Bin Xu
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
Research Subjects
Bin Xu has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Yanping Li
Reproductive Medicine Center, Xiangya Hospital of Central South University, Changsha, People’s Republic of China
11 shared publications - Jing Zhao
Reproductive Medicine Center, Xiangya Hospital of Central South University, Changsha, China
7 shared publications - Hui Li
Department of Reproductive Medicine, Xiangya Hospital, Central South University, Changsha, Hunan, China
6 shared publications - Jingjing Chen
Department of Reproductive Medicine, Xiangya Hospital, Central South University, Changsha, Hunan, China
4 shared publications Qiong Zhang
Center for Reproductive Medicine, Xiangya Hospital of Central South University , Changsha 410000
4 shared publications- Dabao Xu
Department of Obstetrics and Gynecology, The Third Xiangya Hospital of Central South University, PR China
3 shared publications
Latest Publications
- Open Access
- 1 August 2026
- Published by Taylor & Francis in International Journal of Women's Health
- Vol. ume 18, 1-21
Global, Regional, and National Burden of Acute Abdominal Conditions in Females—Appendicitis, Ectopic Pregnancy, Paralytic Ileus and Intestinal Obstruction, and Urolithiasis, 1990–2021: A Systematic Analysis for the Global Burden of Disease Study 2021 with ARIMA Projections to 2030
Background: Acute abdominal conditions--appendicitis, ectopic pregnancy, paralytic ileus and intestinal obstruction, and urolithiasis--are surgical emergencies carrying substantial risks of morbidity, mortality, and long-term disability. An integrated quantification of their collective burden may help identify shared vulnerabilities in global emergency surgical systems. Methods: Data from the Global Burden of Diseases, Injuries, and Risk Factors Study 2021 were used to estimate incidence, prevalence, mortality, years lived with disability, and disability-adjusted life-years (DALYs) for the four conditions from 1990 to 2021 across 204 countries, 21 regions, and five Socio-demographic Index (SDI) quintiles. Temporal trends were quantified using the estimated annual percentage change, and Autoregressive Integrated Moving Average models projected key indicators to 2030. All analyses complied with GATHER guidelines. Results: In 2021, the combined age-standardised incidence rate was stable at 1,200 (95% UI 1,150– 1,250) per 100,000 (EAPC − 0.05%, 95% CI − 0.10 to 0.00), while absolute cases rose by 85%, from 50 million in 1990 to 92 million in 2021. Age-standardised DALY rate declined (EAPC − 1.2%, 95% CI − 1.4 to − 1.0) and global deaths fell by 20% (18– 22%). Highest incidence rates occurred in Eastern Europe and Central Asia; the heaviest DALY burdens were in sub-Saharan Africa and South Asia. A bimodal age pattern was observed, peaking at 15– 49 years (ectopic pregnancy, appendicitis) and ≥ 70 years (paralytic ileus and intestinal obstruction). Projections to 2030 indicate continued DALY declines for appendicitis, ectopic pregnancy, and paralytic ileus, whereas urolithiasis DALYs may oscillate or rise, particularly in middle- and high-SDI settings. Conclusion: The association between population growth and ageing and rising absolute case volumes suggests demographic shifts as a major driver of the escalating burden of acute abdominal conditions. Pronounced geographic and SDI-related disparities were observed alongside distinct age profiles across the four aetiologies, underscoring the need for context-specific planning of future emergency surgical and preventive care.
- 1 August 2026
- Published by Taylor & Francis in International Journal of Women's Health
- Vol. ume 18, 1-21
Open AccessGlobal, Regional, and National Burden of Acute Abdominal Conditions in Females—Appendicitis, Ectopic Pregnancy, Paralytic Ileus and Intestinal Obstruction, and Urolithiasis, 1990–2021: A Systematic Analysis for the Global Burden of Disease Study 2021 with ARIMA Projections to 2030
Background: Acute abdominal conditions--appendicitis, ectopic pregnancy, paralytic ileus and intestinal obstruction, and urolithiasis--are surgical emergencies carrying substantial risks of morbidity, mortality, and long-term disability. An integrated quantification of their collective burden may help identify shared vulnerabilities in global emergency surgical systems. Methods: Data from the Global Burden of Diseases, Injuries, and Risk Factors Study 2021 were used to estimate incidence, prevalence, mortality, years lived with disability, and disability-adjusted life-years (DALYs) for the four conditions from 1990 to 2021 across 204 countries, 21 regions, and five Socio-demographic Index (SDI) quintiles. Temporal trends were quantified using the estimated annual percentage change, and Autoregressive Integrated Moving Average models projected key indicators to 2030. All analyses complied with GATHER guidelines. Results: In 2021, the combined age-standardised incidence rate was stable at 1,200 (95% UI 1,150– 1,250) per 100,000 (EAPC − 0.05%, 95% CI − 0.10 to 0.00), while absolute cases rose by 85%, from 50 million in 1990 to 92 million in 2021. Age-standardised DALY rate declined (EAPC − 1.2%, 95% CI − 1.4 to − 1.0) and global deaths fell by 20% (18– 22%). Highest incidence rates occurred in Eastern Europe and Central Asia; the heaviest DALY burdens were in sub-Saharan Africa and South Asia. A bimodal age pattern was observed, peaking at 15– 49 years (ectopic pregnancy, appendicitis) and ≥ 70 years (paralytic ileus and intestinal obstruction). Projections to 2030 indicate continued DALY declines for appendicitis, ectopic pregnancy, and paralytic ileus, whereas urolithiasis DALYs may oscillate or rise, particularly in middle- and high-SDI settings. Conclusion: The association between population growth and ageing and rising absolute case volumes suggests demographic shifts as a major driver of the escalating burden of acute abdominal conditions. Pronounced geographic and SDI-related disparities were observed alongside distinct age profiles across the four aetiologies, underscoring the need for context-specific planning of future emergency surgical and preventive care.
- Open Access
- 3 February 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e78245
Integrated Prediction System for Individualized Ovarian Stimulation and Ovarian Hyperstimulation Syndrome Prevention: Algorithm Development and Validation
Background: Accurately predicting ovarian response and determining the optimal starting dose of follicle-stimulating hormone (FSH) remain critical yet challenging for effective ovarian stimulation. Currently, there is a lack of a comprehensive model capable of simultaneously forecasting the number of oocytes retrieved (NOR) and assessing the risk of early-onset moderate-to-severe ovarian hyperstimulation syndrome (OHSS). Objective: This study aimed to establish an integrated mode capable of forecasting the NOR and assessing the risk of early-onset moderate-to-severe OHSS across varying starting doses of FSH. Methods: This prognostic study included patients undergoing their first ovarian stimulation cycles at 2 independent in vitro fertilization clinics. Automated classifiers were used for variable selection. Machine learning models (11 for NOR and 11 for OHSS) were developed and validated using internal (n=6401) and external (n=3805) datasets. Shapley additive explanation was applied for variable interpretation. The best-performing models were incorporated into a web-based prediction tool. Results: For NOR prediction, 17 variables were selected, with the gradient boosting regressor achieving the highest performance (internal dataset: R2=0.7978; external dataset: R2=0.7924). For OHSS prediction, 19 variables were identified, and the LightGBM model demonstrated superior performance (internal dataset: area under the receiver operating characteristic curve=0.7588; external dataset: area under the receiver operating characteristic curve=0.7287). Shapley additive explanation analysis highlighted the FSH starting dose to BMI ratio and baseline antral follicle count as key predictors for NOR and OHSS, respectively. Dose-response curves were generated to visualize predicted outcomes with varying FSH starting doses. The models were implemented in a user-friendly, research-oriented online prototype, individualized ovarian stimulation guide (InOvaSGuide). Conclusions: This study introduces an integrated framework for predicting NOR and early-onset moderate-to-severe OHSS risk across different FSH doses. Future prospective evaluation is needed before clinical implementation.
- 3 February 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28, e78245
Open AccessIntegrated Prediction System for Individualized Ovarian Stimulation and Ovarian Hyperstimulation Syndrome Prevention: Algorithm Development and Validation
Background: Accurately predicting ovarian response and determining the optimal starting dose of follicle-stimulating hormone (FSH) remain critical yet challenging for effective ovarian stimulation. Currently, there is a lack of a comprehensive model capable of simultaneously forecasting the number of oocytes retrieved (NOR) and assessing the risk of early-onset moderate-to-severe ovarian hyperstimulation syndrome (OHSS). Objective: This study aimed to establish an integrated mode capable of forecasting the NOR and assessing the risk of early-onset moderate-to-severe OHSS across varying starting doses of FSH. Methods: This prognostic study included patients undergoing their first ovarian stimulation cycles at 2 independent in vitro fertilization clinics. Automated classifiers were used for variable selection. Machine learning models (11 for NOR and 11 for OHSS) were developed and validated using internal (n=6401) and external (n=3805) datasets. Shapley additive explanation was applied for variable interpretation. The best-performing models were incorporated into a web-based prediction tool. Results: For NOR prediction, 17 variables were selected, with the gradient boosting regressor achieving the highest performance (internal dataset: R2=0.7978; external dataset: R2=0.7924). For OHSS prediction, 19 variables were identified, and the LightGBM model demonstrated superior performance (internal dataset: area under the receiver operating characteristic curve=0.7588; external dataset: area under the receiver operating characteristic curve=0.7287). Shapley additive explanation analysis highlighted the FSH starting dose to BMI ratio and baseline antral follicle count as key predictors for NOR and OHSS, respectively. Dose-response curves were generated to visualize predicted outcomes with varying FSH starting doses. The models were implemented in a user-friendly, research-oriented online prototype, individualized ovarian stimulation guide (InOvaSGuide). Conclusions: This study introduces an integrated framework for predicting NOR and early-onset moderate-to-severe OHSS risk across different FSH doses. Future prospective evaluation is needed before clinical implementation.
- Open Access
- 25 November 2025
- Published by Springer Nature in BMC Pregnancy and Childbirth
- Vol. 26 (1)
Melatonin supplementation and outcomes of assisted reproductive technology: a systematic review and meta-analysis
Background: Melatonin (MT) is involved in the regulation of various important biological processes related to reproduction. Many studies have investigated the effects of MT supplementation on outcomes following assisted reproductive technology (ART), yielding conflicting results. The aim of this systematic review was to synthesize evidence from clinical studies regarding the impact of MT on the key outcomes of ART. Methods: PubMed, Embase, Web of Science, and Google scholar were searched. Clinical trials that studied the effect of MT supplementation on outcomes following ART and published in English from inception to April 2020, were included. One author assessed the risk of bias in the studies using the Cochrane Collaboration checklist. Dichotomous outcomes were analyzed as risk ratios (RR) using the Mantel-Haenszel statistical method and a random/fixed effect model. Continuous outcomes were analyzed as Mean Difference (MD) using the Inverse Variance statistical method. The funnel plot was used to assess the publication bias. Results: Eleven studies conducted between 2008 and 2019 were included in this meta-analysis. Clinical pregnancy rate (CPR), live birth rate (LBR), Miscarriage rate (MR), fertilization rate (FR), Number of oocytes retrieved, MII oocytes, top-quality embryos was reported in 10, 3, 6, 7, 9, 8, and 6 studies, respectively. MT supplementation significantly increased the CPR (RR, 1.24; 95% confidence interval [CI], 1.04, 1.47), the number of MII oocytes (MD, 1.39; 95% CI, 0.74, 2.04), the number of top-quality embryos (MD, 0.56; 95% CI, 0.24, 0.88), and the FR (4 studies with RR, 1.10; 95% CI, 1.03, 1.17; 3 studies with MD, 0.13; 95% CI, 0.01, 0.24). However, there was no significant difference in LBR (RR, 1.23; 95% CI, 0.85, 1.80), the number of oocytes retrieved (MD, 0.58; 95% CI, -0.12, 1.27), and the MR (RR, 0.96; 95% CI, 0.50, 1.82). When studies were sub-grouped by the control interventions, either myoinositol(MI) plus folic acid (FA) or placebo/no treatment, MT supplementation increased number of MII oocytes (MT + MI + FA vs. MI + FA, MD, 0.91; 95% CI, 0.40, 1.41; MT vs. Placebo/no treatment, MD, 2.06; 95% CI, 0.73, 3.39) and number of top-quality embryos (MT + MI + FA vs. MI + FA, MD, 0.70; 95% CI, 0.24, 1.16; MT vs. Placebo/no treatment, MD, 0.33; 95% CI, 0.11, 0.54), while the CPR remained comparable between groups(MT + MI + FA vs. MI + FA, RR, 1.22; 95% CI, 0.96, 1.54; MT vs. Placebo/No treatment, RR, 1.26; 95% CI, 0.97, 1.62). When sub-group analysis was performed basing on women’s characteristics, MT supplementation showed no significant beneficial effect on CPR in women with polycystic ovary syndrome (PCOS) (RR, 1.18; 95% CI, 0.92, 1.52), normal ovarian function (RR, 1.15; 95% CI, 0.87, 1.53), or a history of low fertilization rates or poor-quality embryos (RR, 1.71; 95% CI, 0.95, 3.07). However, MT supplementation increased the number of MII oocytes in women with PCOS (MD, 0.97; 95% CI, 0.22, 1.73), but this benefit was not observed in women with normal ovarian function (MD, 1.49; 95% CI, -0.33, 3.31). Conclusions: As the outcomes of ART were influenced by multiple factors, MT supplementation may not significantly improve clinical pregnancy or live birth rate. However, MT appears to have a positive effect on oocyte and embryo quality, particularly in women with PCOS or decreased ovarian reserve (DOR), at least to some extent. Nevertheless, further well-designed, large-scale studies are needed before MT can be recommended for routine use in clinical practice.
- 25 November 2025
- Published by Springer Nature in BMC Pregnancy and Childbirth
- Vol. 26 (1)
Open AccessMelatonin supplementation and outcomes of assisted reproductive technology: a systematic review and meta-analysis
Background: Melatonin (MT) is involved in the regulation of various important biological processes related to reproduction. Many studies have investigated the effects of MT supplementation on outcomes following assisted reproductive technology (ART), yielding conflicting results. The aim of this systematic review was to synthesize evidence from clinical studies regarding the impact of MT on the key outcomes of ART. Methods: PubMed, Embase, Web of Science, and Google scholar were searched. Clinical trials that studied the effect of MT supplementation on outcomes following ART and published in English from inception to April 2020, were included. One author assessed the risk of bias in the studies using the Cochrane Collaboration checklist. Dichotomous outcomes were analyzed as risk ratios (RR) using the Mantel-Haenszel statistical method and a random/fixed effect model. Continuous outcomes were analyzed as Mean Difference (MD) using the Inverse Variance statistical method. The funnel plot was used to assess the publication bias. Results: Eleven studies conducted between 2008 and 2019 were included in this meta-analysis. Clinical pregnancy rate (CPR), live birth rate (LBR), Miscarriage rate (MR), fertilization rate (FR), Number of oocytes retrieved, MII oocytes, top-quality embryos was reported in 10, 3, 6, 7, 9, 8, and 6 studies, respectively. MT supplementation significantly increased the CPR (RR, 1.24; 95% confidence interval [CI], 1.04, 1.47), the number of MII oocytes (MD, 1.39; 95% CI, 0.74, 2.04), the number of top-quality embryos (MD, 0.56; 95% CI, 0.24, 0.88), and the FR (4 studies with RR, 1.10; 95% CI, 1.03, 1.17; 3 studies with MD, 0.13; 95% CI, 0.01, 0.24). However, there was no significant difference in LBR (RR, 1.23; 95% CI, 0.85, 1.80), the number of oocytes retrieved (MD, 0.58; 95% CI, -0.12, 1.27), and the MR (RR, 0.96; 95% CI, 0.50, 1.82). When studies were sub-grouped by the control interventions, either myoinositol(MI) plus folic acid (FA) or placebo/no treatment, MT supplementation increased number of MII oocytes (MT + MI + FA vs. MI + FA, MD, 0.91; 95% CI, 0.40, 1.41; MT vs. Placebo/no treatment, MD, 2.06; 95% CI, 0.73, 3.39) and number of top-quality embryos (MT + MI + FA vs. MI + FA, MD, 0.70; 95% CI, 0.24, 1.16; MT vs. Placebo/no treatment, MD, 0.33; 95% CI, 0.11, 0.54), while the CPR remained comparable between groups(MT + MI + FA vs. MI + FA, RR, 1.22; 95% CI, 0.96, 1.54; MT vs. Placebo/No treatment, RR, 1.26; 95% CI, 0.97, 1.62). When sub-group analysis was performed basing on women’s characteristics, MT supplementation showed no significant beneficial effect on CPR in women with polycystic ovary syndrome (PCOS) (RR, 1.18; 95% CI, 0.92, 1.52), normal ovarian function (RR, 1.15; 95% CI, 0.87, 1.53), or a history of low fertilization rates or poor-quality embryos (RR, 1.71; 95% CI, 0.95, 3.07). However, MT supplementation increased the number of MII oocytes in women with PCOS (MD, 0.97; 95% CI, 0.22, 1.73), but this benefit was not observed in women with normal ovarian function (MD, 1.49; 95% CI, -0.33, 3.31). Conclusions: As the outcomes of ART were influenced by multiple factors, MT supplementation may not significantly improve clinical pregnancy or live birth rate. However, MT appears to have a positive effect on oocyte and embryo quality, particularly in women with PCOS or decreased ovarian reserve (DOR), at least to some extent. Nevertheless, further well-designed, large-scale studies are needed before MT can be recommended for routine use in clinical practice.
- Top Cited PapersOpen Access
- 9 June 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e72062
Large Language Models in Medical Diagnostics: Scoping Review With Bibliometric Analysis
Background: The integration of large language models (LLMs) into medical diagnostics has garnered substantial attention due to their potential to enhance diagnostic accuracy, streamline clinical workflows, and address health care disparities. However, the rapid evolution of LLM research necessitates a comprehensive synthesis of their applications, challenges, and future directions. Objective: This scoping review aimed to provide an overview of the current state of research regarding the use of LLMs in medical diagnostics. The study sought to answer four primary subquestions, as follows: (1) Which LLMs are commonly used? (2) How are LLMs assessed in diagnosis? (3) What is the current performance of LLMs in diagnosing diseases? (4) Which medical domains are investigating the application of LLMs? Methods: This scoping review was conducted according to the Joanna Briggs Institute Manual for Evidence Synthesis and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Relevant literature was searched from the Web of Science, PubMed, Embase, IEEE Xplore, and ACM Digital Library databases from 2022 to 2025. Articles were screened and selected based on predefined inclusion and exclusion criteria. Bibliometric analysis was performed using VOSviewer to identify major research clusters and trends. Data extraction included details on LLM types, application domains, and performance metrics. Results: The field is rapidly expanding, with a surge in publications after 2023. GPT-4 and its variants dominated research (70/95, 74% of studies), followed by GPT-3.5 (34/95, 36%). Key applications included disease classification (text or image-based), medical question answering, and diagnostic content generation. LLMs demonstrated high accuracy in specialties like radiology, psychiatry, and neurology but exhibited biases in race, gender, and cost predictions. Ethical concerns, including privacy risks and model hallucination, alongside regulatory fragmentation, were critical barriers to clinical adoption. Conclusions: LLMs hold transformative potential for medical diagnostics but require rigorous validation, bias mitigation, and multimodal integration to address real-world complexities. Future research should prioritize explainable artificial intelligence frameworks, specialty-specific optimization, and international regulatory harmonization to ensure equitable and safe clinical deployment.
- 9 June 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e72062
Top Cited PapersOpen AccessLarge Language Models in Medical Diagnostics: Scoping Review With Bibliometric Analysis
Background: The integration of large language models (LLMs) into medical diagnostics has garnered substantial attention due to their potential to enhance diagnostic accuracy, streamline clinical workflows, and address health care disparities. However, the rapid evolution of LLM research necessitates a comprehensive synthesis of their applications, challenges, and future directions. Objective: This scoping review aimed to provide an overview of the current state of research regarding the use of LLMs in medical diagnostics. The study sought to answer four primary subquestions, as follows: (1) Which LLMs are commonly used? (2) How are LLMs assessed in diagnosis? (3) What is the current performance of LLMs in diagnosing diseases? (4) Which medical domains are investigating the application of LLMs? Methods: This scoping review was conducted according to the Joanna Briggs Institute Manual for Evidence Synthesis and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Relevant literature was searched from the Web of Science, PubMed, Embase, IEEE Xplore, and ACM Digital Library databases from 2022 to 2025. Articles were screened and selected based on predefined inclusion and exclusion criteria. Bibliometric analysis was performed using VOSviewer to identify major research clusters and trends. Data extraction included details on LLM types, application domains, and performance metrics. Results: The field is rapidly expanding, with a surge in publications after 2023. GPT-4 and its variants dominated research (70/95, 74% of studies), followed by GPT-3.5 (34/95, 36%). Key applications included disease classification (text or image-based), medical question answering, and diagnostic content generation. LLMs demonstrated high accuracy in specialties like radiology, psychiatry, and neurology but exhibited biases in race, gender, and cost predictions. Ethical concerns, including privacy risks and model hallucination, alongside regulatory fragmentation, were critical barriers to clinical adoption. Conclusions: LLMs hold transformative potential for medical diagnostics but require rigorous validation, bias mitigation, and multimodal integration to address real-world complexities. Future research should prioritize explainable artificial intelligence frameworks, specialty-specific optimization, and international regulatory harmonization to ensure equitable and safe clinical deployment.
- Open Access
- 30 May 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Integrated Prediction System for Individualized Ovarian Stimulation and Ovarian Hyperstimulation Syndrome Prevention: Algorithm Development and Validation (Preprint)
BACKGROUND Accurately predicting ovarian response and determining the optimal starting dose of follicle-stimulating hormone (FSH) remain critical yet challenging for effective ovarian stimulation. Currently, there is a lack of a comprehensive model capable of simultaneously forecasting the number of oocytes retrieved (NOR) and assessing the risk of early-onset moderate-to-severe ovarian hyperstimulation syndrome (OHSS). OBJECTIVE This study aimed to establish an integrated mode capable of forecasting the NOR and assessing the risk of early-onset moderate-to-severe OHSS across varying starting doses of FSH. METHODS This prognostic study included patients undergoing their first ovarian stimulation cycles at 2 independent in vitro fertilization clinics. Automated classifiers were used for variable selection. Machine learning models (11 for NOR and 11 for OHSS) were developed and validated using internal (n=6401) and external (n=3805) datasets. Shapley additive explanation was applied for variable interpretation. The best-performing models were incorporated into a web-based prediction tool. RESULTS For NOR prediction, 17 variables were selected, with the gradient boosting regressor achieving the highest performance (internal dataset: R2=0.7978; external dataset: R2=0.7924). For OHSS prediction, 19 variables were identified, and the LightGBM model demonstrated superior performance (internal dataset: area under the receiver operating characteristic curve=0.7588; external dataset: area under the receiver operating characteristic curve=0.7287). Shapley additive explanation analysis highlighted the FSH starting dose to BMI ratio and baseline antral follicle count as key predictors for NOR and OHSS, respectively. Dose-response curves were generated to visualize predicted outcomes with varying FSH starting doses. The models were implemented in a user-friendly, research-oriented online prototype, individualized ovarian stimulation guide (InOvaSGuide). CONCLUSIONS This study introduces an integrated framework for predicting NOR and early-onset moderate-to-severe OHSS risk across different FSH doses. Future prospective evaluation is needed before clinical implementation.
- 30 May 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessIntegrated Prediction System for Individualized Ovarian Stimulation and Ovarian Hyperstimulation Syndrome Prevention: Algorithm Development and Validation (Preprint)
BACKGROUND Accurately predicting ovarian response and determining the optimal starting dose of follicle-stimulating hormone (FSH) remain critical yet challenging for effective ovarian stimulation. Currently, there is a lack of a comprehensive model capable of simultaneously forecasting the number of oocytes retrieved (NOR) and assessing the risk of early-onset moderate-to-severe ovarian hyperstimulation syndrome (OHSS). OBJECTIVE This study aimed to establish an integrated mode capable of forecasting the NOR and assessing the risk of early-onset moderate-to-severe OHSS across varying starting doses of FSH. METHODS This prognostic study included patients undergoing their first ovarian stimulation cycles at 2 independent in vitro fertilization clinics. Automated classifiers were used for variable selection. Machine learning models (11 for NOR and 11 for OHSS) were developed and validated using internal (n=6401) and external (n=3805) datasets. Shapley additive explanation was applied for variable interpretation. The best-performing models were incorporated into a web-based prediction tool. RESULTS For NOR prediction, 17 variables were selected, with the gradient boosting regressor achieving the highest performance (internal dataset: R2=0.7978; external dataset: R2=0.7924). For OHSS prediction, 19 variables were identified, and the LightGBM model demonstrated superior performance (internal dataset: area under the receiver operating characteristic curve=0.7588; external dataset: area under the receiver operating characteristic curve=0.7287). Shapley additive explanation analysis highlighted the FSH starting dose to BMI ratio and baseline antral follicle count as key predictors for NOR and OHSS, respectively. Dose-response curves were generated to visualize predicted outcomes with varying FSH starting doses. The models were implemented in a user-friendly, research-oriented online prototype, individualized ovarian stimulation guide (InOvaSGuide). CONCLUSIONS This study introduces an integrated framework for predicting NOR and early-onset moderate-to-severe OHSS risk across different FSH doses. Future prospective evaluation is needed before clinical implementation.
- Open Access
- 22 April 2025
- Published by Oxford University Press (OUP) in Stem Cells Translational Medicine
- Vol. 14 (4)
hUC-MSCs loaded collagen scaffold for refractory thin endometrium caused by Asherman syndrome: a double-blind randomized controlled trial
In this single-center, double-blinded, randomized controlled trial, we investigated whether human umbilical cord-derived mesenchymal stromal cells loaded collagen scaffolds (hUC-MSC/CS) could improve the cumulative live-birth rate (cLBR) in infertile women with refractory thin endometrium (RTE). We randomly assigned 25 subfertile women with RTE, in a 1:1 ratio, to receive hysteroscopic adhesiolysis and plowing plus either hUC-MSC/CS or saline/CS (control) for intrauterine implantation. Uterine fluid was collected on the embryo transfer day for RNA-sequencing to explore the potential mechanisms by which hUC-MSCs exert their effects. The primary outcome was the cLBR. Live births occurred in 3 out of 11 women in the hUC-MSC/CS group and in 1 out of 13 women in the control group (27.3% vs 7.7%; relative risk [RR], 3.55; 95% confidence interval [CI], 0.43 to 29.42; P = .30). The cumulative frequencies of clinical pregnancy were 5/11 and 1/13 in the hUC-MSC/CS group and control group, respectively (45.5% vs. 7.7%; RR, 5.91; 95% CI, 0.81-43.28; P = .06). Two of 11 participants developed urticaria in the hUC-MSC/CS group. Enrichment analysis showed that T-cell activation had the largest proportion in the biological process category. Kyoto Encyclopedia of Genes and Genomes pathway analysis showed that most genes were related to cytokine-cytokine receptor interaction. In conclusion, there was a non-significant trend toward a higher cLBR with hUC-MSC/CS compared to controls, potentially through the cytokine-cytokine receptor interaction pathway. hUC-MSCs appeared to be relatively safe in a 1-year follow-up. Therefore, this novel therapy can be proposed to patients with RTE.
- 22 April 2025
- Published by Oxford University Press (OUP) in Stem Cells Translational Medicine
- Vol. 14 (4)
Open AccesshUC-MSCs loaded collagen scaffold for refractory thin endometrium caused by Asherman syndrome: a double-blind randomized controlled trial
In this single-center, double-blinded, randomized controlled trial, we investigated whether human umbilical cord-derived mesenchymal stromal cells loaded collagen scaffolds (hUC-MSC/CS) could improve the cumulative live-birth rate (cLBR) in infertile women with refractory thin endometrium (RTE). We randomly assigned 25 subfertile women with RTE, in a 1:1 ratio, to receive hysteroscopic adhesiolysis and plowing plus either hUC-MSC/CS or saline/CS (control) for intrauterine implantation. Uterine fluid was collected on the embryo transfer day for RNA-sequencing to explore the potential mechanisms by which hUC-MSCs exert their effects. The primary outcome was the cLBR. Live births occurred in 3 out of 11 women in the hUC-MSC/CS group and in 1 out of 13 women in the control group (27.3% vs 7.7%; relative risk [RR], 3.55; 95% confidence interval [CI], 0.43 to 29.42; P = .30). The cumulative frequencies of clinical pregnancy were 5/11 and 1/13 in the hUC-MSC/CS group and control group, respectively (45.5% vs. 7.7%; RR, 5.91; 95% CI, 0.81-43.28; P = .06). Two of 11 participants developed urticaria in the hUC-MSC/CS group. Enrichment analysis showed that T-cell activation had the largest proportion in the biological process category. Kyoto Encyclopedia of Genes and Genomes pathway analysis showed that most genes were related to cytokine-cytokine receptor interaction. In conclusion, there was a non-significant trend toward a higher cLBR with hUC-MSC/CS compared to controls, potentially through the cytokine-cytokine receptor interaction pathway. hUC-MSCs appeared to be relatively safe in a 1-year follow-up. Therefore, this novel therapy can be proposed to patients with RTE.
- Open Access
- 2 February 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Large Language Models in Medical Diagnostics: Scoping Review With Bibliometric Analysis (Preprint)
BACKGROUND The integration of large language models (LLMs) into medical diagnostics has garnered substantial attention due to their potential to enhance diagnostic accuracy, streamline clinical workflows, and address health care disparities. However, the rapid evolution of LLM research necessitates a comprehensive synthesis of their applications, challenges, and future directions. OBJECTIVE This scoping review aimed to provide an overview of the current state of research regarding the use of LLMs in medical diagnostics. The study sought to answer four primary subquestions, as follows: (1) Which LLMs are commonly used? (2) How are LLMs assessed in diagnosis? (3) What is the current performance of LLMs in diagnosing diseases? (4) Which medical domains are investigating the application of LLMs? METHODS This scoping review was conducted according to the Joanna Briggs Institute Manual for Evidence Synthesis and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Relevant literature was searched from the Web of Science, PubMed, Embase, IEEE Xplore, and ACM Digital Library databases from 2022 to 2025. Articles were screened and selected based on predefined inclusion and exclusion criteria. Bibliometric analysis was performed using VOSviewer to identify major research clusters and trends. Data extraction included details on LLM types, application domains, and performance metrics. RESULTS The field is rapidly expanding, with a surge in publications after 2023. GPT-4 and its variants dominated research (70/95, 74% of studies), followed by GPT-3.5 (34/95, 36%). Key applications included disease classification (text or image-based), medical question answering, and diagnostic content generation. LLMs demonstrated high accuracy in specialties like radiology, psychiatry, and neurology but exhibited biases in race, gender, and cost predictions. Ethical concerns, including privacy risks and model hallucination, alongside regulatory fragmentation, were critical barriers to clinical adoption. CONCLUSIONS LLMs hold transformative potential for medical diagnostics but require rigorous validation, bias mitigation, and multimodal integration to address real-world complexities. Future research should prioritize explainable artificial intelligence frameworks, specialty-specific optimization, and international regulatory harmonization to ensure equitable and safe clinical deployment.
- 2 February 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessLarge Language Models in Medical Diagnostics: Scoping Review With Bibliometric Analysis (Preprint)
BACKGROUND The integration of large language models (LLMs) into medical diagnostics has garnered substantial attention due to their potential to enhance diagnostic accuracy, streamline clinical workflows, and address health care disparities. However, the rapid evolution of LLM research necessitates a comprehensive synthesis of their applications, challenges, and future directions. OBJECTIVE This scoping review aimed to provide an overview of the current state of research regarding the use of LLMs in medical diagnostics. The study sought to answer four primary subquestions, as follows: (1) Which LLMs are commonly used? (2) How are LLMs assessed in diagnosis? (3) What is the current performance of LLMs in diagnosing diseases? (4) Which medical domains are investigating the application of LLMs? METHODS This scoping review was conducted according to the Joanna Briggs Institute Manual for Evidence Synthesis and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Relevant literature was searched from the Web of Science, PubMed, Embase, IEEE Xplore, and ACM Digital Library databases from 2022 to 2025. Articles were screened and selected based on predefined inclusion and exclusion criteria. Bibliometric analysis was performed using VOSviewer to identify major research clusters and trends. Data extraction included details on LLM types, application domains, and performance metrics. RESULTS The field is rapidly expanding, with a surge in publications after 2023. GPT-4 and its variants dominated research (70/95, 74% of studies), followed by GPT-3.5 (34/95, 36%). Key applications included disease classification (text or image-based), medical question answering, and diagnostic content generation. LLMs demonstrated high accuracy in specialties like radiology, psychiatry, and neurology but exhibited biases in race, gender, and cost predictions. Ethical concerns, including privacy risks and model hallucination, alongside regulatory fragmentation, were critical barriers to clinical adoption. CONCLUSIONS LLMs hold transformative potential for medical diagnostics but require rigorous validation, bias mitigation, and multimodal integration to address real-world complexities. Future research should prioritize explainable artificial intelligence frameworks, specialty-specific optimization, and international regulatory harmonization to ensure equitable and safe clinical deployment.
- 6 January 2024
- Published by Elsevier in American Journal of Obstetrics and Gynecology
- Vol. 231 (1) , 36-50.35
Effects and safety of hyaluronic acid gel on intrauterine adhesion and fertility after intrauterine surgery: a systematic review and meta-analysis with trial sequential analysis of randomized controlled trials
- 6 January 2024
- Published by Elsevier in American Journal of Obstetrics and Gynecology
- Vol. 231 (1) , 36-50.35
Effects and safety of hyaluronic acid gel on intrauterine adhesion and fertility after intrauterine surgery: a systematic review and meta-analysis with trial sequential analysis of randomized controlled trials
- 15 September 2023
- Published by Oxford University Press (OUP) in Human Reproduction
- Vol. 38 (11) , 2247-2258
Association between ovarian reserve and spontaneous miscarriage and their shared genetic architecture
STUDY QUESTION: Can potential mechanisms involved in the likely concurrence of diminished ovarian reserve (DOR) and miscarriage be identified using genetic data? SUMMARY ANSWER: Concurrence between ovarian reserve and spontaneous miscarriage was observed, and may be attributed to shared genetic risk loci enriched in antigen processing and presentation and autoimmune disease pathways. WHAT IS KNOWN ALREADY: Previous studies have shown that lower serum anti-Müllerian hormone (AMH) levels are associated with increased risk of embryo aneuploidy and spontaneous miscarriage, although findings have not been consistent across all studies. A recent meta-analysis suggested that the association between DOR and miscarriage may not be causal, but rather a result of shared underlying causes such as clinical conditions or past exposure. Motivated by this hypothesis, we conducted the present analysis to explore the concurrence between DOR and miscarriage, and to investigate potential mechanisms using genetic data. STUDY DESIGN, SIZE, DURATION: Three data sources were used in the study: the clinical IVF data were retrospectively collected from an academically affiliated Reproductive Medicine Center (17 786 cycles included); the epidemiological data from the UK Biobank (UKB), which is a large-scale, population-based, prospective cohort study (35 316 white women included), were analyzed; and individual-level genotype data from the UKB were extracted for further analysis. PARTICIPANTS/MATERIALS, SETTING, METHODS: There were three modules of analysis. First, clinical IVF data were used to test the association between ovarian reserve biomarkers and the subsequent early spontaneous miscarriage risk. Second, the UKB data were used to test the association of spontaneous miscarriage history and early menopause. Third, individual-level genotype data from the UKB were analyzed to identify specific pleiotropic genes which affect the development of miscarriage and menopause. MAIN RESULTS AND THE ROLE OF CHANCE: In the analysis of clinical IVF data, the risk of early spontaneous miscarriage was 1.57 times higher in the group with AMH < 1.1 ng/ml group (P < 0.001), 1.62 times for antral follicular count <5 (P < 0.001), and 1.39 times for FSH ≥10 mIU/ml (P < 0.001) in comparison with normal ovarian reserve groups. In the analysis of UKB data, participants with a history of three or more miscarriages had a one-third higher risk of experiencing early menopause (odds ratio: 1.30, 95% CI 1.13–1.49, P < 0.001), compared with participants without spontaneous miscarriage history. We identified 158 shared genetic risk loci that affect both miscarriage and menopause, which enrichment analysis showed were involved in antigen processing and presentation and autoimmune disease pathways. LIMITATIONS, REASONS FOR CAUTION: The analyses of the UKB data were restricted to participants of European ancestry, as 94.6% of the cohort were of white ethnicity. Further studies are needed in non-white populations. Additionally, maternal age at the time of spontaneous miscarriage was not available in the UKB cohort, therefore we adjusted for age at baseline assessment in the models instead. It is known that miscarriage rate in IVF is higher compared to natural conception, highlighting a need for caution when generalizing our findings from the IVF cohort to the general population. WIDER IMPLICATIONS OF THE FINDINGS: Our findings have implications for IVF clinicians in terms of patient counseling on the prognosis of IVF treatment, as well as for genetic counseling regarding miscarriage. Our results highlight the importance of further research on the shared genetic architecture and common pathophysiological basis of DOR and miscarriage, which may lead to new therapeutic opportunities. STUDY FUNDING/COMPETING INTEREST(S): This work was supported by the Hunan Youth Science and Technology Innovation Talent Project (2020RC3060), the International Postdoctoral Exchange Fellowship Program (Talent-Introduction Program, YJ20220220), the fellowship of China Postdoctoral Science Foundation (2022M723564), and the Natural Science Foundation of Hunan Province, China (2023JJ41016). This work has been accepted for poster presentation at the 39th Annual Meeting of ESHRE, Copenhagen, Denmark, 25–28 June 2023 (Poster number: P-477). The authors declare no conflict of interest. TRIAL REGISTRATION NUMBER: N/A.
- 15 September 2023
- Published by Oxford University Press (OUP) in Human Reproduction
- Vol. 38 (11) , 2247-2258
Association between ovarian reserve and spontaneous miscarriage and their shared genetic architecture
STUDY QUESTION: Can potential mechanisms involved in the likely concurrence of diminished ovarian reserve (DOR) and miscarriage be identified using genetic data? SUMMARY ANSWER: Concurrence between ovarian reserve and spontaneous miscarriage was observed, and may be attributed to shared genetic risk loci enriched in antigen processing and presentation and autoimmune disease pathways. WHAT IS KNOWN ALREADY: Previous studies have shown that lower serum anti-Müllerian hormone (AMH) levels are associated with increased risk of embryo aneuploidy and spontaneous miscarriage, although findings have not been consistent across all studies. A recent meta-analysis suggested that the association between DOR and miscarriage may not be causal, but rather a result of shared underlying causes such as clinical conditions or past exposure. Motivated by this hypothesis, we conducted the present analysis to explore the concurrence between DOR and miscarriage, and to investigate potential mechanisms using genetic data. STUDY DESIGN, SIZE, DURATION: Three data sources were used in the study: the clinical IVF data were retrospectively collected from an academically affiliated Reproductive Medicine Center (17 786 cycles included); the epidemiological data from the UK Biobank (UKB), which is a large-scale, population-based, prospective cohort study (35 316 white women included), were analyzed; and individual-level genotype data from the UKB were extracted for further analysis. PARTICIPANTS/MATERIALS, SETTING, METHODS: There were three modules of analysis. First, clinical IVF data were used to test the association between ovarian reserve biomarkers and the subsequent early spontaneous miscarriage risk. Second, the UKB data were used to test the association of spontaneous miscarriage history and early menopause. Third, individual-level genotype data from the UKB were analyzed to identify specific pleiotropic genes which affect the development of miscarriage and menopause. MAIN RESULTS AND THE ROLE OF CHANCE: In the analysis of clinical IVF data, the risk of early spontaneous miscarriage was 1.57 times higher in the group with AMH < 1.1 ng/ml group (P < 0.001), 1.62 times for antral follicular count <5 (P < 0.001), and 1.39 times for FSH ≥10 mIU/ml (P < 0.001) in comparison with normal ovarian reserve groups. In the analysis of UKB data, participants with a history of three or more miscarriages had a one-third higher risk of experiencing early menopause (odds ratio: 1.30, 95% CI 1.13–1.49, P < 0.001), compared with participants without spontaneous miscarriage history. We identified 158 shared genetic risk loci that affect both miscarriage and menopause, which enrichment analysis showed were involved in antigen processing and presentation and autoimmune disease pathways. LIMITATIONS, REASONS FOR CAUTION: The analyses of the UKB data were restricted to participants of European ancestry, as 94.6% of the cohort were of white ethnicity. Further studies are needed in non-white populations. Additionally, maternal age at the time of spontaneous miscarriage was not available in the UKB cohort, therefore we adjusted for age at baseline assessment in the models instead. It is known that miscarriage rate in IVF is higher compared to natural conception, highlighting a need for caution when generalizing our findings from the IVF cohort to the general population. WIDER IMPLICATIONS OF THE FINDINGS: Our findings have implications for IVF clinicians in terms of patient counseling on the prognosis of IVF treatment, as well as for genetic counseling regarding miscarriage. Our results highlight the importance of further research on the shared genetic architecture and common pathophysiological basis of DOR and miscarriage, which may lead to new therapeutic opportunities. STUDY FUNDING/COMPETING INTEREST(S): This work was supported by the Hunan Youth Science and Technology Innovation Talent Project (2020RC3060), the International Postdoctoral Exchange Fellowship Program (Talent-Introduction Program, YJ20220220), the fellowship of China Postdoctoral Science Foundation (2022M723564), and the Natural Science Foundation of Hunan Province, China (2023JJ41016). This work has been accepted for poster presentation at the 39th Annual Meeting of ESHRE, Copenhagen, Denmark, 25–28 June 2023 (Poster number: P-477). The authors declare no conflict of interest. TRIAL REGISTRATION NUMBER: N/A.
- Open Access
- 9 November 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-6
Chromosomal polymorphisms have no negative effect on reproductive outcomes after IVF/ICSI-ET/FET
The present study aimed to explore whether chromosomal polymorphisms (CPs) have negative effects on reproductive outcomes of in vitro fertilization/intracytoplasmic sperm injection-embryo transfer (IVF/ICSI-ET)/frozen-thawing embryo transfer (FET)? We conducted a retrospective study consisting of 21,867 assisted reproductive technology treatment cycles, among which, fresh embryo transfer cycles accounted for 10,400, and the rest were FET cycles. According to karyotype of CPs, the former was grouped as: group 1 (male carrier, n = 425), group 2 (female carrier, n = 262), and group 3 (couple without CPs, n = 9713). Accordingly, FET cycles were divided into 3 groups: group 4 (male carrier, n = 298), group 5 (female carrier, n = 311), and group 6 (couple without CPs, n = 10,858). The embryo implantation rate (IR), clinical pregnancy rate (CPR), live birth rate (LBR), and early miscarriage rate (EMR) were compared among the groups. In fresh embryo transfer cycles after IVF/ICSI, there were no significant differences in the infertility duration, BMI, basal FSH, no. of oocyte, no. of 2PN, endometrial thickness on trigger day, serum E2, P, and LH level on trigger day (P > 0.05). The female age, no. of 2PN embryo cleavage, top-quality embryo, and no. of embryo transferred were significantly different among groups (P < 0.05). The IR was 38.8%, 36.2%, and 34.0% in groups 1, 2, and 3, respectively. The CPR was 55.1%, 52.3%, and 49.7%, respectively. The LBR was 36.9%, 37.4%, and 36.4%, respectively. The CPR and LBR showed no significant differences among groups. The IR was lower and the EMR was higher in group 3 than those of groups 1 and 2. Binary logistic regression analysis indicated that female age, no. of embryo transferred, EMT, LH, and P on the trigger day were independently factors associated with CPR. Besides, no. of embryo transferred, and EMT on trigger day were associated with LBR, while the CPs was not related with CPR and LBR after IVF/ICSI-ET. In FET cycles, the infertility duration was similar (P > 0.05), but the female age, BMI, no. of embryo transferred were significantly different among groups (P > 0.05). The IR was 24.3%, 23.6% and 22.3% in group 4, 5, and 6, receptivity. The CPR was 31.8%, 30.9%, and 30.0%, the LBR was 23.8%,26.3%, and 23.8%, while the EMR was 12.6%, 13.1%, 14.4%, respectively. The IR, CPR, EMR, and LBR showed no significant differences among groups (P > 0.05). Binary logistic regression analysis indicated that female age, infertility duration, and no. of embryo transferred were independently factors affecting CPR and LBR after FET. The CPs were not associated with CPR and LBR after FET. The results suggested that uniparental carrying of CPs have no effects on the reproductive outcomes after IVF/ICSI-ET/FET. However, it is not clear whether both parents carrying CPs would affect pregnancy outcome.
- 9 November 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-6
Open AccessChromosomal polymorphisms have no negative effect on reproductive outcomes after IVF/ICSI-ET/FET
The present study aimed to explore whether chromosomal polymorphisms (CPs) have negative effects on reproductive outcomes of in vitro fertilization/intracytoplasmic sperm injection-embryo transfer (IVF/ICSI-ET)/frozen-thawing embryo transfer (FET)? We conducted a retrospective study consisting of 21,867 assisted reproductive technology treatment cycles, among which, fresh embryo transfer cycles accounted for 10,400, and the rest were FET cycles. According to karyotype of CPs, the former was grouped as: group 1 (male carrier, n = 425), group 2 (female carrier, n = 262), and group 3 (couple without CPs, n = 9713). Accordingly, FET cycles were divided into 3 groups: group 4 (male carrier, n = 298), group 5 (female carrier, n = 311), and group 6 (couple without CPs, n = 10,858). The embryo implantation rate (IR), clinical pregnancy rate (CPR), live birth rate (LBR), and early miscarriage rate (EMR) were compared among the groups. In fresh embryo transfer cycles after IVF/ICSI, there were no significant differences in the infertility duration, BMI, basal FSH, no. of oocyte, no. of 2PN, endometrial thickness on trigger day, serum E2, P, and LH level on trigger day (P > 0.05). The female age, no. of 2PN embryo cleavage, top-quality embryo, and no. of embryo transferred were significantly different among groups (P < 0.05). The IR was 38.8%, 36.2%, and 34.0% in groups 1, 2, and 3, respectively. The CPR was 55.1%, 52.3%, and 49.7%, respectively. The LBR was 36.9%, 37.4%, and 36.4%, respectively. The CPR and LBR showed no significant differences among groups. The IR was lower and the EMR was higher in group 3 than those of groups 1 and 2. Binary logistic regression analysis indicated that female age, no. of embryo transferred, EMT, LH, and P on the trigger day were independently factors associated with CPR. Besides, no. of embryo transferred, and EMT on trigger day were associated with LBR, while the CPs was not related with CPR and LBR after IVF/ICSI-ET. In FET cycles, the infertility duration was similar (P > 0.05), but the female age, BMI, no. of embryo transferred were significantly different among groups (P > 0.05). The IR was 24.3%, 23.6% and 22.3% in group 4, 5, and 6, receptivity. The CPR was 31.8%, 30.9%, and 30.0%, the LBR was 23.8%,26.3%, and 23.8%, while the EMR was 12.6%, 13.1%, 14.4%, respectively. The IR, CPR, EMR, and LBR showed no significant differences among groups (P > 0.05). Binary logistic regression analysis indicated that female age, infertility duration, and no. of embryo transferred were independently factors affecting CPR and LBR after FET. The CPs were not associated with CPR and LBR after FET. The results suggested that uniparental carrying of CPs have no effects on the reproductive outcomes after IVF/ICSI-ET/FET. However, it is not clear whether both parents carrying CPs would affect pregnancy outcome.