Junji Kumasawa

Junji Kumasawa

Auto-generated profile, based on publicly available information

Junji Kumasawa

Department of Critical Care Medicine, Sakai City Medical Center, 1-1-1, Ebaraji-cho, Nishi-ku, Sakai, Osaka, 593-8304, Japan; Kyoto University Graduate School of Medicine, Human Health Science, Yoshida-Konoe-cho, Sakyo-ku, Kyoto, 606-8501, Japan

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
15 May 2017 in Clinical and Experimental Nephrology
Verified
Journal Article
Open access
PDF file
HOHiroaki Ogata
Hiroaki Ogata
JKJunji Kumasawa
Junji Kumasawa
SFShingo Fukuma
Shingo Fukuma
MMMasahide Mizobuchi
Masahide Mizobuchi
EKEriko Kinugasa
Eriko Kinugasa
TATadao Akizawa
Tadao Akizawa
Background The cardiothoracic ratio (CTR) is a non-invasive left ventricular hypertrophy index. However, whether CTR associates with cardiovascular disease (CVD) and mortality in hemodialysis (HD) populations is unclear. Methods Using a Mineral and Bone disorder Outcomes Study for Japanese CKD Stage 5D Patients (MBD-5D Study) subcohort, 2266 prevalent HD patients (age 62.8 years, female 38.0%, HD duration 9.4 years) with secondary hyperparathyroidism (SHPT) whose baseline CTR had been recorded were selected. We evaluated associations between CTR and all-cause death, CVD death, or composite events in HD patients. Results CTR was associated significantly with various background and laboratory characteristics. All-cause death, CVD-related death, and composite events increased across the CTR quartiles (Q). Adjusted hazard risk (HR) for all-cause death was 1.4 (95% confidential interval, 0.9–2.1) in Q2, 1.9 (1.3–2.9) in Q3, and 2.6 (1.7–4.0) in Q4, respectively (Q1 as a reference). The corresponding adjusted HR for CVD-related death was 1.8 (0.8–4.2), 3.1 (1.4–6.8), and 3.5 (1.6–7.9), and that for composite outcome was 1.2 (1.0–1.6), 1.7 (1.3–2.2), and 1.8 (1.5–2.3), respectively. Exploratory analysis revealed that there were relationships between CTR and age, sex, body mass index, comorbidity of CVD, dialysis duration, dialysate calcium level and intact parathyroid hormone, phosphorus, hemoglobin, and usage of phosphate binder. Conclusion CTR correlated with all-cause death, CVD death, and composite events in HD patients with SHPT.

24 July 2017 in Journal of Anesthesia
Verified
Journal Article
Closed access
PDF file
YKYutaka Kondo
Yutaka Kondo
JKJunji Kumasawa
Junji Kumasawa
AKAtsushi Kawaguchi
Atsushi Kawaguchi
RSRyutaro Seo
Ryutaro Seo
ENEishu Nango
Eishu Nango
SHSatoru Hashimoto
Satoru Hashimoto

1 November 2023 in medRxiv
Verified
Preprint
Open access
PDF file
Yuki KataokaYuki Kataoka
Yuki Kataoka
RSRyuhei So
Ryuhei So
Masahiro BannoMasahiro Banno
Masahiro Banno
JKJunji Kumasawa
Junji Kumasawa
HSHidehiro Someko
Hidehiro Someko
TFToshi A. Furukawa
Toshi A. Furukawa
Systematic reviews (SRs) are a critical component of evidence-based medicine, but the process of screening titles and abstracts is time-consuming. This study aimed to develop and externally validate a method using large language models to classify abstracts for diagnostic test accuracy (DTA) systematic reviews, thereby reducing the human workload. We used a previously collected dataset for developing DTA abstract classifiers and applied prompt engineering. We developed an optimized meta-prompt for Generative Pre-trained Transformer (GPT)-3.5-turbo and GPT-4 to classify abstracts. In the external validation dataset 1, the prompt with GPT-3.5 turbo showed a sensitivity of 0.988, and a specificity of 0.298. GPT-4 showed a sensitivity of 0.982, and a specificity of 0.677. In the external validation dataset 2, GPT-3.5 turbo showed a sensitivity of 0.919, and a specificity of 0.434. GPT-4 showed a sensitivity of 0.806, and a specificity of 0.740. If we included eligible studies from among the references of the identified studies, GPT-3.5 turbo had no critical misses, while GPT-4 had some misses. Our study indicates that GPT-3.5 turbo can be effectively used to classify abstracts for DTA systematic reviews. Further studies using other dataset are warranted to confirm our results. Additionally, we encourage the use of our framework and publicly available dataset for further exploration of more effective classifiers using other LLMs and prompts (https://github.com/youkiti/ARE/). What is already known - Title and abstract screening in systematic reviews (SRs) consumes significant time. - Several attempts using machine learning to reduce this process in diagnostic test accuracy (DTA) SRs exist, but they have not yielded positive results in external validation. What is new - We aimed to develop and externally validate optimized meta-prompt for GPT-3.5-turbo and GPT-4 to classify abstracts for DTA SRs. - Through an iterative approach across three training datasets, an optimal meta-prompt capable of identifying DTA studies with remarkable sensitivity and specificity was developed. - The accuracy reproduced in the external validation datasets. Potential Impact for Readers - The developed meta-prompt can lessen the need for humans to read abstracts for DTA SRs, saving significant time and resources.


17 September 2013 in Annals of Internal Medicine
Verified
Journal Article
Closed access
PDF file
JKJunji Kumasawa
Junji Kumasawa
NKNoriaki Kurita
Noriaki Kurita
SFShunichi Fukuhara
Shunichi Fukuhara

21 April 2016 in BMC Research Notes
Verified
Journal Article
Open access
PDF file
NKNoriaki Kurita
Noriaki Kurita
MMMinoru Murakami
Minoru Murakami
SSSayaka Shimizu
Sayaka Shimizu
JKJunji Kumasawa
Junji Kumasawa
TATeruhisa Azuma
Teruhisa Azuma
SFShunichi Fukuhara
Shunichi Fukuhara
Desire to attend graduate school for academic research training following the mandatory two-year clinical internship is unknown among young Japanese physicians who work at community hospitals after their internship. The aim of this study is to determine opinions and factors regarding pursuing higher education through graduate school among young physicians who work at community hospitals after their two-year internship. This cross-sectional survey was conducted among young physicians working at community hospitals after their two-year internship. We examined the percentage of young physicians considering higher education through graduate school, the planned timing and field of enrollment among those wanting to enroll, and reasons for not continuing their education among those with no such plans. The association between desire to enroll in graduate school and background characteristics was examined using modified least-squares regression to estimate proportion difference. Among 127 (73.2 % internal medicine specialists, median age 30 years) physicians in 33 hospitals, 71 (55.9 %) stated that they wished to enroll in graduate school. The most frequently reported timing was 7–8 years after graduation from medical school. Those who stated no desire to attend graduate school cited concerns about the quality of training or not having enough knowledge to choose an appropriate laboratory or field, among other reasons. Increased number of years since graduating medical school [adjusted proportion difference (PD) −6.0 %, 95 % confidence interval (95 % CI) −9.8 to −2.3 %], being a woman with children [adjusted PD −53.4 %, 95 % CI −87.3 to −19.5 % (vs. a man not having children)], and completing their two-year internship at both university and community hospitals [adjusted PD −40.3 %, 95 % CI −72.5 to −8.0 % (vs. internship only at community hospitals)] were associated with a reduction in desire to enroll in graduate school. We identified a growing trend in desire among young physicians to attend graduate school. Attracting those young physicians who express no desire to attend graduate school, however, will require establishment of more flexible graduate school programs which address their concerns.