Yoshiharu Horie

Yoshiharu Horie

Auto-generated profile, based on publicly available information

Yoshiharu Horie

Medical Department AstraZeneca K.K. Osaka Japan

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 July 2021 in JMIR Medical Informatics
Verified
Journal Article
Open access
SMSHIGEO MURO
SHIGEO MURO
MIMasato Ishida
Masato Ishida
YHYoshiharu Horie
Yoshiharu Horie
WTWataru Takeuchi
Wataru Takeuchi
SNShunki Nakagawa
Shunki Nakagawa
Tetsuhisa KitamuraTetsuhisa Kitamura
Tetsuhisa Kitamura
Background: Airflow limitation is a critical physiological feature in chronic obstructive pulmonary disease (COPD), for which long-term exposure to noxious substances, including tobacco smoke, is an established risk. However, not all long-term smokers develop COPD, meaning that other risk factors exist. Objective: This study aimed to predict the risk factors for COPD diagnosis using machine learning in an annual medical check-up database. Methods: In this retrospective observational cohort study (ARTDECO [Analysis of Risk Factors to Detect COPD]), annual medical check-up records for all Hitachi Ltd employees in Japan collected from April 1998 to March 2019 were analyzed. Employees who provided informed consent via an opt-out model were screened and those aged 30 to 75 years without a prior diagnosis of COPD/asthma or a history of cancer were included. The database included clinical measurements (eg, pulmonary function tests) and questionnaire responses. To predict the risk factors for COPD diagnosis within a 3-year period, the Gradient Boosting Decision Tree machine learning (XGBoost) method was applied as a primary approach, with logistic regression as a secondary method. A diagnosis of COPD was made when the ratio of the prebronchodilator forced expiratory volume in 1 second (FEV1) to prebronchodilator forced vital capacity (FVC) was <0.7 during two consecutive examinations. Results: Of the 26,101 individuals screened, 1213 met the exclusion criteria, and thus, 24,815 individuals were included in the analysis. The top 10 predictors for COPD diagnosis were FEV1/FVC, smoking status, allergic symptoms, cough, pack years, hemoglobin A1c, serum albumin, mean corpuscular volume, percent predicted vital capacity, and percent predicted value of FEV1. The areas under the receiver operating characteristic curves of the XGBoost model and the logistic regression model were 0.956 and 0.943, respectively. Conclusions: Using a machine learning model in this longitudinal database, we identified a number of parameters as risk factors other than smoking exposure or lung function to support general practitioners and occupational health physicians to predict the development of COPD. Further research to confirm our results is warranted, as our analysis involved a database used only in Japan.

23 July 2020 in Movement Disorders Clinical Practice
Verified
Journal Article
Open access
PDF file
SSShoichi Sasaki
Shoichi Sasaki
YHYoshiharu Horie
Yoshiharu Horie
Background: The association between olfactory dysfunction and disease duration and severity in Parkinson's disease (PD) remains controversial.Objective: The objective of this study was to examine the relationship between olfactory dysfunction and disease severity and duration in patients with recently diagnosed parkinsonism and patients with PD with a previous diagnosis.Methods: Olfactory function was evaluated in 79 patients with recently diagnosed parkinsonism, 71 patients with PD with a previous diagnosis—with patients in both groups free of cognitive impairment—and 128 age‐matched controls. The Odor‐Stick Identification Test for Japanese score was counted as the numbers of correct answers, responses of indistinguishable, and responses of odorless. Parkinsonism was evaluated using the Movement Disorder Society Criteria, the Unified Parkinson Disease Rating Scale (UPDRS) Part III, and 123iodine‐labeled N‐(3‐fluoropropyl)‐2β‐carbomethoxy‐3β‐(4‐iodophenyl) nortropane single photon emission computed tomography (DaTscan).Results: In the patients with recently diagnosed parkinsonism having the UPDRS Part III score ≥5 (mean [standard deviation: SD] score: 6.3 [1.9]) and with a positive DaTscan, the mean (SD) numbers of correct answers, responses of indistinguishable and responses of odorless were 4.3 (2.2), 1.6 (2.0), and 1.2 (2.2), respectively. In patients with PD with a previous diagnosis (mean [SD] UPDRS Part III score: 10.9 [3.2]), these numbers were 2.5 (2.2), 2.2 (2.5), and 3.8 (4.6), respectively. The patients with PD with a previous diagnosis showed more significant deterioration than the patients with recently diagnosed parkinsonism in the numbers of correct answers and responses of odorless (P < 0.0001). Olfaction in the combined patient group was significantly impaired compared with age‐matched controls in each category (P < 0.0001).Conclusions: These findings imply a close association between olfactory dysfunction and disease severity and duration in PD.

8 April 2026 in International Journal of Cancer
Verified
Journal Article
Open access
RKRyo Koto
Ryo Koto
AYAtsuki Yamamoto
Atsuki Yamamoto
ITIkumi Takashima
Ikumi Takashima
CIChikako Iwao
Chikako Iwao
JSJung‐Ho Shin
Jung‐Ho Shin
YHYoshiharu Horie
Yoshiharu Horie
Greater understanding of the prognosis for patients with lung cancer in Japan according to detection pathway would clarify the effect of population‐based screening programs and the related diagnostic environment. We explored the clinical characteristics and net survival (NS) of patients with lung cancer by detection pathway with Japanese National Cancer Registry (NCR) database of patients registered between 2016 and 2020. Eligible patients were categorized based on detection pathway: Screening, Incidental, and Other groups. We estimated 1, 2, and 3 year NS. Of 565,044 patients with lung cancer in the NCR, 563,244 met the eligibility criteria and were included in this study: 95,081 patients in the Screening (16.9%), 254,569 in the Incidental (45.2%), and 200,850 in the Other groups (35.7%). In the Screening, Incidental, and Other groups, 69.5%, 88.6%, and 83.4% of patients, respectively, were aged ≥ 65 years, and 50.5%, 49.6%, and 14.0% of patients, respectively, had localized disease at diagnosis. The most common initial treatment was surgery in the Screening (62.5%) and Incidental (46.2%) groups, while chemotherapy was most frequently used in the Other group (48.6%). The 3 year NS was 75.2% (95% confidence interval: 74.9%–75.6%) in the Screening group, 56.3% (56.0%–56.6%) in the Incidental group, and 29.4% (29.2%–29.7%) in the Other group. In the subgroup analysis, 3 year NS for localized disease was 96.3% (95.9%–96.7%), 83.9% (83.6%–84.3%), and 76.9% (76.1%–77.6%), respectively, for the Screening, Incidental, and Other groups. Screening may play a key role in early detection and is likely to contribute to survival outcomes.

20 November 2025 in Cancer Medicine
Verified
Journal Article
Open access
KFKazuki Fukuzawa
Kazuki Fukuzawa
YSYoshimitsu Shimomura
Yoshimitsu Shimomura
LZLing Zha
Ling Zha
HSHaruka Shida
Haruka Shida
MHManabu Hayama
Manabu Hayama
YHYoshiharu Horie
Yoshiharu Horie
Background: Patients with lung cancer tend to have a high incidence of arterial thromboembolism (ATE). However, although ATE is a life‐threatening disease, there are insufficient studies evaluating patient characteristics associated with a high risk of ATE in patients with lung cancer. This study aimed to examine the incidence of ATE in Japanese patients with lung cancer and evaluate the incidence of ATE by histological types and treatment patterns to identify high‐risk groups. Methods: We conducted a retrospective cohort study using the National Health Insurance (NHI), Employees' Health Insurance (EHI), and Later‐Stage Elderly Healthcare System (LSEHS) claims databases from April 1, 2014, to March 31, 2022. The patients were followed up for up to 5 years after lung cancer diagnosis. The cumulative incidence rates of ATE were calculated, and the histological subtype and treatment pattern were analyzed using the Cox proportional hazards model. Results: Among 6340 patients in EHI, 10,857 patients in NHI, and 13,039 patients in LSEHS, patient characteristics such as the median age and the prevalence of comorbidities were different among the three databases. The cumulative incidence rates at 1 year were 1.7%, 4.5%, and 7.2%, and those at 5 years were 3.5%, 8.0%, and 15.9%. The adjusted hazard ratio for small cell lung cancer compared to non‐small cell lung cancer was 2.84 (95% confidence interval (CI): 1.46–5.54) in EHI, 0.93 (95% CI: 0.65–1.32) in NHI, and 1.11 (95% CI: 0.85–1.46) in LSEHS. The risk of ATE of RT + systemic therapy and systemic therapy only was higher than surgery only, with a statistically significant difference in the three databases. Conclusion: In any database, the cumulative incidence of ATE increased over the 5 year observation period. Patients who received RT + systemic therapy and those who received systemic therapy only had a higher risk of ATE than those who underwent surgery only.

5 October 2020 in JMIR Preprints
Verified
Preprint
Open access
SMSHIGEO MURO
SHIGEO MURO
MIMasato Ishida
Masato Ishida
YHYoshiharu Horie
Yoshiharu Horie
WTWataru Takeuchi
Wataru Takeuchi
SNShunki Nakagawa
Shunki Nakagawa
Tetsuhisa KitamuraTetsuhisa Kitamura
Tetsuhisa Kitamura
BACKGROUND Airflow limitation is a critical physiological feature in chronic obstructive pulmonary disease (COPD), for which long-term exposure to noxious substances, including tobacco smoke, is an established risk. However, not all long-term smokers develop COPD, meaning that other risk factors exist. OBJECTIVE This study aimed to predict the risk factors for COPD diagnosis using machine learning in an annual medical check-up database. METHODS In this retrospective observational cohort study (ARTDECO [Analysis of Risk Factors to Detect COPD]), annual medical check-up records for all Hitachi Ltd employees in Japan collected from April 1998 to March 2019 were analyzed. Employees who provided informed consent via an opt-out model were screened and those aged 30 to 75 years without a prior diagnosis of COPD/asthma or a history of cancer were included. The database included clinical measurements (eg, pulmonary function tests) and questionnaire responses. To predict the risk factors for COPD diagnosis within a 3-year period, the Gradient Boosting Decision Tree machine learning (XGBoost) method was applied as a primary approach, with logistic regression as a secondary method. A diagnosis of COPD was made when the ratio of the prebronchodilator forced expiratory volume in 1 second (FEV1) to prebronchodilator forced vital capacity (FVC) was <0.7 during two consecutive examinations. RESULTS Of the 26,101 individuals screened, 1213 met the exclusion criteria, and thus, 24,815 individuals were included in the analysis. The top 10 predictors for COPD diagnosis were FEV1/FVC, smoking status, allergic symptoms, cough, pack years, hemoglobin A1c, serum albumin, mean corpuscular volume, percent predicted vital capacity, and percent predicted value of FEV1. The areas under the receiver operating characteristic curves of the XGBoost model and the logistic regression model were 0.956 and 0.943, respectively. CONCLUSIONS Using a machine learning model in this longitudinal database, we identified a number of parameters as risk factors other than smoking exposure or lung function to support general practitioners and occupational health physicians to predict the development of COPD. Further research to confirm our results is warranted, as our analysis involved a database used only in Japan.

27 October 2025 in BMC Cancer
Verified
Journal Article
Open access
PDF file
THTakehiro Hirai
Takehiro Hirai
YHYoshiharu Horie
Yoshiharu Horie
TKTetsuhisa Kitamura
Tetsuhisa Kitamura
Lung cancer is one of the most common cancers worldwide and is associated with a high mortality rate. Addressing unmet medical needs in lung cancer treatments from the perspective of patient preferences has become an important research focus. Discrete Choice Experiments (DCEs) are commonly used to assess patient preferences, therefore it is essential to interpret their analysis methods and results appropriately. This scoping review was intended to examine DCE studies on patient preferences for lung cancer treatments published after the release of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) ESTIMATE checklist in 2016, and to identify challenges in applying the checklist. Literatures meeting the eligibility criteria for DCE studies on patient preferences for lung cancer treatments were identified through PubMed and Cochrane Library searches and reviewed independently by two reviewers according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) Scoping Review extension. All literatures were reviewed and assessed using the ISPOR ESTIMATE checklist. Twelve DCE studies published between January 2017 and June 2022 were identified. All studies used efficacy and side effects as attributes while less than half studies used cost. Our assessment revealed that all studies completed the checklists of Interpretation, Method, and Assumptions domains. However, the checklists in Evaluation, Stochastic, and Trade-offs domains were not completed in 83.3%, 41.7%, and 33.3% of the 12 studies, respectively. This suggests that further consideration is needed to ensure the appropriateness of the statistical analysis and interpretation of the results of the DCE studies for lung cancer treatments. Our scoping review using the ISPOR ESTIMATE checklists found that the consideration points remained the focus of Evaluation, Stochastic, and Trade-offs in DCE studies on patient preferences for lung cancer treatments.