Yu Li

Yu Li

Auto-generated profile, based on publicly available information

Yu Li

General Medicine, Chiba University Hospital, Chiba, JPN

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
8 April 2024 in JMIR Medical Education
Verified
Journal Article
Open access
FFFumitoshi Fukuzawa
Fumitoshi Fukuzawa
YYYasutaka Yanagita
Yasutaka Yanagita
DYDaiki Yokokawa
Daiki Yokokawa
SUShun Uchida
Shun Uchida
SYShiho Yamashita
Shiho Yamashita
MIMasatomi Ikusaka
Masatomi Ikusaka
Background: Medical history contributes approximately 80% to a diagnosis, although physical examinations and laboratory investigations increase a physician’s confidence in the medical diagnosis. The concept of artificial intelligence (AI) was first proposed more than 70 years ago. Recently, its role in various fields of medicine has grown remarkably. However, no studies have evaluated the importance of patient history in AI-assisted medical diagnosis. Objective: This study explored the contribution of patient history to AI-assisted medical diagnoses and assessed the accuracy of ChatGPT in reaching a clinical diagnosis based on the medical history provided. Methods: Using clinical vignettes of 30 cases identified in The BMJ , we evaluated the accuracy of diagnoses generated by ChatGPT. We compared the diagnoses made by ChatGPT based solely on medical history with the correct diagnoses. We also compared the diagnoses made by ChatGPT after incorporating additional physical examination findings and laboratory data alongside history with the correct diagnoses. Results: ChatGPT accurately diagnosed 76.6% (23/30) of the cases with only the medical history, consistent with previous research targeting physicians. We also found that this rate was 93.3% (28/30) when additional information was included. Conclusions: Although adding additional information improves diagnostic accuracy, patient history remains a significant factor in AI-assisted medical diagnosis. Thus, when using AI in medical diagnosis, it is crucial to include pertinent and correct patient histories for an accurate diagnosis. Our findings emphasize the continued significance of patient history in clinical diagnoses in this age and highlight the need for its integration into AI-assisted medical diagnosis systems.

1 November 2017 in Cleveland Clinic Journal of Medicine
Verified
Journal Article
Closed access
YLYu Li
Yu Li
HFHiraku Funakoshi
Hiraku Funakoshi
TSTakashi Shiga
Takashi Shiga
SFShigeki Fujitani
Shigeki Fujitani
A 52-year-old woman with diabetes presents with pain in her lower abdomen and back.

14 July 2022 in Journal of General and Family Medicine
Verified
Journal Article
Open access
YLYu Li
Yu Li
KSKiyoshi Shikino
Kiyoshi Shikino
JTJiro Terada
Jiro Terada
YKYusuke Katsumata
Yusuke Katsumata
TKToru Kinouchi
Toru Kinouchi
MIMasatomi Ikusaka
Masatomi Ikusaka
Background: We aimed to assess differences in health literacy between those who improved CPAP non‐adherent and those who remained non‐adherent.Methods: We included patients newly diagnosed with sleep apnea syndrome who had started CPAP therapy between February 2019 and October 2020 with ≥6 follow‐up months or who self‐interrupted CPAP therapy <6 months. We recorded the CPAP wearing time after 3 and 6 months. Patients were divided into the CPAP adherent (using CPAP for ≥4 h per night) and non‐adherent (self‐interrupted CPAP therapy/using CPAP for <4 h per night) groups. We compared the European Health Literacy Survey Questionnaire 47 (HLS‐EU‐Q47) score between those who were CPAP non‐adherent after 3 months and become CPAP adherent after 6 months, and those who remained non‐adherent after 6 months.Results: At 3 months, 34 patients were CPAP non‐adherent. After 6 months, there were 7 and 27 patients in the CPAP adherent and non‐adherent groups, respectively. There was a significant difference in the HLS‐EU‐Q47 score between the patients who became adherent to CPAP and who remained non‐adherent after 6 months.Conclusion: Previously non‐adherent patients who subsequently became adherent tended to have higher health literacy.

23 February 2024 in Diagnosis
Verified
Journal Article
Closed access
DYDaiki Yokokawa
Daiki Yokokawa
YYYasutaka Yanagita
Yasutaka Yanagita
YLYu Li
Yu Li
SYShiho Yamashita
Shiho Yamashita
KSKiyoshi Shikino
Kiyoshi Shikino
MIMasatomi Ikusaka
Masatomi Ikusaka
Article For any disease a human can imagine, ChatGPT can generate a fake report was published on February 23, 2024 in the journal Diagnosis (volume 0, issue 0).

31 January 2025 in JMIR Medical Education
Verified
Journal Article
Open access
KSKiyoshi Shikino
Kiyoshi Shikino
KYKazuyo Yamauchi
Kazuyo Yamauchi
NANobuyuki Araki
Nobuyuki Araki
ISIkuo Shimizu
Ikuo Shimizu
HKHajime Kasai
Hajime Kasai
SIShoichi Ito
Shoichi Ito
Background: Japan faces a health care delivery challenge due to physician maldistribution, with insufficient physicians practicing in rural areas. This issue impacts health care access in remote areas and affects patient outcomes. Educational interventions targeting students’ career decision-making can potentially address this problem by promoting interest in rural medicine. We hypothesized that community-based problem-based learning (PBL) using real-patient videos could foster students’ understanding of community health care and encourage positive attitudes toward rural health care. Objective: This study investigated the impact of community-based PBL on medical students’ understanding and engagement with rural health care, focusing on their knowledge, skills, and career orientation. Methods: Participants were 113 fourth-year medical students from Chiba University, engaged in a transition course between preclinical and clinical clerkships from October 24 to November 2, 2023. The students were randomly divided into 16 groups (7-8 participants per group). Each group participated in two 3-hour PBL sessions per week over 2 consecutive weeks. Quantitative data were collected using pre- and postintervention questionnaires, comprehension tests, and tutor-assessed rubrics. Self-assessment questionnaires evaluated the students’ interest in community health care and their ability to envision community health care settings before and after the intervention. Qualitative data from the students’ semistructured interviews after the PBL sessions assessed the influence of PBL experience on clinical clerkship in community hospitals. Statistical analysis included median (IQR), effect sizes, and P values for quantitative outcomes. Thematic analysis was used for qualitative data. Results: Of the 113 participants, 71 (62.8%) were male and 42 (37.2%) female. The total comprehension test scores improved significantly (pretest: median 4.0, IQR 2.5-5.0; posttest: median 5, IQR 4-5; P<.001; effect size r=0.528). Rubric-based assessments showed increased knowledge application (pretest: median 8, IQR 7-9; posttest: median 8, IQR 8-8; P<.001; r=0.494) and self-directed learning (pretest: median 8, IQR 7-9; posttest: median 8, IQR 8-8; P<.001; r=0.553). Self-assessment questionnaires revealed significant improvements in the students’ interest in community health care (median 3, IQR 3-4 to median 4, IQR 3-4; P<.001) and their ability to envision community health care settings (median 3, IQR 3-4 to median 4, IQR 3-4; P<.001). Thematic analysis revealed key themes, such as “empathy in patient care,” “challenges in home health care,” and “professional identity formation.” Conclusions: Community-based PBL with real-patient videos effectively enhances medical students’ understanding of rural health care settings, clinician roles, and the social needs of rural patients. This approach holds potential as an educational strategy to address physician maldistribution. Although this study suggests potential for fostering positive attitudes toward rural health care, further research is needed to assess its long-term impact on students’ career trajectories.

26 February 2025 in BMC Medical Education
Verified
Journal Article
Open access
PDF file
JKJumpei Kojima
Jumpei Kojima
TUTakanori Uehara
Takanori Uehara
YOYoshiyuki Ohira
Yoshiyuki Ohira
TTTomoko Tsukamoto
Tomoko Tsukamoto
KSKiyoshi Shikino
Kiyoshi Shikino
MIMasatomi Ikusaka
Masatomi Ikusaka
The “5-microskills” instructional method for clinical reasoning does not incorporate a step for learners’ critical reflection on their predicted hypotheses. This study aimed to correct this shortcoming by inserting a third step in which learners conduct critical self-examinations and furnish evidence that contradicts their predicted hypotheses, resulting in the “6-microskills” method. In this cross-sectional study, changes in learners’ confidence in their predicted hypotheses were measured and examined to modify confirmation bias and diagnoses. A total of 108 medical students were presented with one randomly assigned clinical vignette from a set of eight, having to: (1) describe their first impression; (2) provide evidence for it; and (3) finally identify inconsistencies/state evidence against it. Participants rated their confidence in their diagnosis at each of the three steps on a 10 point scale, and results were analyzed using a two-way ANOVA with repeated measures for two between-participant levels (correct or incorrect diagnosis) and three within-participant factors (diagnostic steps). The Bonferroni method was used for multiple comparison tests. Mean confidence scores were 5.01 (Step 1), 5.20 (Step 2), and 4.98 (Step 3); multiple comparisons showed a significant difference between Steps 1–2 (P =.04) and 2–3 (P =.01). Verbalization of evidence in favor of the predicted hypothesis (Step 2) and against it (Step 3) prompted changes in diagnosis in four cases of misdiagnosis (three at Step 2, one at Step 3). The 6-microskills method, which added a part encouraging learners to verbalize why something “does not fit” with a predicted diagnosis, may effectively correct the confirmation bias associated with diagnostic predictions.