kenshiro Shiraishi

    Department of Radiology, Teikyo University School of Medicine, Tokyo, JPN

    11 shared publications
  • Shinobu Kumagai

    Central of Radiology, Teikyo University Hospital, Tokyo, JPN

    7 shared publications
  • Takeshi Takata

    Advanced Comprehensive Research Organization, Teikyo University , Tokyo 173-8605 , Japan

    7 shared publications
  • Takenori Kobayashi

    Graduate School of Medical Care and Technology, Teikyo University, Tokyo, Japan

    6 shared publications
  • Asuka Oyama

    Health Care Division, Health and Counseling Center, Osaka University, Osaka, Japan

    5 shared publications
  • Kenshiro Taguchi

    Graduate Degree Program of Health Data Science, Teikyo University, 2-11-1 Kaga, Itabashi, 173-8605, Japan, 81 339641211 ext 41601

    5 shared publications

Latest Publications

  • Open Access

    One-Year Trajectory of Step Counts and Weight Loss in Adults With Overweight/Obesity: Retrospective Cohort Study

    by Kenshiro Taguchi, Asuka Oyama, Jun'IChi Kotoku, Hiroshi Toki, Ryohei Yamamoto

    Background: Being overweight and obese are major health concerns worldwide, contributing to lifestyle-related diseases such as hypertension, dyslipidemia, type 2 diabetes, and cardiovascular disease. Increasing physical activity is an effective strategy for weight management. However, earlier step count studies have remained limited to small populations, short-term measurements of 1‐2 weeks, and mainly cross-sectional comparisons of average step counts. The effects of long-term step count changes on weight loss remain unclear. Objective: This study was conducted to assess the effects of long-term patterns of step counts on weight loss using data from the “Asmile” mobile health app in Japan. We hypothesized that participants with continuously increasing step counts over time would have a higher likelihood of significant weight reduction than participants who show steady or fluctuating patterns, even if their average step counts were similar. Methods: We analyzed data of 2778 Asmile users aged 40‐74 years with BMI ≥25 kg/m² who underwent a specific health checkup during fiscal years 2019‐2023 and who had valid step count records for 10‐14 months. Step count trajectories, reflecting long-term trends in physical activity, were classified using a latent class mixed model into four patterns: UP (increasing), FLAT (steady), DOWN (decreasing), and UP/DOWN (increasing then decreasing). Logistic regression was applied to estimate odds ratios for achieving ≥3% weight loss, with step trajectory as the explanatory variable and weight loss as the outcome. Results: Among participants, 1601 (57.6%) were men and 1177 (42.4%) were women, with respective mean ages of 65.8 (SD 7.9) and 64 (SD 8.2) years. Step count trajectories were distributed as 28.5% UP , 36.2% FLAT , 20.1% DOWN , and 15.2% UP/DOWN . Compared with the FLAT group, participants in the UP group had a significantly higher likelihood of achieving ≥3% weight loss (adjusted odds ratio 2.45, 95% CI 1.78‐3.38). Conclusions: Long-term tracking of step counts using the Asmile app revealed distinct activity patterns. Continuous increases in step counts were associated with the greatest likelihood of weight loss, emphasizing the importance of sustained physical activity. These findings support the use of long-term step monitoring to guide interventions for obesity and lifestyle-related disease prevention.

  • Open Access

    A Practical Study of Data Requirements for Self-Supervised Learning in Medical Image Analysis

    by Mitsuki Hommyo, Takumasa Tsuji, Shinobu Kumagai, Kenshiro Shiraishi, Jun'Ichi Kotoku

    Deep learning has become the mainstream approach for medical image analysis, but the availability of annotated datasets remains constrained in many clinical scenarios. Even when using such limited data, however, self-supervised learning (SSL) has yielded promising results. Nevertheless, comprehensive investigations into the number of images required for effective pretraining and fine-tuning in SSL frameworks remain inadequate. Especially lacking are such studies assessing factors such as class diversity, dataset scale, and task complexity. For this study, we evaluated the performance of widely adopted contrastive SSL models, particularly assessing their applicability in limited data settings by examining how the number of images used for pretraining and fine-tuning influences the accurate execution of binary classification tasks. Our findings indicate that among the three studied methods, Simple Siamese Representation Learning (SimSiam) achieved high accuracy based on only four training images. Achieving such performance would likely require at least 10,000 pretraining images. These findings offer practical insights into optimizing SSL-based pipelines for medical image analysis, particularly in scenarios involving rare diseases or severely scarce data.

  • Open Access

    One-year Trajectory of Step Counts and Weight Loss in Adults with Overweight/Obesity: A Retrospective Cohort Study (Preprint)

    by Kenshiro Taguchi, Asuka Oyama, Jun'Ichi Kotoku, Hiroshi Toki, Ryohei Yamamoto

    BACKGROUND Overweight and obesity has become a medical and social problem in economically developed and developing countries, including Japan. Overweight and obesity are associated with lifestyle-related diseases such as hypertension, dyslipidemia, type 2 diabetes, and arteriosclerotic cardiovascular disease. Increasing physical activity is an extremely effective method to promote weight loss as a common way to de-obesify and thereby improve lifestyle-related diseases. Earlier studies of step counts and weight were aimed at small populations and short-term step counts, which were recorded for 1–2 weeks while volunteers wore pedometers. Nevertheless, the relation between long-term and longitudinal step count changes and weight loss remains unclear. OBJECTIVE For this study, we analyzed effects of long-term steps on weight loss by accessing the database of "Asmile": a mobile health application provided by Osaka prefecture. METHODS We targeted Asmile users in their 40s to 70s who had undergone a Specific Health Checkup during fiscal years 2019–2023. Of these, we selected 2,778 users with a BMI of 25 kg/m2 or higher and continuous step count data of 10–14 months. Then, we conducted a cluster analysis of trajectories of a year's accumulated data of steps using a Latent Class Mixed Model (LCMM). Finally, odds ratios of step trajectories for weight loss were calculated using logistic regression with step trajectories as explanatory variables and more than 3% weight loss or not as the response variable. RESULTS Of the 2,778 members, 1,601 (57.6%) were men and 1,177 (42.4%) were women, with respective mean ages of 65.8±7.9 and 64.0±8.2. Step counts of around one year were classified into the four latent classes of UP, FLAT, DOWN, and UP/DOWN, respectively representing increasing, steady, decreasing, and increasing/decreasing trajectories by LCMM. Logistic regression shows UP as having an adjusted odds ratio of 2.45 (95% confidence interval: 1.78–3.38), compared to FLAT. CONCLUSIONS Asmile provided a reliable dataset that enabled tracking of users’ long-term step counts. We provided an effective analytical method to classify different tendencies of users' step counts, which can be explained easily as physical activities. The results revealed that different clusters of step count changes can differently affect weight loss.

  • Open Access

    Effects of Mobile Health Care App "Asmile" on Physical Activity of 80,689 Users in Osaka Prefecture, Japan: Longitudinal Observational Study

    by Asuka Oyama, Kenshiro Taguchi, Hiroe Seto, Reiko Kanaya, Jun'Ichi Kotoku, Miyae Yamakawa, Hiroshi Toki, Ryohei Yamamoto

    Background: Lifestyle-related diseases can be controlled by improving individuals’ lifestyles; however, improving and maintaining a healthy lifestyle is difficult. Mobile health (mHealth) applications have recently attracted attention as tools for maintaining and improving health, and their use may also increase physical activity. Objective: This study aimed to verify the effect of registration in Asmile, an mHealth application provided by the Osaka Prefectural Government, on step counts using a Causal Impact approach based on the step count data recorded in the Asmile application. Methods: This observational study included Osaka residents aged 20-79 years, newly registered to Asmile, between the fiscal years 2020 and 2023. Of these, 80,689 participants with step count records for 4 weeks before and after the day they registered to Asmile were included in the analysis. We used daily step counts that were automatically transferred from a standard smartphone health care app into Asmile. We used a Causal Impact model to estimate the increase in step count after registration to Asmile. Results: Of the 80,689 participants analyzed, 38.5% (31,082/80,689) were men, and the mean age was 51.6 (SD 13.2) years. The mean step count before registration was 5923 (SD 4860) steps per day, with the highest proportion of new users registered in spring (38,389/80,689, 47.6%) and in fiscal year 2020 (34,491/80,689, 42.7%). The analysis revealed that the effect of Asmile registration on step counts was 360 steps (95% CI 331-389) per day and 10,041 steps (95% CI 9632-10,450) over 4 weeks. Stratified analysis showed that the impact of increased step count was more pronounced in younger groups and groups with fewer step counts before registration. Conversely, the effect of registration on step count was relatively minor in the groups registered in summer or winter. Conclusions: This study demonstrates increased physical activity among users registered with the Asmile app. These findings suggest that mHealth apps such as Asmile can effectively promote healthier lifestyles and potentially reduce the risk of lifestyle-related diseases.


  • Open Access

    In silico prediction of loop-mediated isothermal amplification using a generalized linear model

    by Kenshiro Taguchi, Satoru Michiyuki, Takumasa Tsuji, Jun’iChi Kotoku

    Loop-mediated isothermal amplification (LAMP), a DNA amplification technique under isothermal conditions, provides the important benefits of high sensitivity, specificity, rapidity, and simplicity. Maximizing LAMP features necessitates the design of a complex LAMP primer set (LPS) consisting of four primers for six regions of a given target DNA. Furthermore, the LPS of a given target DNA is designed with LPS design support software such as Primer Explorer. However, even if the design is completed, we still must do many in vitro experiments and evaluations. Consequently, designing LPS often fails to achieve high performance, including efficient amplification. For this study, we examined in silico LAMP: a generalized linear model to predict DNA amplification from LPS. Using logistic regression with elastic net regularization, we identified factors that strongly affect LPS design. These factors, combined with domain knowledge for LPS design, led to the creation of LAMP kernel variables that are highly essential for high LAMP reaction. In silico LAMP, constructed using logistic regression with LAMP kernel variables, allows classification and performance prediction of LPS with an area under the curve of 0.86. These results suggest that a high LAMP reaction can be predicted using LAMP kernel variables and generalized linear regression model. Moreover, an LPS with high performance can be constructed without in vitro experimentation.


  • Open Access

    Diffusion equation quantification: selective enhancement algorithm for bone metastasis lesions in CT images

    by Yusuke Anetai, Kentaro Doi, Hideki Takegawa, Yuhei Koike, Midori Yui, Asami Yoshida, Kazuki Hirota, Ken Yoshida, Teiji Nishio, Jun’iChi Kotoku, et al.

    Objective. Diffusion equation (DE) imaging processing is promising to enhance images showing lesions of bone metastasis (LBM). The Perona–Malik diffusion (PMD) model, which has been widely used and studied, is an anisotropic diffusion processing method to denoise or extract objects from an image effectively. However, the smoothing characteristics of PMD or its related method hinder extraction and enhancement of soft tissue regions of medical image such as computed tomography (CT), typically leaving an indistinct region with ambient tissues. Moreover, PMD expands the border region of the objects. A novel diffusion methodology must be used to enhance the LBM region effectively. Approach. For this study, we originally developed a DE quantification (DEQ) method that uses a filter function to selectively provide modulated diffusion according to the original locations of objects in an image. The structural similarity index measure (SSIM) and Lie derivative image analysis L-value map were used to evaluate image quality and processing. Main results. We determined superellipse function with its order n = 4 as a better performing filter for the LBM region. DEQ was found to be more effective at contrasting LBM for various LBM CT images than PMD or its improved models when the filter was a positive exponential similar function. DEQ yields enhancement agreeing with the indications of positron emission tomography despite complex LBM comprising osteoblastic, osteoclastic, mixed tissues, and metal artifacts, which is innovative. Moreover, DEQ retained high quality of image (SSIM> 0.95), and achieved a low mean value of the L-value (Significance. Our method improved the visibility of mixed tissue lesions, which can assist computer visional framework and can help radiologists to produce accurate diagnose of LBM regions which are frequently overlooked in radiology findings because of the various degrees of visibility in CT images.


  • Open Access

    Cone Beam Computed Tomography Image-Quality Improvement Using “One-Shot” Super-resolution

    by Takumasa Tsuji, Soichiro Yoshida, Mitsuki Hommyo, Asuka Oyama, Shinobu Kumagai, Kenshiro Shiraishi, Jun’iChi Kotoku

    Cone beam computed tomography (CBCT) images are convenient representations for obtaining information about patients’ internal organs, but their lower image quality than those of treatment planning CT images constitutes an important shortcoming. Several proposed CBCT image-quality improvement methods based on deep learning require large amounts of training data. Our newly developed model using a super-resolution method, “one-shot” super-resolution (OSSR) based on the “zero-shot” super-resolution method, requires only small amounts of training data to improve CBCT image quality using only the target CBCT image and the paired treatment planning CT image. For this study, pelvic CBCT images and treatment planning CT images of 30 prostate cancer patients were used. We calculated the root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) to evaluate image-quality improvement and normalized mutual information (NMI) as a quantitative evaluation of positional accuracy. Our proposed method can improve CBCT image quality without requiring large amounts of training data. After applying our proposed method, the resulting RMSE, PSNR, SSIM, and NMI between the CBCT images and the treatment planning CT images were as much as 0.86, 1.05, 1.03, and 1.31 times better than those obtained without using our proposed method. By comparison, CycleGAN exhibited values of 0.91, 1.03, 1.02, and 1.16. The proposed method achieved performance equivalent to that of CycleGAN, which requires images from approximately 30 patients for training. Findings demonstrated improvement of CBCT image quality using only the target CBCT images and the paired treatment planning CT images.


  • Open Access

    Effects of Mobile Health Care App "Asmile" on Physical Activity of 80,689 Users in Osaka Prefecture, Japan: Longitudinal Observational Study (Preprint)

    by Asuka Oyama, Kenshiro Taguchi, Hiroe Seto, Reiko Kanaya, Jun'Ichi Kotoku, Miyae Yamakawa, Hiroshi Toki, Ryohei Yamamoto

    BACKGROUND Lifestyle-related diseases can be controlled by improving individuals’ lifestyles; however, improving and maintaining a healthy lifestyle is difficult. Mobile health (mHealth) applications have recently attracted attention as tools for maintaining and improving health, and their use may also increase physical activity. OBJECTIVE This study aimed to verify the effect of registration in Asmile, an mHealth application provided by the Osaka Prefectural Government, on step counts using a Causal Impact approach based on the step count data recorded in the Asmile application. METHODS This observational study included Osaka residents aged 20-79 years, newly registered to Asmile, between the fiscal years 2020 and 2023. Of these, 80,689 participants with step count records for 4 weeks before and after the day they registered to Asmile were included in the analysis. We used daily step counts that were automatically transferred from a standard smartphone health care app into Asmile. We used a Causal Impact model to estimate the increase in step count after registration to Asmile. RESULTS Of the 80,689 participants analyzed, 38.5% (31,082/80,689) were men, and the mean age was 51.6 (SD 13.2) years. The mean step count before registration was 5923 (SD 4860) steps per day, with the highest proportion of new users registered in spring (38,389/80,689, 47.6%) and in fiscal year 2020 (34,491/80,689, 42.7%). The analysis revealed that the effect of Asmile registration on step counts was 360 steps (95% CI 331-389) per day and 10,041 steps (95% CI 9632-10,450) over 4 weeks. Stratified analysis showed that the impact of increased step count was more pronounced in younger groups and groups with fewer step counts before registration. Conversely, the effect of registration on step count was relatively minor in the groups registered in summer or winter. CONCLUSIONS This study demonstrates increased physical activity among users registered with the Asmile app. These findings suggest that mHealth apps such as Asmile can effectively promote healthier lifestyles and potentially reduce the risk of lifestyle-related diseases.

  • Open Access

    Classification of chest X-ray images by incorporation of medical domain knowledge into operation branch networks

    by Takumasa Tsuji, Yukina Hirata, Kenya Kusunose, Masataka Sata, Shinobu Kumagai, Kenshiro Shiraishi, Jun’iChi Kotoku

    Background: This study was conducted to alleviate a common difficulty in chest X-ray image diagnosis: The attention region in a convolutional neural network (CNN) does not often match the doctor’s point of focus. The method presented herein, which guides the area of attention in CNN to a medically plausible region, can thereby improve diagnostic capabilities. Methods: The model is based on an attention branch network, which has excellent interpretability of the classification model. This model has an additional new operation branch that guides the attention region to the lung field and heart in chest X-ray images. We also used three chest X-ray image datasets (Teikyo, Tokushima, and ChestX-ray14) to evaluate the CNN attention area of interest in these fields. Additionally, after devising a quantitative method of evaluating improvement of a CNN’s region of interest, we applied it to evaluation of the proposed model. Results: Operation branch networks maintain or improve the area under the curve to a greater degree than conventional CNNs do. Furthermore, the network better emphasizes reasonable anatomical parts in chest X-ray images. Conclusions: The proposed network better emphasizes the reasonable anatomical parts in chest X-ray images. This method can enhance capabilities for image interpretation based on judgment.


  • Open Access

    New perforated radiation shield for anesthesiologists: Monte Carlo simulation of effects

    by Ayaka Yanagawa, Takeshi Takata, Taichi Onimaru, Takahiro Honjo, Takeyuki Sajima, Akihito Kakinuma, Akihisa Kataoka, Jun’ichi Kotoku

    Catheterization for structural heart disease (SHD) requires fluoroscopic guidance, which exposes health care professionals to radiation exposure risk. Nevertheless, existing freestanding radiation shields for anesthesiologists are typically simple, uncomfortable rectangles. Therefore, we devised a new perforated radiation shield that allows anesthesiologists and echocardiographers to access a patient through its apertures during SHD catheterization. No report of the relevant literature has described the degree to which the anesthesiologist’s radiation dose can be reduced by installing radiation shields. For estimating whole-body doses to anesthesiologists and air dose distributions in the operating room, we used a Monte Carlo system for a rapid dose-estimation system used with interventional radiology. The simulations were performed under four conditions: no radiation shield, large apertures, small apertures and without apertures. With small apertures, the doses to the lens, waist and neck surfaces were found to be comparable to those of a protective plate without an aperture, indicating that our new radiation shield copes with radiation protection and work efficiency. To simulate the air-absorbed dose distribution, results indicated that a fan-shaped area of the dose rate decrease was generated in the area behind the shield, as seen from the tube sphere. For the aperture, radiation was found to wrap around the backside of the shield, even at a height that did not match the aperture height. The data presented herein are expected to be of interest to all anesthesiologists who might be involved in SHD catheterization. The data are also expected to enhance their understanding of radiation exposure protection.