Apr. 2026

Dec. 2024

Dec. 2024

Aug. 2024

Mar. 2024

Aug. 2022

Mar. 2019

Feb. 2018
Outstanding performance award
Intestinal content signal suppression method in diffusion weighted images
Daisuke Hirahara

Machine learning models using 18F-FDG PET/CT radiomics for RAS mutation prediction and prognostic stratification in colorectal cancer.
Masatoyo Nakajo; Daisuke Hirahara; Kenji Baba; Mitsuho Hirahara; Yukari Eizuru; Atsushi Tani; Koji Takumi; Kiyohisa Kamimura; Fumiko Kanzaki; Takao Ohtsuka; Takashi Yoshiura
The British journal of radiology, 03 Jul. 2026
OBJECTIVE: To evaluate a machine learning (ML) model that integrates clinical data and 2-deoxy-2-[18F]fluoro-D-glucose (18F-FDG)-PET radiomic features for predicting RAS mutation status and prognosis in patients with colorectal cancer (CRC). METHODS: This retrospective study included 90 patients (mean age, 64 years, 60 men) with CRC who underwent pretreatment 18F-FDG-PET/CT. Radiomic features were extracted from PET images. Ten clinical variables and 49 radiomic features were analyzed. Using the AutoGluon ML framework, three models (clinical, radiomics, and combined) were developed with 10-fold cross-validation. RAS mutation prediction was evaluated by the AUC, and interpretability was assessed using SHAP analysis. Survival outcomes were evaluated using a RSF model, and risk scores for disease progression were calculated based on true RAS and ML-predicted RAS models. RESULTS: The radiomics ML model demonstrated the highest performance with an AUC of 0.675. SHAP analysis identified NGTDM_Complexity, kurtosis, and skewness as key contributors. For survival analysis, the C-indices based on true and ML-predicted RAS status were 0.685 and 0.675, respectively. A strong correlation was observed between the risk scores derived from the two models (r = 0.99, ρ = 0.98). Kaplan-Meier analysis demonstrated clear separation between high- and low-risk groups for both models (log-rank P<.01), with comparable survival curves. CONCLUSION: Although the radiomics-based ML model demonstrated moderate performance in predicting RAS mutation status, it provided prognostic stratification comparable to that based on true genetic profiles. ADVANCES IN KNOWLEDGE: 18F-FDG PET-based radiomics with ML may have potential as a noninvasive approach for genetic profiling and prognostic assessment in CRC, although further validation is required.

P0367 Machine learning analysis for predicting pathological complete response after chemoradiotherapy in patients with muscle-invasive bladder cancer
M. Fujiwara; H. Tanaka; D. Hirahara; T. Saho; E. Takaya; T. Ito; E. Muta; M. Urushido; S. Matsumoto; M. Kobayashi; Y. Nakamura; W. Chen; Y. Ishikawa; H. Fukushima; S. Fukuda; Y. Waseda; Y. Yasuda; S. Yoshida; Y. Fujii
European Urology, Mar. 2026

Development and Evaluation of a Urinary Na/K Ratio Prediction Model: A Systematic Comparison from Attention-Based Deep Learning to Classical Ensemble Approaches.
Emi Yuda; Itaru Kaneko; Daisuke Hirahara
Bioengineering (Basel, Switzerland), 21 Feb. 2026
The urinary sodium-to-potassium (Na/K) ratio is a clinically established predictor of blood pressure and cardiovascular risk. This study aimed to develop and rigorously evaluate machine learning models for estimating the urinary Na/K ratio using four easily obtainable physiological variables: body weight, systolic blood pressure, diastolic blood pressure, and pulse rate. A dataset of 82 participants was analyzed under a nested cross-validation framework to ensure strict generalization assessment. We first designed an attention-based deep learning model (MIDIP: Multi-Integrated Deep Ion Prediction). Although MIDIP showed reduced training error, nested validation revealed performance instability, indicating overfitting in this small-sample setting. We then compared classical machine learning models and ensemble strategies. Among all configurations, simple averaging of Random Forest, Gradient Boosting, and Linear Regression (Group A) achieved the best performance (MAE = 1.756, RMSE = 2.349, R2 = 0.390). In contrast, incorporating a Transformer model (Group B) degraded performance (MAE = 1.855, R2 = 0.294). Similarly, adaptive weighting (AWE) did not improve accuracy (Group A: MAE = 1.836, R2 = 0.266; Group B: MAE = 2.133, R2 = 0.035). These results demonstrate that, under limited sample conditions (N = 82), model simplicity and equal-weight ensemble integration provide superior generalization compared to attention-based or adaptively weighted deep architectures. The findings underscore the importance of strict validation and controlled model complexity when developing clinically applicable prediction models from small datasets.

Time-dependent diffusion-weighted imaging assessment of tumor grading and isocitrate dehydrogenase genotypes in adult-type diffuse gliomas.
Kiyohisa Kamimura; Tsubasa Nakano; Masanori Nakajo; Junki Kamizono; Tomohito Hasegawa; Daiki Tobo; Akie Mukai; Yoshiki Kamimura; Fumitaka Ejima; Hiroaki Nagano; Koji Takumi; Masatoyo Nakajo; Nayuta Higa; Hajime Yonezawa; Ryosuke Hanaya; Mari Kirishima; Akihide Tanimoto; Hirokazu Otsuka; Daisuke Hirahara; Hiroshi Imai; Thorsten Feiweier; Takashi Yoshiura
Japanese journal of radiology, 05 Jan. 2026
BACKGROUND: This study aimed to investigate the usefulness of time-dependent diffusion magnetic resonance imaging (MRI) parameters compared with the conventional apparent diffusion coefficient (ADC) in distinguishing tumor grade and isocitrate dehydrogenase (IDH) genotypes of adult-type diffuse gliomas. METHODS: This retrospective study included 102 patients with adult-type diffuse gliomas. ADC maps obtained using diffusion-weighted imaging at short (7.1 ms) and long (44.5 ms) diffusion times (ADC7.1ms and ADC44.5ms) and maps of ADC changes (cADC) and relative ADC changes (rcADC) between the two diffusion times were generated. The mean, 5th, and 95th percentile values of each parameter were compared between low-grade (LGGs) and high-grade gliomas (HGGs) and between IDH-mutant and IDH-wildtype gliomas. The discriminative performance was assessed using receiver operating characteristic (ROC) analysis, and correlation with Ki-67 labeling index (Ki-67LI) was assessed using Spearman's rank correlation. Multivariable logistic regression analyses were conducted to predict HGGs and IDH-wildtype gliomas. RESULTS: In HGGs, the mean and 5th percentile values of ADC44.5ms and ADC7.1ms were significantly lower, whereas cADC and rcADC indices were significantly higher than those in LGGs. Performance of the mean rcADC (area under the ROC curve: 0.925; 95% confidence interval: 0.855-0.967) was significantly better than any index of conventional ADCs for tumor grade classification. The mean rcADC demonstrated the strongest correlation with Ki-67LI (ρ = 0.542, p < 0.0001). Moreover, the 95th percentile of rcADC was an independent predictor of IDH-wildtype gliomas after adjustment for age and sex, was useful for distinguishing IDH-wildtype from IDH-mutant gliomas CONCLUSIONS: The mean rcADC showed the strongest correlation with the Ki-67 LI and achieved better diagnostic performance than conventional PGSE-based ADC for differentiating LGGs from HGGs. In multivariable analyses, the mean and 95th percentile of rcADC were identified as independent predictors of HGGs and IDH-wildtype gliomas, respectively.

Artificial intelligence in oncological positron emission tomography: advancing image analysis and interpretation.
M Nakajo; D Hirahara; M Hirahara; Y Eizuru; A Tani; F Kanzaki; K Takumi; K Kamimura; T Yoshiura
Clinical radiology, Jan. 2026
Functional and metabolic information provided by positron emission tomography (PET) imaging, such as patient diagnosis, tumour staging, and treatment evaluation, plays an important role in the clinical management of patients with cancer. Nonetheless, its clinical efficacy may be inhibited by differences in image quality and limitations in quantitative robustness. Artificial intelligence (AI) has transformed oncological PET imaging by improving image quality and facilitating a more consistent extraction of quantitative metrics. Recent research emphasises the value of AI in improving diagnostic accuracy and prognostic modelling. However, to ensure that AI-based PET analysis is successfully implemented in clinical practice, challenges such as imaging data standardisation, the development of reliable explainability methods, and the establishment of regulatory frameworks must be addressed. To optimise individualised care, future progress will likely be based on multimodal integration, federated learning, and probabilistic deep learning. Overall, this review highlights both the current progress and the remaining challenges of AI in oncological PET, aiming to provide a balanced perspective for future clinical translation.

CLIP-Guided Clustering with Archetype-Based Similarity and Hybrid Segmentation for Robust Indoor Scene Classification
Emi Yuda; Naoya Morikawa; Itaru Kaneko; Daisuke Hirahara
Electronics, Nov. 2025

Electron Density and Effective Atomic Number as Quantitative Biomarkers for Differentiating Malignant Brain Tumors: An Exploratory Study with Machine Learning.
Tsubasa Nakano; Daisuke Hirahara; Tomohito Hasegawa; Kiyohisa Kamimura; Masanori Nakajo; Junki Kamizono; Koji Takumi; Masatoyo Nakajo; Fumitaka Ejima; Ryota Nakanosono; Ryoji Yamagishi; Fumiko Kanzaki; Hiroki Muraoka; Nayuta Higa; Hajime Yonezawa; Ikumi Kitazono; Jihun Kwon; Gregor Pahn; Eran Langzam; Ko Higuchi; Takashi Yoshiura
Tomography (Ann Arbor, Mich.), 29 Oct. 2025

Noninvasive Optical Flow Analysis of White Blood Cell Dynamics for Enhanced COVID-19 Screening.
Emi Yuda; Yutaka Yoshida; Itaru Kaneko; Daisuke Hirahara; Junichiro Hayano
Journal of Advanced Computational Intelligence and Intelligent Informatics, Sep. 2025

Machine-Learning Insights from the Framingham Heart Study: Enhancing Cardiovascular Risk Prediction and Monitoring
Emi Yuda; Itaru Kaneko; Daisuke Hirahara
Applied Sciences, Aug. 2025

Machine learning-based prognostic modeling in gallbladder cancer using clinical data and pre-treatment [18F]-FDG-PET-radiomic features.
Masatoyo Nakajo; Daisuke Hirahara; Megumi Jinguji; Tetsuya Idichi; Mitsuho Hirahara; Atsushi Tani; Koji Takumi; Kiyohisa Kamimura; Takao Ohtsuka; Takashi Yoshiura
Japanese journal of radiology, May 2025

Predictive Potential of Contrast-Enhanced MRI-Based Delta-Radiomics for Chemoradiation Responsiveness in Muscle-Invasive Bladder Cancer.
Kohei Isemoto; Yuma Waseda; Motohiro Fujiwara; Koichiro Kimura; Daisuke Hirahara; Tatsunori Saho; Eichi Takaya; Yuki Arita; Thomas C Kwee; Shohei Fukuda; Hajime Tanaka; Soichiro Yoshida; Yasuhisa Fujii
Diagnostics (Basel, Switzerland), 21 Mar. 2025

Advancements in artificial intelligence for prostate cancer: Optimizing diagnosis, treatment, and prognostic assessment
Yuki Arita; Christian Roest; Thomas C. Kwee; Ramesh Paudyal; Alfonso Lema-Dopico; Stefan Fransen; Daisuke Hirahara; Eichi Takaya; Ryo Ueda; Lisa Ruby; Noam Nissan; Lawrence H. Schwartz; Amita Shukla-Dave; Oguz Akin
Asian Journal of Urology, Feb. 2025
OBJECTIVE: This review provides a comprehensive overview of the current research landscape on artificial intelligence (AI) in prostate cancer (PCa) management, highlighting its potential to enhance diagnosis, improve medical image quality, facilitate risk stratification, and aid prognosis. The review also identifies opportunities and challenges associated with integrating AI into clinical practice. METHODS: This review synthesizes findings from recent studies on AI applications in PCa management. It examines the use of machine learning and deep learning techniques in diagnostic imaging, surgical skill assessment, and outcome prediction. The analysis emphasizes empirical evidence demonstrating the efficacy and limitations of AI models in clinical settings. RESULTS: AI, particularly machine learning and deep learning algorithms, is improving diagnostic accuracy by analyzing medical images with greater efficiency and precision compared to traditional methods. AI-based tools are also being developed for surgical skill assessment, offering objective evaluations and feedback to surgeons. Additionally, AI applications in predicting patient outcomes are facilitating the creation of personalized treatment plans. Empirical evidence shows that AI models exhibit higher sensitivity and specificity in detecting clinically significant PCa, outperforming conventional diagnostic techniques. CONCLUSION: AI holds significant promise for transforming PCa management by improving diagnostic accuracy, personalizing treatment plans, and enhancing patient outcomes. While the evidence underscores its potential, challenges such as the need for larger, more diverse datasets and addressing implementation barriers remain critical. Despite these hurdles, the benefits of AI in PCa management represent a compelling area for future research and clinical integration.

Development and Evaluation of a Virtual Reality-Based Teaching Material for Interprofessional Education: A Case Study on Swallowing Videofluorography.
Shotaro Komaki; Shogo Baba; Yuuko Yotsumoto; Takayuki Yamashita; Susumu Takayoshi; Hisashi Niidome; Megumi Imamura; Megumi Mihara; Daisuke Hirahara
Japanese journal of comprehensive rehabilitation science, 2025
UNLABELLED: Komaki S, Baba S, Yotsumoto Y, Yamashita T, Takayoshi S, Niidome H, Imamura M, Mihara M, Hirahara D. Development and Evaluation of a Virtual Reality-Based Teaching Material for Interprofessional Education: A Case Study on Swallowing Videofluorography. Jpn J Compr Rehabil Sci 2025; 16: 37-45. OBJECTIVE: With the advancement of team-based medical care, effective interprofessional collaboration (IPC) has become increasingly important. However, implementing IPC without appropriate interprofessional education (IPE) remains challenging. This study aimed to clarify the educational effects of a virtual reality-based IPE teaching material on students' awareness and understanding of IPC. Specifically, it evaluated changes in awareness before and after lectures and the impact of different viewing environments on educational effectiveness. METHODS: A total of 224 students from six medical and welfare-related departments participated in lectures focused on IPC using a swallowing videofluorography scenario. The students were divided into three groups based on viewing environment: personal computer (PC), classroom screen, and virtual reality (VR). All groups viewed the same teaching material. Awareness and understanding of IPC were assessed before and after the lectures using a 12-item self-administered questionnaire. RESULTS: Post-lecture comparisons revealed significant increases in scores across all questionnaire items. In comparing the viewing environments, the VR group scored significantly higher than the PC group in the areas of "problem-solving skills" and "respectful attitude." CONCLUSION: This study suggests that VR-based teaching materials are effective in enhancing students' awareness and understanding of interprofessional collaboration.

Applying deep learning-based ensemble model to [18F]-FDG-PET-radiomic features for differentiating benign from malignant parotid gland diseases.
Masatoyo Nakajo; Daisuke Hirahara; Megumi Jinguji; Mitsuho Hirahara; Atsushi Tani; Hiromi Nagano; Koji Takumi; Kiyohisa Kamimura; Fumiko Kanzaki; Masaru Yamashita; Takashi Yoshiura
Japanese journal of radiology, Jan. 2025

Monitoring prostate cancer after low-dose-rate hemigland brachytherapy with delta-radiomics of diffusion-weighted magnetic resonance imaging
Kotaro Shimada; Motohiro Fujiwara; Daisuke Hirahara; Eichi Takaya; Soichiro Yoshida; Yasuhisa Fujii
International Journal of Urology, Dec. 2024

Inflammasome activation aggravates choroidal neovascularization.
Ryan D Makin; Ivana Apicella; Roshni Dholkawala; Shinichi Fukuda; Shuichiro Hirahara; Yoshio Hirano; Younghee Kim; Ayami Nagasaka; Yosuke Nagasaka; Siddharth Narendran; Felipe Pereira; Akhil Varshney; Shao-Bin Wang; Jayakrishna Ambati; Bradley D Gelfand
Angiogenesis, Nov. 2024
Inflammasome activation is implicated in diseases of aberrant angiogenesis such as age-related macular degeneration (AMD), though its precise role in choroidal neovascularization (CNV), a characteristic pathology of advanced AMD, is ill-defined. Reports on inhibition of inflammasome constituents on CNV are variable and the precise role of inflammasome in mediating pathological angiogenesis is unclear. Historically, subretinal injection of inflammasome agonists alone has been used to investigate retinal pigmented epithelium (RPE) degeneration, while the laser photocoagulation model has been used to study pathological angiogenesis in a model of CNV. Here, we report that the simultaneous introduction of any of several disease-relevant inflammasome agonists (Alu or B2 RNA, Alu cDNA, or oligomerized amyloid β (1-40)) exacerbates laser-induced CNV. These activities were diminished or abrogated by genetic or pharmacological targeting of inflammasome signaling constituents including P2rx7, Nlrp3, caspase-1, caspase-11, and Myd88, as well as in myeloid-specific caspase-1 knockout mice. Alu RNA treatment induced inflammasome activation in macrophages within the CNV lesion, and increased accumulation of macrophages in an inflammasome-dependent manner. Finally, IL-1β neutralization prevented inflammasome agonist-induced chemotaxis, macrophage trafficking, and angiogenesis. Collectively, these observations support a model wherein inflammasome stimulation promotes and exacerbates CNV and may be a therapeutic target for diseases of angiogenesis such as neovascular AMD.

Comprehensive Data Augmentation Approach Using WGAN-GP and UMAP for Enhancing Alzheimer’s Disease Diagnosis
Emi Yuda; Tomoki Ando; Itaru Kaneko; Yutaka Yoshida; Daisuke Hirahara
Electronics, 16 Sep. 2024, [Reviewed]

Impact of Downsampling Size and Interpretation Methods on Diagnostic Accuracy in Deep Learning Model for Breast Cancer Using Digital Breast Tomosynthesis Images.
Ryusei Inamori; Tomofumi Kaneno; Ken Oba; Eichi Takaya; Daisuke Hirahara; Tomoya Kobayashi; Kurara Kawaguchi; Maki Adachi; Daiki Shimokawa; Kengo Takahashi; Hiroko Tsunoda; Takuya Ueda
The Tohoku journal of experimental medicine, 25 Jul. 2024

Machine Learning Analysis of Predictors for Inhaled Nitric Oxide Therapy Administration Time Post Congenital Heart Disease Surgery: A Single-Center Observational Study.
Shuhei Niiyama; Takahiro Nakashima; Kentaro Ueno; Daisuke Hirahara; Masatoyo Nakajo; Yutaro Madokoro; Mitsuhito Sato; Kenshin Shimono; Takahiro Futatsuki; Yasuyuki Kakihana
Cureus, Jul. 2024

The usefulness of deep learning-based ensemble learning method using 18F-FDG-PET/CT radiomic features for differentiating between benign and malignant parotid gland diseases
Masatoyo Nakajo; Daisuke Hirahara; Megumi Jinguji; Mitsuho Hirahara; Atsushi Tani; Takashi Yoshiura
JOURNAL OF NUCLEAR MEDICINE, 01 Jun. 2024

ARE THERE DIFFERENCES IN MRI FINDINGS BETWEEN CRIBRIFORM AND NON-CRIBRIFORM CANCER? AN ANALYSIS USING RADIOMICS AND DELTA-RADIOMICS
Koki Enomoto; Soichiro Yoshida; Haruto Izumi; Sho Uehara; Yoh Matsuoka; Kohei Yamamoto; Daisuke Hirahara; Tatsunori Saho; Eichi Takaya; Shohei Fukuda; Yuma Waseda; Hajime Tanaka; Kenichi Ohashi; Yasuhisa Fujii
JOURNAL OF UROLOGY, May 2024

DELTA-RADIOMICS ANALYSIS IN COMPARISON TO RADIOMICS ANALYSIS USING DYNAMIC COMPUTED TOMOGRAPHY FOR PREOPERATIVE RISK STRATIFICATION IN UPPER URINARY TRACT UROTHELIAL CARCINOMA
Motohiro Fujiwara; Daisuke Hirahara; Tatsunori Saho; Eichi Takaya; Shunya Matsumoto; Kasumi Yoshitomi; Masaki Kobayashi; Yuki Nakamura; Bo Fan; Yudai Ishikawa; Shohei Fukuda; Yuma Waseda; Hajime Tanaka; Soichiro Yoshida; Yasuhisa Fujii
JOURNAL OF UROLOGY, May 2024

Machine learning approach using 18F-FDG-PET-radiomic features and the visibility of right ventricle 18F-FDG uptake for predicting clinical events in patients with cardiac sarcoidosis.
Masatoyo Nakajo; Daisuke Hirahara; Megumi Jinguji; Satoko Ojima; Mitsuho Hirahara; Atsushi Tani; Koji Takumi; Kiyohisa Kamimura; Mitsuru Ohishi; Takashi Yoshiura
Japanese journal of radiology, 16 Mar. 2024

Canine Model of Ischemia-Induced Ventricular Tachycardia.
Annie M Hirahara; Muhammad S Khan; Yuki Ishidoya; Douglas Smego; Orvelin Roman; Ravi Ranjan; Lauren Krueger; Guillaume Hoareau; Craig H Selzman; Derek J Dosdall
The Journal of surgical research, Jan. 2024
INTRODUCTION: Despite advances in antiarrhythmia therapies, ventricular tachycardia (VT) is a leading cause of sudden cardiac death. Investigation into the characteristics and new treatments for this arrhythmia is required to improve outcomes and a reproducible model of VT would be useful in these endeavors. We therefore created a canine model of ischemia-induced VT. MATERIALS AND METHODS: A pacing lead was implanted in the right ventricle in canines (n = 13) and the left anterior descending artery was occluded in two locations for 2 h and subsequently released to create an ischemia-reperfusion injury. In the 10 dogs that survived the first 48 h following the initial study, a terminal study was conducted 4-7 d later and VT was induced using premature stimulation or burst pacing through the right ventricle lead. The arrhythmia was terminated using either antitachycardia pacing or a defibrillatory shock. Multiple inductions into sustained VT were attempted. RESULTS: Sustained VT was induced in eight of 10 dogs with an average cycle length of 335 ± 70 bpm. Multiple episodes of VT were induced. Episodes of VT exhibited different electrocardiogram morphologies and cycle lengths in individual animals. CONCLUSIONS: This canine model provides a consistent technique for inducing multiple episodes of sustained VT. It may be useful for investigating VT mechanisms and testing novel therapeutics and treatments for patients with VT.

All-in-one platform for AI R&D in medical imaging, encompassing data collection, selection, annotation, and pre-processing.
Changhee Han; Kyohei Shibano; Wataru Ozaki; Keishiro Osaki; Takafumi Haraguchi; Daisuke Hirahara; Shumon Kimura; Yasuyuki Kobayashi; Gento Mogi
CoRR, 2024

Configurationally Nonselective Aquation of a Mononuclear Ru(II) Chloro Complex to Aquo Complex Isomers with Distinctive Aspects in Photoisomerization, Redox, and Catalytic Water Oxidation.
Yuta Tsubonouchi; Keisuke Inaba; Norihisa Hoshino; Masanari Hirahara; Debraj Chandra; Zaki N Zahran; Masayuki Yagi
Inorganic chemistry, 30 Oct. 2023
distal-[Ru(EtOtpy)(pynp)Cl]+ (d-EtO1Cl) (EtOtpy = 4'-ethoxy-2,2':6',2″-terpyridine, pynp = 2-(2-pyridyl)-1,8-naphthyridine), and distal/proximal-[Ru(EtOtpy)(pynp)OH2]2+ (d/p-EtO1H2O) complexes were newly synthesized to investigate the synergistic influence of the geometric configuration coupled with substituent introduction of an ethoxy (EtO) group on the physicochemical properties and reactions of the Ru(II) complexes. Configurationally nonselective aquation of d-EtO1Cl was uniquely observed to form d/p-EtO1H2O isomers in water, in contrast to configurationally selective aquation of distal-[Ru(tpy)(pynp)Cl]+ (d-1Cl, tpy = 2,2':6',2″-terpyridine) without the EtO group [Yamazaki, H. . J. Am. Chem. Soc. 2011, 133, 8846-8849].The kinetic profiles of the aquation reactions of d-EtO1Cl were well analyzed using a sequential reversible reaction model assuming the reversible interconversion between d/p-EtO1H2O isomers via d-EtO1Cl. The observed equilibrium constant (Kiso) of isomerization between p/d-EtO1H2O was calculated from the kinetic analysis as Kiso = 0.45, which is consistent with the final concentration ratio (1:0.43) of p/d-EtO1H2O generated in the aquation reaction of d-EtO1Cl. The irreversible photoisomerization from d-EtO1H2O to p-EtO1H2O was observed in water with an internal quantum yield (Φ) of 0.44% at 520 nm. Electrochemical measurements showed that d-EtO1H2O undergoes a 2-step oxidation reaction of 1H+-coupled 1e- processes of RuII-OH2/RuIII-OH and RuIII-OH/RuIV═O at pH 1.3-9.7, whereas p-EtO1H2O undergoes a 1-step oxidation reaction of a 2H+-coupled 2e- process of RuII-OH2/RuIV═O in the pH range of 1.8-11.5. Any redox potential of d/p-EtO1H2O isomers was decreased by the electro-donating EtO substitution, compared with distal/proximal-[Ru(tpy)(pynp)OH2]2+ (d/p-1H2O). The turnover frequency (kO2 = 1.7 × 10-2 s-1) of d-EtO1H2O for water oxidation catalysis is higher than that (3.5 × 10-4 s-1) of p-EtO1H2O by a factor of 48.6. The kO2 value (1.7 × 10-2 s-1) for d-EtO1H2O is 4.5-fold higher than those of d-1H2O (3.8 × 10-3 s-1). The higher kO2 value of d-EtO1H2O compared with d-1H2O could be explained by the fast oxidation rate from RuIV═O to RuV═O involved in the rate-determining step due to the electron-donating EtO group.

Application of Machine Learning Analyses Using Clinical and [18F]-FDG-PET/CT Radiomic Characteristics to Predict Recurrence in Patients with Breast Cancer.
Kodai Kawaji; Masatoyo Nakajo; Yoshiaki Shinden; Megumi Jinguji; Atsushi Tani; Daisuke Hirahara; Ikumi Kitazono; Takao Ohtsuka; Takashi Yoshiura
Molecular imaging and biology, Oct. 2023

The usefulness of machine-learning-based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features for predicting prognosis in patients with laryngeal cancer.
Masatoyo Nakajo; Hiromi Nagano; Megumi Jinguji; Yoshiki Kamimura; Keiko Masuda; Koji Takumi; Atsushi Tani; Daisuke Hirahara; Keisuke Kariya; Masaru Yamashita; Takashi Yoshiura
The British journal of radiology, 10 Jul. 2023

USEFULNESS OF DELTA-RADIOMICS APPLIED TO CONTRAST- AND NON-ENHANCED MRI FOR PREDICTING THERAPEUTIC EFFICACY IN MUSCLE-INVASIVE BLADDER CANCER
Kohei Isemoto; Yuma Waseda; Motohiro Fujiwara; Koichiro Kimura; Daisuke Hirahara; Tatsunori Saho; Eichi Takaya; Shohei Fukuda; Hajime Tanaka; Soichiro Yoshida; Minato Yokoyama; Yasuhisa Fujii
JOURNAL OF UROLOGY, Apr. 2023

[The Fundamentals of Diffusion Weighted Imaging (DWI) in the Mammary Region and Its Application to Artificial Intelligence (AI)].
Daisuke Hirahara
Nihon Hoshasen Gijutsu Gakkai zasshi, 2023

Radiomics model of diffusion-weighted whole-body imaging with background signal suppression (DWIBS) for predicting axillary lymph node status in breast cancer
Takafumi Haraguchi; Yasuyuki Kobayashi; Daisuke Hirahara; Tatsuaki Kobayashi; Eichi Takaya; Mariko Takishita Nagai; Hayato Tomita; Jun Okamoto; Yoshihide Kanemaki; Koichiro Tsugawa
Journal of X-ray science and technology, 2023

Distinctive Aspects in Aquation, Proton-Coupled Redox, and Photoisomerization Reactions between Geometric Isomers of Mononuclear Ruthenium Complexes with a Large-π-Conjugated Tetradentate Ligand.
Yuta Tsubonouchi; Takeumi Watanabe; Kazuha Yoshida; Shunsuke Watabe; Keisuke Inaba; Masanari Hirahara; Tsubasa Hatanaka; Yasuhiro Funahashi; Debraj Chandra; Norihisa Hoshino; Zaki N Zahran; Masayuki Yagi
Inorganic chemistry, 05 Sep. 2022
Geometric isomers of mononuclear ruthenium(II) complexes, distal-/proximal-[Ru(tpy)(dpda)Cl]+ (d-/p-RuCl, tpy = 2,2':6',2″-terpyridine, dpda = 2,7-bis(2-pyridyl)-1,8-diazaanthracene), were newly synthesized to comprehensively investigate the geometric and electronic structures and distinctive aspects in various reactions between isomers. The ultraviolet (UV)-visible absorption spectra of d-/p-RuCl isomers show intense bands for metal-to-ligand charge transfer (MLCT) at close wavelengths of 576 and 573 nm, respectively. However, time-dependent density functional theory (TD-DFT) calculations suggest that the MLCT transition of d-RuCl involves mainly single transitions to the π* orbital of the dpda ligand in contrast to mixing of the π* orbitals of the dpda and tpy ligands for p-RuCl. The aquation reaction (1.5 × 10-3 s-1) of p-RuCl to yield proximal-[Ru(tpy)(dpda)(OH2)]2+ (p-RuH2O) is faster than that (5.3 × 10-6 s-1) of d-RuCl in D2O/CD3OD (4:1 v/v) by three orders of magnitude, which resulted from the longer Ru-Cl bond by 0.017 Å and the distorted angle (100.2(3)°) of Cl-Ru-N (a nitrogen of dpda, being on a tpy plane) due to the steric repulsion between Cl and dpda for p-RuCl. Electrochemical measurements showed that d-RuH2O undergoes a 2-step oxidation reaction of 1H+-coupled 1e- processes of RuII-OH2/RuIII-OH and RuIII-OH/RuIV═O at pH 1-9, whereas p-RuH2O undergoes a 1-step oxidation reaction of a 2H+-coupled 2e- process of RuII-OH2/RuIV═O in the pH range of pH 1-10. The irreversible photoisomerization from d-RuH2O to p-RuH2O was observed in aqueous solution with an internal quantum yield (Φ) of 5.4 × 10-3% at 520 nm, which is lower compared with Φ = 1.1-2.1% of mononuclear Ru(II) aquo complexes with similar bidentate ligands instead of dpda by three orders of magnitude. This is possibly ascribed to the faster nonradiative decay rate from the excited 3MLCT state to the ground state for d-RuH2O due to the lower π* level of dpda ligands according to the energy-gap law: the rate decreases exponentially with the increasing energy gap.

Deep learning approach of diffusion-weighted imaging as an outcome predictor in laryngeal and hypopharyngeal cancer patients with radiotherapy-related curative treatment: a preliminary study.
Hayato Tomita; Tatsuaki Kobayashi; Eichi Takaya; Sono Mishiro; Daisuke Hirahara; Atsuko Fujikawa; Yoshiko Kurihara; Hidefumi Mimura; Yasuyuki Kobayashi
European radiology, Aug. 2022

The Usefulness of Machine Learning-Based Evaluation of Clinical and Pretreatment [18F]-FDG-PET/CT Radiomic Features for Predicting Prognosis in Hypopharyngeal Cancer.
Masatoyo Nakajo; Kodai Kawaji; Hiromi Nagano; Megumi Jinguji; Akie Mukai; Hiroshi Kawabata; Atsushi Tani; Daisuke Hirahara; Masaru Yamashita; Takashi Yoshiura
Molecular imaging and biology, 21 Jul. 2022

機械学習を用いたサルとヒトの心拍変動指標の分類と識別性　　　　　　　　　　　　　　　
Feb. 2022

機械学習を用いたサルとヒトの心拍変動指標の分類と識別性　　　　　　　　　　　　　　　
Feb. 2022

Discrimination between Cynomolgus Monkey (Macaca Fascicularis) and Humans using Machine Learning and Heart Rate Variability Indices.
Itaru Kaneko; Daisuke Hirahara; Junichiro Hayano; Oscar Martinez Mozos; Emi Yuda
WAC, 2022

DDX17 is an essential mediator of sterile NLRC4 inflammasome activation by retrotransposon RNAs.
Shao-Bin Wang; Siddharth Narendran; Shuichiro Hirahara; Akhil Varshney; Felipe Pereira; Ivana Apicella; Meenakshi Ambati; Vidya L Ambati; Praveen Yerramothu; Kameshwari Ambati; Yosuke Nagasaka; Dionne Argyle; Peirong Huang; Kirstie L Baker; Kenneth M Marion; Kartik Gupta; Bo Liu; David R Hinton; Scott W Canna; Tamer Sallam; Srinivas R Sadda; Nagaraj Kerur; Bradley D Gelfand; Jayakrishna Ambati
Science immunology, 03 Dec. 2021
Detection of microbial products by multiprotein complexes known as inflammasomes is pivotal to host defense against pathogens. Nucleotide-binding domain leucine-rich repeat (NLR) CARD domain containing 4 (NLRC4) forms an inflammasome in response to bacterial products; this requires their detection by NLR family apoptosis inhibitory proteins (NAIPs), with which NLRC4 physically associates. However, the mechanisms underlying sterile NLRC4 inflammasome activation, which is implicated in chronic noninfectious diseases, remain unknown. Here, we report that endogenous short interspersed nuclear element (SINE) RNAs, which promote atrophic macular degeneration (AMD) and systemic lupus erythematosus (SLE), induce NLRC4 inflammasome activation independent of NAIPs. We identify DDX17, a DExD/H box RNA helicase, as the sensor of SINE RNAs that licenses assembly of an inflammasome comprising NLRC4, NLR pyrin domain–containing protein 3, and apoptosis-associated speck-like protein–containing CARD and induces caspase-1 activation and cytokine release. Inhibiting DDX17-mediated NLRC4 inflammasome activation decreased interleukin-18 release in peripheral blood mononuclear cells of patients with SLE and prevented retinal degeneration in an animal model of AMD. Our findings uncover a previously unrecognized noncanonical NLRC4 inflammasome activated by endogenous retrotransposons and provide potential therapeutic targets for SINE RNA–driven diseases.

Effect of the Pixel Interpolation Method for Downsampling Medical Images on Deep Learning Accuracy
Daisuke Hirahara; Eichi Takaya; Mizuki Kadowaki; Yasuyuki Kobayashi; Takuya Ueda
Journal of Computer and Communications, 30 Nov. 2021, [Reviewed]

Machine learning based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features to predict prognosis of cervical cancer patients.
Masatoyo Nakajo; Megumi Jinguji; Atsushi Tani; Erina Yano; Chin Khang Hoo; Daisuke Hirahara; Shinichi Togami; Hiroaki Kobayashi; Takashi Yoshiura
Abdominal radiology (New York), 25 Nov. 2021

Nucleoside reverse transcriptase inhibitors and Kamuvudines inhibit amyloid-β induced retinal pigmented epithelium degeneration.
Siddharth Narendran; Felipe Pereira; Praveen Yerramothu; Ivana Apicella; Shao-Bin Wang; Kameshwari Ambati; Shuichiro Hirahara; Younghee Kim; Meenakshi Ambati; Vidya L Ambati; Peirong Huang; Akhil Varshney; Yosuke Nagasaka; Shinichi Fukuda; Kirstie L Baker; Kenneth M Marion; Jan M Deussing; Srinivas R Sadda; Bradley D Gelfand; Jayakrishna Ambati
Signal transduction and targeted therapy, 14 Apr. 2021
Nonfibrillar amyloid-β oligomers (AβOs) are a major component of drusen, the sub-retinal pigmented epithelium (RPE) extracellular deposits characteristic of age-related macular degeneration (AMD), a common cause of global blindness. We report that AβOs induce RPE degeneration, a clinical hallmark of geographic atrophy (GA), a vision-threatening late stage of AMD that is currently untreatable. We demonstrate that AβOs induce activation of the NLRP3 inflammasome in the mouse RPE in vivo and that RPE expression of the purinergic ATP receptor P2RX7, an upstream mediator of NLRP3 inflammasome activation, is required for AβO-induced RPE degeneration. Two classes of small molecule inflammasome inhibitors-nucleoside reverse transcriptase inhibitors (NRTIs) and their antiretrovirally inert modified analog Kamuvudines-both inhibit AβOs-induced RPE degeneration. These findings crystallize the importance of P2RX7 and NLRP3 in a disease-relevant model of AMD and identify inflammasome inhibitors as potential treatments for GA.

Application of a Machine Learning Approach for the Analysis of Clinical and Radiomic Features of Pretreatment [18F]-FDG PET/CT to Predict Prognosis of Patients with Endometrial Cancer.
Masatoyo Nakajo; Megumi Jinguji; Atsushi Tani; Hidehiko Kikuno; Daisuke Hirahara; Shinichi Togami; Hiroaki Kobayashi; Takashi Yoshiura
Molecular imaging and biology, 24 Mar. 2021

Application of a machine learning approach to characterization of liver function using 99mTc-GSA SPECT/CT.
Masatoyo Nakajo; Megumi Jinguji; Atsushi Tani; Daisuke Hirahara; Hiroaki Nagano; Koji Takumi; Takashi Yoshiura
Abdominal radiology (New York), 06 Mar. 2021

Deep Learning for the Preoperative Diagnosis of Metastatic Cervical Lymph Nodes on Contrast-Enhanced Computed ToMography in Patients with Oral Squamous Cell Carcinoma.
Hayato Tomita; Tsuneo Yamashiro; Joichi Heianna; Toshiyuki Nakasone; Tatsuaki Kobayashi; Sono Mishiro; Daisuke Hirahara; Eichi Takaya; Hidefumi Mimura; Sadayuki Murayama; Yasuyuki Kobayashi
Cancers, 03 Feb. 2021

Effects of data count and image scaling on Deep Learning training
Daisuke Hirahara; Eichi Takaya; Taro Takahara; Takuya Ueda
PeerJ Computer Science, 16 Nov. 2020

Create medical AI (super resolution and image generation)
Daisuke Hirahara
Communication of the Imaging Group of the JSRT, Apr. 2020, [Reviewed], [Invited]

[Progress of Radiological Technology by AI].
Chisako Muramatsu; Ikuo Kawashita; Ryohei Nakayama; Naoki Kamiya; Daisuke Hirahara
Nihon Hoshasen Gijutsu Gakkai zasshi, 2020

MRI画像を用いた乳がん診断補助用AI（悪性疑い検出）の開発　　　　　　　　　　　　　　　
Feb. 2019, [Reviewed]

DWIBSによる乳がん検診のAI診断の試み　　　　　　　　　　　　　　　
Feb. 2019, [Reviewed], [Invited]

Preliminary assessment for the development of CADe system for brain tumor in MRI images utilizing transfer learning in Xception model
Daisuke Hirahara
2019 IEEE 8th Global Conference on Consumer Electronics (GCCE), 2019, [Reviewed]

Fundamental study on preliminary image processing at time development of CNN using chest radiography.
Daisuke Hirahara; Emi Yuda; Taro Takahara; Yasuyuki Kobayashi
IEEE LifeTech2019, 2019, [Reviewed]

DWIBS画像を用いた乳がん診断補助用AI（悪性疑い検出）の開発と性能評価
Jan. 2019, [Reviewed], [Invited]

海馬および海馬傍回の萎縮度判定解析ソフトにおける永久磁石装置撮像パラメータ標準化に関する研究　　　　　　　　　　　　　　　
Oct. 2018, [Reviewed]

無症候性野球競技者における上腕骨内側上顆下端の病態と円回内筋・浅指屈筋の実態について
Aug. 2018, [Reviewed]

MRIを用いた体幹筋トレーニングの違いによる筋活動と腹腔内面積の変化-ドローインVSブレーシング-　　　　　　　　　　　　　　　
Aug. 2018, [Reviewed]

大腿四頭筋セッティングにおける膝蓋上嚢の形態変化の観察-核磁気共鳴画像法(MRI)による横断像と矢状断像を用いて-
Aug. 2018, [Reviewed]

医療系専門学校におけるプログラミング教育の実践
Aug. 2018, [Reviewed]

Vibe codingによる医療AIソフトウエア開発の可能性—特集 医療AIのさらなる普及に向けて : 医療におけるAI活用の現在地と将来展望 ; 医療AIを加速させる技術の現状
Jul. 2026

機械学習を用いた持続的腎代替療法におけるLife-timeの予測
Sep. 2025

【課題解決に向けたAI活用の最前線 働き方改革から2040年問題まで医療の課題に立ち向かう】医療におけるAI研究開発の最新動向と将来展望 医療におけるノーコードAI開発の実際と今後の可能性
Jun. 2025

AIと循環制御医学の融合 研究イノベーションから臨床実装まで
Jun. 2025

人工知能の軌跡と奇跡-医療の未来への新たなるステップ- 医用画像におけるAIの軌跡と未来への展望　　　　　　　　　　　　　　　
Mar. 2025

医用画像解析の未来を描くマルチモーダル時代のLLM入門　　　　　　　　　　　　　　　
Mar. 2025

医用画像解析の未来を描くマルチモーダル時代のLLM入門
2025

診療放射線技師のための生成AI入門
2025

診療放射線養成課程学⽣の臨床実習前後における概念変化の分析：機械学習アプローチ　　　　　　　　　　　　　　　
24 Aug. 2024
Corresponding

診療放射線養成課程学⽣の臨床実習前後における概念変化の可視化 ：次元削減⼿法を⽤いた分析　　　　　　　　　　　　　　　
24 Aug. 2024
Corresponding

特集 DXが理学療法にもたらす未来 教育現場でのDX活用の実際と可能性
15 Apr. 2024

Comparative Analysis of Noise Reduction Models for Deep Learning-Based COVID-19 Detection in Low-Dose CT Imaging
柿原妃菜子; 久保田莉奈; 尾崎望未; 堂免功要; 平原大助; 平原大助; 平原大助; 平原大助; 平原大助
日本診療放射線技師会誌, 2024

乳がん検診へのRadiomics応用の可能性 RadiomicsのMRIへの適用　　　　　　　　　　　　　　　
Nov. 2023

指先毛細血管オプティカルフローによる感染症のスクリーニング手法　　　　　　　　　　　　　　　
Nov. 2023

ヒトの心拍変動指標の個人識別性 機械学習を用いたサル指標との比較　　　　　　　　　　　　　　　
Nov. 2023

Virtual Reality(VR)を活用した多職種連携能力向上のための教材開発とその評価　　　　　　　　　　　　　　　
Nov. 2023

「 ニュウセン リョウイキ 」 ロンブン トクシュウゴウ
平原大助; 平原大助; 平原大助
日本放射線技術学会雑誌, Nov. 2023
Japanese Society of Radiological Technology

多職種連携教育におけるVR(Virtual Reality)の有用な活用方法について VF(videofluoroscopic examination of swallowing)検査における協働場面を通して
Sep. 2023

【医療AIの普及拡大とさらなる展開 医療からヘルスケアサービスまで発展に向けた現状と展望】医療AIのための人材育成の現状と展望 診療放射線技師養成校における医療AI人材育成
Jun. 2023

Virtual Realityの臨場感がもたらす多職種連携教育の可能性 専門職養成課程におけるVRを用いた具体的教育方略の考案　　　　　　　　　　　　　　　
Apr. 2023

Considering the Possibility of Interprofessional Education from the Perspective of Highly Realistic Virtual Reality Education-Devise a Teaching Methods that Incorporates Virtual Reality Technology into Professional Training Courses-
Yamashita Takayuki; Hirahara Daisuke; Shimoi Toshinori
Journal of Physical Therapy Education(Japan), 01 Apr. 2023
Introduction: To provide high -quality medical and welfare services, it is essential to practice cooperation between related occupations, and education in cooperation with related occupations is becoming common in vocational schools. However, interprofessional education (IPE) has many environmental and management issues, such as different departments in each school, varied training methods for different occupations, and difficulty in making various adjustments. Methods: We introduced virtual reality (VR) technology as teaching materials for IPE. We investigated the impressions of students and faculty members who used VR teaching materials, and analyzed the results via text mining. Results: Respondents want VR material that allows them to experience a simulated clinical experience. For IPE, there was a requirement for teaching materials that could help in understanding the flow of collaboration and related occupations. Discussion: Owing to its high sense of presence and reproducibility, VR can revolutionize existing educational strategies. In the future, further enrichment of VR teaching materials and verification of educational effects are necessary., Japanese Society of Physical Therapy Education

理学療法士養成校におけるVR教材の効果的なデザイン
2023

デルタラジオミクスを用いた筋層浸潤性膀胱癌に対する放射線化学療法の治療効果予測
2023

RadiomicsのMRIへの適用
2023

放射線部門のサスティナビリティを向上するデジタル戦略 最新動向から見えてきた診療の質向上と業務効率化のヒント 放射線診療におけるAIの最新動向
2023

学習済みNAFNet(Nonlinear Activation Free Network)を用いたMRCP画像のアーチファクト軽減に関する検討
2023

Data Augmentationの拡張確率が分類精度に与える影響
2023

Screening for infections using the optical flow method of fingertip capillaries
平原大助; 平原大助; 金子格; 早野順一郎; 湯田恵美; 湯田恵美
医療情報学連合大会論文集(CD-ROM), 2023

Machine Learning Models for the Personal Identification using Monkey and Human’s Heart Rate Variability Data
平原大助; 平原大助; 金子格; 早野順一郎; 湯田恵美; 湯田恵美
医療情報学連合大会論文集(CD-ROM), 2023

Development and Evaluation of Teaching Materials for Improving Multidisciplinary Collaboration Skills Using Virtual Reality (VR)
小牧祥太郎; 馬場祥吾; 四元祐子; 山下喬之; 平原大助
医療情報学連合大会論文集(CD-ROM), 2023

Report on Production of Virtual Reality Teaching Materials for Interprofessional Education: 2nd Report
四元祐子; 山下喬之; 平原大助; 佐々木信幸
日本作業療法学会抄録集(CD-ROM), 2023

見落とされがちなAIの基礎知識とMRI分野のAI
15 Nov. 2022

医療系専門職養成教育現場における多職種連携教育Virtual Reality(VR)システムの有用性の調査 医療専門職を対象とした事前検証　　　　　　　　　　　　　　　
Nov. 2022

多職種連携教育VRコンテンツ制作の報告　　　　　　　　　　　　　　　
Aug. 2022

中間赤外線波動追跡計算を用いた宇宙機搭載用イマージョンエシェル高分散分光器における波長分解能の評価
2022

Investigation of the usefulness of a multidisciplinary collaborative educational Virtual Reality (VR) system in the field of medical professional development education.-Preliminary validation for medical professionals-
小牧祥太郎; 馬場祥吾; 四元祐子; 山下喬之; 平原大助
医療情報学連合大会論文集(CD-ROM), 2022

Classification and Discriminability of Heart Rate Variability Indices in Monkeys and Humans using Machine Learning
金子格; 平原大助; 吉田豊; 早野順一郎; MOZOS Oscar Martinez; 湯田恵美
電子情報通信学会技術研究報告(Web), 2022

MRI領域のAI画像診断
2021

医療・ヘルスケア領域におけるAI-AIの基礎から活用まで 人工知能の基礎知識 II 医療・ヘルスケア従事者の精神的負担を減らす画像認識技術のスタンダード,「畳み込みニューラルネットワーク」の基礎知識
2021

医療・ヘルスケア領域におけるAI-AIの基礎から活用まで 人工知能の基礎知識 III「Transformer」...Google翻訳の精度を革命的に向上させた機械学習モデルは,医療やヘルスケアの世界も塗り替える可能性
2021

人工知能基礎知識 I(機械学習概論)
2021

The potential of AI as a diagnostic aid in MRI breast cancer screening
平原大助; 平原大助; 平原大助; 高原太郎
日本乳癌検診学会誌(Web), 2021

腫瘤形成性乳癌の乳房トモシンセシス画像のAI解析によるセンチネルリンパ節転移リスクの予測
2021

The potential of AI as a diagnostic aid in MRI breast cancer screening
Hirahara Daisuke; Takahara Taro
Nihon Nyugan Kenshin Gakkaishi (Journal of Japan Association of Breast Cancer Screening), 2021
More than 60,000 Japanese women are diagnosed with breast cancer annually, and 13,000 die from the disease.Therefore, breast cancer screening is very important to achieve mortality reduction. We are conducting research and development on deep learning of DWIBS, an image with excellent contrast that emphasizes microscopic water diffusion, and mammary MRI images with various contrasts such as T1WI and T2WI. Using the deep learning model Xception, we developed a diagnostic aid model for fat-suppressed T2-weighted images and diffusion-weighted images. The combination of AI to assist diagnosis by taking advantage of the characteristics of MRI images may further contribute to the goal of reducing the mortality rate of breast cancer screening., Japan Association of Breast Cancer Screening

Comparison of noise2noise and noise2clear in MRI images
Hirahara Daisuke
Transactions of Japanese Society for Medical and Biological Engineering, 2021
Purpose:To compare performance of denoising processing using two denoising AI models (noise2noise and noise2clear) on whole body MR imaging by Diffusion Weighted Imaging with Background Suppression .Materials and Methods:DWIBS images of MRI examination for cancer screening were targeted. For AI denoising processing were applied to 516 DWIBS images using two AI models (noise2noise and noise2clear), where 456 images were used for training and 60 images were used for testing. The image quality of the original images and processed images by the two AI models were assessed by SSIM (Structural similarity) values. Results:The SSIM values for original and noise2clear and noise2noise-processed images were 0.4, 0.804, and 0.907 respectively.Conclusion:AI Denoising processing with noise2noise as well as noise2clear was useful for whole body MR imaging by DWIBS., Japanese Society for Medical and Biological Engineering

AI技術の医療分野における活用の現状と未来 : 放射線診断学の視点から—特集 形成外科で人工知能(AI)・バーチャルリアリティー(VR)を活用する! ; 人工知能(AI)・機械学習・ディープラーニング
Oct. 2020

Apply Deep Learning to Radiological Technology Research Generate
平原大助
日本放射線技術学会総会学術大会予稿集, 2020

乳がん検診としてのMRI検査(+診断補助AI)の可能性
2020

非造影MRIを用いた「痛くない・恥ずかしくない乳がん検診」の設計と実現
2020

無症候性野球競技者における棘上筋筋力と棘上筋volumeとの関係性
2020

無症候性野球競技者における棘下筋Volumeと2nd肢位での外旋筋力との関係性―Fijiを用いた三次元的解析―
2020

無症候性野球競技者における棘下筋Volumeの三次元的分析-画像処理ソフトFuji(by Image J)を用いて-
2019

MRIのRFパルスによる火傷の影響評価
2019

乳腺超音波の診断補助用AIの開発と性能評価
2019

今から始めるディープラーニング-基礎から実践まで,押さえておきたいポイント II ディープラーニングによる研究に向けて 1.ディープラーニングで必要となるハードウエア,ソフトウエア,プラットフォーム,プログラミング言語
2019

大腿四頭筋セッティングは膝蓋上嚢の柔軟性に影響を与えるか－核磁気共鳴画像法(MRI)による解析にて筋力と可動域の関係生を紐解く－
2019

新時代の乳腺画像診断 MRI:乳房MRIの現状と将来展望について
2019

Body DWIにおける至適脂肪抑制法の初期検討
2016

CXDIの機種の違いによるREX値の比較
2015

フラットパネルディテクタの蛍光体層の違いによるREX値およびEI値の比較
2015

マルチショットEPI法を用いたFLAIR法の最適化
2008

174 グラジエントエコータイプEPIの有用性の検討(MR検査磁化率強調画像, 第35回日本放射線技術学会秋季学術大会プログラム)
20 Sep. 2007

Techniques and Methods for Safe and Effective Clipping of Aneurysmal Neck
NISHI Toru; KOGA Kazunari; YAMASHIRO Shigeo; HAMADA Kazuya; GOTO Tomoaki; KAJI Masatomo; MUTA Daisuke; HIRABARU Masashi; HIRAYAMA Takahisa; KURATSU Jun-ichi; FUJIOKA Shodo
Surgery for Cerebral Stroke, 2007
Clipping an aneurysmal neck is considered an optimal treatment modality for ruptured and unruptured cerebral aneurysms. However, various anatomical variations in the relationships of parent artery, branched arteries, shape of the aneurysmal dome and aneurysmal neck frequently make it difficult to apply a simple clipping method using a single clip. To achieve safe and effective neck clipping, we use several techniques such as the clipping on wrapping method (COW method), tentative parallel clipping method (TPC method) and combination clipping with fenestrated clips (CCFC method). In this study, we analyzed the safety and effectiveness of these methods. In 515 surgeries for cerebral aneurysm performed between January 2001 and December 2003, the COW method, TPC method and CCFC method were used in 61 cases, 11 cases and 26 cases, respectively. In patients treated by the COW method, slight stenosis of the parent arteries was found on long-term follow-up studies in only 2 patients. In patients treated by the TPC method and/or CCFC method, complete obliteration of the aneurysm and preservation of blood flow through the parent arteries and branched arteries were confirmed.
These 3 techniques are safe and effective methods for aneurysmal neck clipping.
, The Japanese Society on Surgery for Cerebral Stroke

グラジエントエコータイプEPIの有用性の検討
2007

Internal Carotid Aneurysm with Circumferential Dilatation
FUJIOKA Shodo; NISHI Toru; KOGA Kazunari; YAMASHIRO Shigeo; YAMAMOTO Haruaki; GOTO Tomoaki; KAJI Masatomo; MUTA Daisuke; HIRABARU Masashi; HIRAYAMA Takahisa; KURATSU Junichi
Surgery for Cerebral Stroke, 2006
We present 3 cases of ruptured aneurysms with circumferential dilatation of the internal carotid artery (IC). Two of the cases presented repeated hemorrhage following neck clipping of the berry aneurysms arising from the anterior wall of IC. In the second surgery, it was confirmed that the newly formed aneurysm extended circumferentially from the primary aneurysm, which had been obliterated completely in the initial surgery.
Case 1 died of rerupture of the aneurysm following the second fundus-wrapping surgery. In Case 2, the aneurysm was obliterated safely using a fenestrated clip in the second surgery. Case 3 with subarachnoid hemorrhage underwent craniotomy because of progressive dilatation of C2 segment. Intraoperatively multiple blister-like bulging spots were identified on the surface of circumferentially dilated IC. It was repaired by so-called clipping on wrapping method. In the surgery, the clip produced mild narrowing of IC in spite of careful application of the clip to avoid kink. This finding suggested that this type of aneurysm was formed by a special mechanism such as arterial dissection.
We stress the necessity of close postoperative observation even if neck obliteration seems to have been complete in the IC anterior wall aneurysm. Furthermore we recommend extraordinary procedures such as a combination of trapping and bypass surgery or the so-called clipping on wrapping method in this type of aneurysm.
, The Japanese Society on Surgery for Cerebral Stroke

インナービジョン　　　　　　　　　　　　　　　
Single work
25 Jun. 2025
9784910561615

サクッとわかる医療AI
Joint editor
11 Jun. 2022
4908083797

Rad Fan 2021 JULY Vol.19 No.7　　　　　　　　　　　　　　　
Daisuke Hirahara, Single work
30 Jun. 2021

Rad Fan 2021 JULY Vol.19 No.7　　　　　　　　　　　　　　　
Daisuke Hirahara, Single work
30 Jun. 2021

MRI応用自在 第4版　　　　　　　　　　　　　　　
Joint work
01 Mar. 2021

PEPARS　　　　　　　　　　　　　　　
Joint work
Oct. 2020

2020-2021年版 はじめての医用画像ディープラーニング -基礎・応用・事例- (医療AIとディープラーニングシリーズ)　　　　　　　　　　　　　　　
Joint work
May 2020

学ぶ！ 究める！ 医療AI─ディープラーニングの基礎から研究最前線まで─ (iv-MOOK vol.1)　　　　　　　　　　　　　　　
Joint work
May 2020

Medical AI and deep learning series. Introduction to medical image processing using Python.　　　　　　　　　　　　　　　
Joint editor
06 Apr. 2020

今から始めるディープラーニング　　　　　　　　　　　　　　　
Daisuke Hirahara, Contributor, ディープラーニングで必要となるハードウェア,ソフトウェア,プラットフォーム,プログラミング言語
INNERVISION, Nov. 2019

診療放射線技師 実用 基礎数学　　　　　　　　　　　　　　　
Joint work
Apr. 2018

生成AIに“伝わる”プロンプト入門 〜論文検索・管理とNotebookLMで踏み出す研究の第一歩〜　　　　　　　　　　　　　　　
25 Oct. 2026, [Invited]

AIリテラシーと正しいワークフロー　　　　　　　　　　　　　　　
16 Oct. 2026, [Invited]

MRI motion artifact補正と不確実性推定のためのbilateral filterを組み込んだmixture of attention U-Net　　　　　　　　　　　　　　　
19 Apr. 2026

AIは画像を「復元」しているのか、「創造」しているのか？～962報の先行研究から紐解く非線形処理評価の真実～　　　　　　　　　　　　　　　
28 Feb. 2026, [Invited]

診療支援・業務効率化AIをその場で実装する ─ DifyとAntigravityを使ったアプリ開発実践　　　　　　　　　　　　　　　
24 Feb. 2026

AIエージェント活用例と人材育成の重要性　　　　　　　　　　　　　　　
04 Feb. 2026, [Invited]

自分だけのAIを作るためのDX戦略 準備編 －無料ツールで始める業務改革の第一歩－　　　　　　　　　　　　　　　
03 Feb. 2026

医療DX推進へ プログラミング不要でAIアプリケーションを作ってみよう！　　　　　　　　　　　　　　　
24 Jan. 2026, [Invited]

医療・介護現場のDX戦略 無料AIツールで始める業務改革　　　　　　　　　　　　　　　
22 Nov. 2025, [Invited]

Strategy for AI 放射線技師に求められるリテラシーとその教育　　　　　　　　　　　　　　　
01 Nov. 2025, [Invited]

AIが変える放射線技師の未来 －日常業務から専門分野まで、“生成AI”活用の最前線－　　　　　　　　　　　　　　　
27 Sep. 2025

AIが変える放射線技師の未来 －日常業務から専門分野まで、“生成AI”活用の最前線－　　　　　　　　　　　　　　　
24 Aug. 2025, [Invited]

医療・介護現場のDX戦略 無料AIツールで始める業務改革　　　　　　　　　　　　　　　
19 Aug. 2025

特別講演１「AIと循環制御医学の融合 ―研究イノベーションから臨床実装まで―」　　　　　　　　　　　　　　　
20 Jun. 2025, [Invited]

医療DX推進のための Dify ワークショップ　　　　　　　　　　　　　　　
23 May 2025, [Invited]

医用画像解析の未来を描くマルチモーダル時代の LLM 入門　　　　　　　　　　　　　　　
13 Apr. 2025

医用画像における AI の軌跡と未来への展望　　　　　　　　　　　　　　　
12 Apr. 2025

平原大助先生と話そう！「Difyで自分のための生成AIを作ろう！」　　　　　　　　　　　　　　　
11 Mar. 2025

研究紹介 & LLM研究活用法　　　　　　　　　　　　　　　
17 Feb. 2025, [Invited]

放射線技師のための生成AI活用講座　　　　　　　　　　　　　　　
18 Jan. 2025, [Invited]

放射線技師のための生成AI活用講座：ChatGPTから始まったLLMの可能性　　　　　　　　　　　　　　　
02 Dec. 2024

AIが拓くMRI画像の新時代：施設間差異の壁を越える標準化の可能性と未来　　　　　　　　　　　　　　　
30 Nov. 2024

放射線技師のための⽣成AI 活⽤講座：ChatGPTから始まったLLMの可能性　　　　　　　　　　　　　　　
19 Oct. 2024, [Invited]

医療画像分析の未来 －ノーコード・ローコードアプローチの活用－　　　　　　　　　　　　　　　
23 Feb. 2024

実践：ノーコード & ローコード　　　　　　　　　　　　　　　
30 Jan. 2024

乳癌検診へのRadiomics応用の可能性　　　　　　　　　　　　　　　
25 Nov. 2023

指先毛細血管オプティカルフローによる 感染症のスクリーニング手法　　　　　　　　　　　　　　　
24 Nov. 2023

ヒトの心拍変動指標の個人識別性ー機械学習を用いたサル指標との比較　　　　　　　　　　　　　　　
23 Nov. 2023

実践：ノーコード　　　　　　　　　　　　　　　
07 Nov. 2023

医療AIにおける解析手法を学ぶ　　　　　　　　　　　　　　　
07 Sep. 2023, [Invited]

未来予想図を片手にAIの世界に入り込みました　　　　　　　　　　　　　　　
27 May 2023, [Invited]

Examples of AI Development using DGX and MONAI, and Multi-Parametric Diagnostic Aid AI for Breakthrough Breast Cancer Screening Tests　　　　　　　　　　　　　　　
Daisuke Hirahara
22 Mar. 2023

GenZer教育にも役立つ医療統計　　　　　　　　　　　　　　　
07 Mar. 2023

web3とは？　　　　　　　　　　　　　　　
16 Jan. 2023

乳腺領域へのAI応用の歴史とこれから　　　　　　　　　　　　　　　
22 Oct. 2022, [Invited]

Often overlooked AI fundamentals and AI in the MRI field　　　　　　　　　　　　　　　
Daisuke Hirahara
JSMRM2022, 11 Sep. 2022

AI 人材育成を実行するための、教育現場から見たNVIDIAソリューション　　　　　　　　　　　　　　　
23 Jun. 2022, [Invited]

AI 人材育成を実行するための、教育現場から見たNVIDIAソリューション　　　　　　　　　　　　　　　
22 Mar. 2022, [Invited]

看護基礎教育におけるICT活用とマネジメント ～その２ 基礎から実践へ～　　　　　　　　　　　　　　　
24 Feb. 2022

AI Course for Medical Professionals　　　　　　　　　　　　　　　
Daisuke Hirahara
Feb. 2022, [Invited]

プログラミング不要でAIを作ってみよう！実践編（分類）　　　　　　　　　　　　　　　
25 Jan. 2022

プログラミング不要でAIを作ってみよう！準備編・実践編（回帰）　　　　　　　　　　　　　　　
11 Jan. 2022

AI Course for Medical Professionals　　　　　　　　　　　　　　　
Daisuke Hirahara
11 Dec. 2021, [Invited]

実践報告ICTを活用した授業の工夫　　　　　　　　　　　　　　　
02 Dec. 2021

DWIBSの短所を解決する人工知能の利用 －ノイズ除去AIを作ってみよう－　　　　　　　　　　　　　　　
03 Oct. 2021

DWIBSの短所を解決する人工知能の利用　　　　　　　　　　　　　　　
Sep. 2021, [Invited]

Comparison of noise2noise and noise2clear in MRI images　　　　　　　　　　　　　　　
The 60th Annual Conference of Japanese Society for Medical and Biological Engineering, 16 Jun. 2021

自然言語・音声認識　　　　　　　　　　　　　　　
09 Mar. 2021

Fundamentals of Artificial Intelligence I (Introduction to Machine Learning)　　　　　　　　　　　　　　　
Daisuke Hirahara
19 Feb. 2021, [Invited]

AI imaging diagnosis in MRI field　　　　　　　　　　　　　　　
Daisuke Hirahara
13 Feb. 2021, [Invited]

なぜAIを学ぶ必要があるのか？　　　　　　　　　　　　　　　
Feb. 2021

RNN　　　　　　　　　　　　　　　
26 Jan. 2021

Possibility of MRI examination (+ diagnostic aid AI) as breast cancer screening　　　　　　　　　　　　　　　
Daisuke Hirahara
23 Nov. 2020, [Invited]

AIって何？（機械学習概論）　　　　　　　　　　　　　　　
27 Oct. 2020

医用画像領域における超解像と画像生成AI　　　　　　　　　　　　　　　
21 Jul. 2020

Evaluation of breast tomosynthesis images for reducing radiation exposure: Preliminary investigation.　　　　　　　　　　　　　　　
Nobuyoshi Tanki; Daisuke Hirahara; Natsuki Tani; Nao Nishiura; Tatsuhiro Gotanda; Rumi Gotanda; Shinya Imai; Yasuyuki Kawaji; Toshizo Katsuda
European Congress of Radiology 2020, Jul. 2020

Creating medical AI (super-resolution and image generation)　　　　　　　　　　　　　　　
Daisuke Hirahara
10 Apr. 2020, [Invited]

深層学習でDWIBS画像の診断分類と画像生成　　　　　　　　　　　　　　　
26 Jan. 2020, [Invited]

深層学習の主な種類と実際の事例　　　　　　　　　　　　　　　
24 Jan. 2020, [Invited]

AIを使ったDWIBSの応用「医療AIの実装と事例」　　　　　　　　　　　　　　　
26 Oct. 2019

医療 AI 実装と事例　　　　　　　　　　　　　　　
02 Jul. 2019

特別企画5:AI実践講座 放射線科は人工知能をどのように学習すればよいかを教えよう！放射線科医の皆さん、人工知能をどのように勉強しますか？　　　　　　　　　　　　　　　
14 Apr. 2019

Fundamental study on preliminary image processing at time development of CNN using chest radiography　　　　　　　　　　　　　　　
Daisuke Hirahara
IEEE LifeTech2019, 14 Mar. 2019

MRI画像を用いた診断補助用AIの開発と性能評価　　　　　　　　　　　　　　　
Feb. 2019

DWIBS画像を用いた乳がん診断補助用AI（悪性疑い検出）の開発と性能評価　　　　　　　　　　　　　　　
Feb. 2019

DWIBSによる乳がん検診のAI診断の試み　　　　　　　　　　　　　　　
Feb. 2019, [Invited]

海馬および海馬傍回の萎縮度判定解析ソフトにおける永久磁石装置撮像パラメータ標準化に関する研究　　　　　　　　　　　　　　　
Nov. 2018

Practice of programming education at medical college　　　　　　　　　　　　　　　
Daisuke Hirahara
25 Aug. 2018

Intestinal content signal suppression method in diffusion weighted images　　　　　　　　　　　　　　　
Daisuke Hirahara
10 Feb. 2018

Fundamental study on optimization of fat suppression method and improvement of distortion in diffusion emphasized image of permanent magnet device　　　　　　　　　　　　　　　
Daisuke Hirahara
Sep. 2017

BodyDWIにおける至適脂肪抑制法の初期検討　　　　　　　　　　　　　　　
Apr. 2016

日立ECHELON RXにおけるBodyDWI　　　　　　　　　　　　　　　
Feb. 2016

フラットパネルディテクタの蛍光体層の違いによるREX値及びEI値の比較　　　　　　　　　　　　　　　
Nov. 2015

DIXON法とDWI-腹部領域を中心に-　　　　　　　　　　　　　　　
Oct. 2015

DIXON法とsubtraction image法における基礎的な比較検討　　　　　　　　　　　　　　　
Sep. 2015

Mac OS用フリーソフトOsiriXを用いたComputed DWI(high b value)画像の検討　　　　　　　　　　　　　　　
Sep. 2014

VSRADにおける入力方向の比較検討　　　　　　　　　　　　　　　
Sep. 2013

ECHELON Vega 1.5T（日立メディコ）におけるVSRADの検討　　　　　　　　　　　　　　　
Sep. 2012

肝臓Gd-EOB-DTPA検査における肝細胞相T1W-Bright Blood法の開発と検討　　　　　　　　　　　　　　　
Sep. 2011

頭部MRI造影における3D-IR－BTFE法と従来法との比較検討　　　　　　　　　　　　　　　
Jul. 2010

頭部造影検査におけるIR付加 ３D－BTFEの有用性　　　　　　　　　　　　　　　
Sep. 2009

マルチショットEPI法を用いたFLAIR法の最適化　　　　　　　　　　　　　　　
Apr. 2008

グラジエントエコータイプEPIの有用性の検討第２報　　　　　　　　　　　　　　　
Nov. 2007

EPI-FLAIR法の有用性　　　　　　　　　　　　　　　
Sep. 2007

グラジエントエコータイプEPIの有用性の検討　　　　　　　　　　　　　　　
Sep. 2007

グラジエントエコータイプEPIの有用性の検討　　　　　　　　　　　　　　　
Apr. 2007

空間的/質的/経時的評価を可能とするbiomarker樹立：全身DWIのdelta-radiomics解析
01 Apr. 2023 - 31 Mar. 2026

Development of Portal Blood Flow Evaluation and Anomaly Detection Methods for Portal Hypertension Using CFD and Artificial Intelligence
Grant-in-Aid for Scientific Research (C)
Saitama Medical University
01 Apr. 2020 - 31 Mar. 2023
A hemodynamic model of the portal venous system and portal hypertension (portal hypertension model) was created on a computer, and simulations were performed under physiological and virtual treatment conditions to investigate the impact on portal venous blood flow. The results are as follows:


When portal vein embolization was performed, an increase in WSS and OSI was observed in the portal vein of the non-embolized lobe, suggesting a correlation with compensatory hypertrophy.In cases of malignant main portal vein stenosis, a pressure gradient occurred before and after the stenosis, resulting in decreased blood flow reaching the periphery. In cases of portal vein thrombosis, there was a tendency for higher WSS at the site of thrombosis.The sclerosing agent demonstrated a pushing effect, filling the inside of the blood vessel.It is believed to have contributed to a better understanding of the pathophysiology of the portal venous system.