OC11.03: Prediction of ovarian reserve using ultrasound imaging‐based radiomics and deep learning: comparison with antral follicle count and anti‐Müllerian hormone

J. Zhang,S. Liu,S. Ran

DOI: https://doi.org/10.1002/uog.27796

2024-09-11

Ultrasound in Obstetrics and Gynecology

Abstract:Ultrasound in Obstetrics &Gynecology, Volume 64, Issue S1, Page 29-29, September 2024.

radiology, nuclear medicine & medical imaging,obstetrics & gynecology,acoustics

What problem does this paper attempt to address?

A Clinician-Friendly Machine Learning System to Predict Ovarian Response and Deploy Individualized Ovarian Stimulation Strategies in IVF

Guiquan Wang,Liming Zhou,Fengcheng Li,Xiaoling Hu,Yun-Xian Yu,Haochao Ying,Ian Chew,Yimin Zhu

DOI: https://doi.org/10.2139/ssrn.4156167

2022-01-01

SSRN Electronic Journal

Abstract:Reproductive health significantly influences both the overall health of individuals (e.g., cardiometabolic, mental and offspring congenital disorders) and human society. Ovarian stimulation (OS), the foundation of successful in vitro fertilization (IVF) treatments, has been complicated by the uncontrollability of ovarian response since OS was invented. This is mainly due to the unpredictable individual variability, long-term and complex therapies, a vast number of choices and limited evidence-based approaches for subgroups of responders. We developed a clinician-friendly machine learning (ML) based decision support system which can assist clinicians in: (1) diagnosing abnormal ovarian response earlier and faster, (2) understanding the pathogenic profiles of risk factors both in a population and at individual level, and (3) deploying individualized OS strategies that potentially strike a balance between the therapeutic effect and economic costs. It showed significant superiority to current ovarian reserve tests and their combinations that are widely used in clinical practice, and achieved excellent performance both in the internal and external validation. This ML system is also explainable, inspirational and still exhibits robustness in a number of stringent situations.Funding Information: This research was supported by the National Key Research and Development Program of China (Grant No. 2018YFC1003201), the Natural Science Foundation of China (Grant nos. 82071604, 81803245), the Fundamental Research Funds for Central Universities (Grant No. 2021FZZX005-23), and the Natural Science Foundation of Zhejiang Province (Grant No. LQ15H04003).Conflict of Interests: The authors declare no competing interests.Ethical Approval: This study was approved by all local institutional review boards (the main center is the Zhejiang University School of Medicine Women’s Hospital, Approval Number: IRB-20200269-R. This study was also registered and approved in the Ningbo Women & Children Hospital, following the main center), waiving the need for informed consent due to the deidentified data.
Anti-Müllerian hormone for prediction of ovarian response in Chinese infertile women undergoing IVF/ICSI cycles: a prospective, multi-centre, observational study

Rong Li,Fei Gong,Yimin Zhu,Wenhui Fang,Jing Yang,Jiayin Liu,Linli Hu,Dongzi Yang,Xiaoyan Liang,Jie Qiao

DOI: https://doi.org/10.1016/j.rbmo.2016.07.003

IF: 4.567

2016-01-01

Reproductive BioMedicine Online

Abstract:This study explored the correlation between serum anti-Müllerian hormone (AMH) concentration and the number of retrieved oocytes after ovarian stimulation in Chinese infertile women undergoing assisted reproductive technology treatment and AMH cut-off values predicting low and high ovarian response. This was a prospective, multi-centre, observational study. A total of 615 subjects were included in nine assisted reproductive centres in China for outcome analysis. Subjects received assisted reproductive technology treatment and used recombinant human FSH (r-HFSH) or r-HFSH plus recombinant LH (rLH) for ovarian stimulation according to conventional treatment regimens. The main outcome variables were correlations between AMH and the number of retrieved oocytes and the cut-off values of AMH predicting low and high ovarian response. Serum AMH concentration was positively correlated with the number of oocytes retrieved in Chinese infertile women treated with IVF/intracytoplasmic sperm injection (ICSI) (Pearson correlation coefficient = 0.4754, P < 0.0001). The optimal AMH cut-off value was 2.6 ng/ml (sensitivity: 81.28%, specificity: 59.51%) in predicting high and normal response, and 1.1 ng/ml (sensitivity: 52.27%, specificity: 87.23%) in predicting low and normal response. In conclusion, serum AMH concentration can be used as a biomarker to predict ovarian response in Chinese infertile women treated with assisted reproductive technology.
Evaluation of oocyte maturity using artificial intelligence quantification of follicle volume biomarker by three-dimensional ultrasound

Xiaowen Liang,Jiamin Liang,Fengyi Zeng,Yan Lin,Yuewei Li,Kuan Cai,Dong Ni,Zhiyi Chen

DOI: https://doi.org/10.1016/j.rbmo.2022.07.012

Abstract:Research question: Can a novel deep learning-based follicle volume biomarker using three-dimensional ultrasound (3D-US) be established to aid in the assessment of oocyte maturity, timing of HCG administration and the individual prediction of ovarian hyper-response? Design: A total of 515 IVF cases were enrolled, and 3D-US scanning was carried out on HCG administration day. A follicle volume biomarker established by means of a deep learning-based segmentation algorithm was used to calculate optimal leading follicle volume for predicting number of mature oocytes retrieved and optimizing HCG trigger timing. Performance of the novel biomarker cut-off value was compared with conventional two-dimensional ultrasound (2D-US) follicular diameter measurements in assessing oocyte retrieval outcome. Moreover, demographics, infertility work-up and ultrasound biomarkers were used to build models for predicting ovarian hyper-response. Results: On the basis of the deep learning method, the optimal cut-off value of the follicle volume biomarker was determined to be 0.5 cm3 for predicting number of mature oocytes retrieved; its performance was significantly better than the conventional method (two-dimensional diameter measurement ≥10 mm). The cut-off value for leading follicle volume to optimize HCG trigger timing was determined to be 3.0 cm3 and was significantly associated with a higher number of mature oocytes retrieved (P = 0.01). Accuracy of the multi-layer perceptron model was better than two-dimensional diameter measurement (0.890 versus 0.785) and other multivariate classifiers in predicting ovarian hyper-response (P < 0.001). Conclusions: Deep learning segmentation methods and multivariate classifiers based on 3D-US were found to be potentially effective approaches for assessing mature oocyte retrieval outcome and individual prediction of ovarian hyper-response.
Assessment and quantification of ovarian reserve on the basis of machine learning models

Ting Ding,Wu Ren,Tian Wang,Yun Han,Wenqing Ma,Man Wang,Fangfang Fu,Yan Li,Shixuan Wang

DOI: https://doi.org/10.3389/fendo.2023.1087429

IF: 6.055

2023-03-16

Frontiers in Endocrinology

Abstract:Background: Early detection of ovarian aging is of huge importance, although no ideal marker or acknowledged evaluation system exists. The purpose of this study was to develop a better prediction model to assess and quantify ovarian reserve using machine learning methods. Methods: This is a multicenter, nationwide population-based study including a total of 1,020 healthy women. For these healthy women, their ovarian reserve was quantified in the form of ovarian age, which was assumed equal to their chronological age, and least absolute shrinkage and selection operator (LASSO) regression was used to select features to construct models. Seven machine learning methods, namely artificial neural network (ANN), support vector machine (SVM), generalized linear model (GLM), K-nearest neighbors regression (KNN), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM) were applied to construct prediction models separately. Pearson's correlation coefficient (PCC), mean absolute error (MAE), and mean squared error (MSE) were used to compare the efficiency and stability of these models. Results: Anti-Müllerian hormone (AMH) and antral follicle count (AFC) were detected to have the highest absolute PCC values of 0.45 and 0.43 with age and held similar age distribution curves. The LightGBM model was thought to be the most suitable model for ovarian age after ranking analysis, combining PCC, MAE, and MSE values. The LightGBM model obtained PCC values of 0.82, 0.56, and 0.70 for the training set, the test set, and the entire dataset, respectively. The LightGBM method still held the lowest MAE and cross-validated MSE values. Further, in two different age groups (20–35 and >35 years), the LightGBM model also obtained the lowest MAE value of 2.88 for women between the ages of 20 and 35 years and the second lowest MAE value of 5.12 for women over the age of 35 years. Conclusion: Machine learning methods combining multi-features were reliable in assessing and quantifying ovarian reserve, and the LightGBM method turned out to be the approach with the best result, especially in the child-bearing age group of 20 to 35 years.

endocrinology & metabolism
Identification of Biomarkers for Predicting Ovarian Reserve of Primordial Follicle Via Transcriptomic Analysis

Li Liu,Biting Liu,Ke Li,Chunyan Wang,Yan Xie,Ning Luo,Lian Wang,Yaoqi Sun,Wei Huang,Zhongping Cheng,Shupeng Liu

DOI: https://doi.org/10.3389/fgene.2022.879974

IF: 3.7

2022-01-01

Frontiers in Genetics

Abstract:Ovarian reserve (OR) is mainly determined by the number of primordial follicles in the ovary and continuously depleted until ovarian senescence. With the development of assisted reproductive technology such as ovarian tissue cryopreservation and autotransplantation, growing demand has arisen for objective assessment of OR at the histological level. However, no specific biomarkers of OR can be used effectively in clinic nowadays. Herein, bulk RNA-seq datasets of the murine ovary with the biological ovarian age (BOA) dynamic changes and single-cell RNA-seq datasets of follicles at different stages of folliculogenesis were obtained from the GEO database to identify gene signature correlated to the primordial follicle pool. The correlations between gene signature expression and OR were also validated in several comparative OR models. The results showed that genes including Lhx8, Nobox, Sohlh1, Tbpl2, Stk31, and Padi6 were highly correlated to the OR of the primordial follicle pool, suggesting that these genes might be used as biomarkers for predicting OR at the histological level.
Prediction of ovarian cancer prognosis using statistical radiomic features of ultrasound images

Ruochen Zuo,Xiuru Li,Jiaqi Hu,Wenqian Wang,Bingjian Lu,Honghe Zhang,Xiaodong Cheng,Weiguo Lu,Jiale Qin,Pengyuan Liu,Yan Lu

DOI: https://doi.org/10.1088/1361-6560/ad4a02

IF: 3.5

2024-05-13

Physics in Medicine and Biology

Abstract:Ovarian cancer is the deadliest gynecologic malignancy worldwide. Ultrasound is the most useful non-invasive test for preoperative diagnosis of ovarian cancer. In this study, by leveraging multiple ultrasound images from the same patient to generate personalized, informative statistical radiomic features, we aimed to develop improved ultrasound image-based prognostic models for ovarian cancer. A total of 2,057 ultrasound images from 514 ovarian cancer patients, including 355 patients with epithelial ovarian cancer, from two hospitals in China were collected for this study. The models were constructed using our recently developed Frequency Appearance in Multiple Univariate pre-Screening (FAMUS) feature selection algorithm and Cox proportional hazards model. The models showed high predictive performance for overall survival (OS) and recurrence-free survival (RFS) in both epithelial and nonepithelial ovarian cancer, with concordance indices (C-index) ranging from 0.773 to 0.794. Radiomic scores predicted 2-year OS and RFS risk groups with significant survival differences (log-rank test, P<1.0e-4 for both validation cohorts). OS and RFS hazard ratios between low- and high-risk groups were 15.994 and 30.692 (internal cohort) and 19.339 and 19.760 (external cohort), respectively. The improved performance of these newly developed prognostic models was mainly attributed to the use of multiple preoperative ultrasound images from the same patient to generate statistical radiomic features, rather than simply using the largest tumor ROI among them. The models also revealed that the roundness of tumor lesion shape was positively correlated with prognosis for ovarian cancer. In summary, the newly developed prognostic models based on statistical radiomic features from ultrasound images were highly predictive of the risk of cancer-related death and possible recurrence not only for patients with epithelial ovarian cancer but also for those with nonepithelial ovarian cancer. They thereby provide reliable, non-invasive markers for individualized prognosis evaluation and clinical decision-making for patients with ovarian cancer.

engineering, biomedical,radiology, nuclear medicine & medical imaging
An Ovarian Reserve Assessment Model Based on Anti-Müllerian Hormone Levels, Follicle-Stimulating Hormone Levels, and Age: Retrospective Cohort Study.

Huiyu Xu,Li Shi,Guoshuang Feng,Zhen Xiao,Lixue Chen,Rong Li,Jie Qiao

DOI: https://doi.org/10.2196/19096

2020-01-01

Abstract:BackgroundPreviously, we reported a model for assessing ovarian reserves using 4 predictors: anti-Müllerian hormone (AMH) level, antral follicle count (AFC), follicle-stimulating hormone (FSH) level, and female age. This model is referred as the AAFA (anti-Müllerian hormone level–antral follicle count–follicle-stimulating hormone level–age) model. ObjectiveThis study aims to explore the possibility of establishing a model for predicting ovarian reserves using only 3 factors: AMH level, FSH level, and age. The proposed model is referred to as the AFA (anti-Müllerian hormone level–follicle-stimulating hormone level–age) model. MethodsOocytes from ovarian cycles stimulated by gonadotropin-releasing hormone antagonist were collected retrospectively at our reproductive center. Poor ovarian response (<5 oocytes retrieved) was defined as an outcome variable. The AFA model was built using a multivariable logistic regression analysis on data from 2017; data from 2018 were used to validate the performance of AFA model. Measurements of the area under the curve (AUC), sensitivity, specificity, positive predictive value, and negative predicative value were used to evaluate the performance of the model. To rank the ovarian reserves of the whole population, we ranked the subgroups according to the predicted probability of poor ovarian response and further divided the 60 subgroups into 4 clusters, A-D, according to cut-off values consistent with the AAFA model. ResultsThe AUCs of the AFA and AAFA models were similar for the same validation set, with values of 0.853 (95% CI 0.841-0.865) and 0.850 (95% CI 0.838-0.862), respectively. We further ranked the ovarian reserves according to their predicted probability of poor ovarian response, which was calculated using our AFA model. The actual incidences of poor ovarian response in groups from A-D in the AFA model were 0.037 (95% CI 0.029-0.046), 0.128 (95% CI 0.099-0.165), 0.294 (95% CI 0.250-0.341), and 0.624 (95% CI 0.577-0.669), respectively. The order of ovarian reserve from adequate to poor followed the order from A to D. The clinical pregnancy rate, live-birth rate, and specific differences in groups A-D were similar when predicted using the AFA and AAFA models. ConclusionsThis AFA model for assessing the true ovarian reserve was more convenient, cost-effective, and objective than our original AAFA model.
OC04.02: Prediction of non‐resectability in ovarian cancer using the predictive index value score: a prospective multicentre study using imaging (ISAAC study)

F. Moro,P. Pinto,V. Chiappa,A. Testa,J. Alcazar,D. Franchi,F. Frühauf,A. Burgetova,M. Mašek,L. Lambert,M. Arraiza,D. Altmanova,G. Avesani,C. Panico,S. Alessi,P. Pricolo,R. Vigorito,G. Calareso,K. Roman,J. Slama,A.M. Vidal Urbinati,A. Fagotti,G. Scambia,F. Raspagliesi,D. Cibula,K. Benesova,J. Jarkovsky,D. Fischerová

DOI: https://doi.org/10.1002/uog.27743

2024-09-11

Ultrasound in Obstetrics and Gynecology

Abstract:Ultrasound in Obstetrics &Gynecology, Volume 64, Issue S1, Page 10-10, September 2024.

radiology, nuclear medicine & medical imaging,obstetrics & gynecology,acoustics
Ultrasound radiomics-based artificial intelligence model to assist in the differential diagnosis of ovarian endometrioma and ovarian dermoid cyst

Lu Liu,Wenjun Cai,Chenyang Zhou,Hongyan Tian,Beibei Wu,Jing Zhang,Guanghui Yue,Yi Hao

DOI: https://doi.org/10.3389/fmed.2024.1362588

IF: 3.9

2024-03-08

Frontiers in Medicine

Abstract:Background Accurately differentiating between ovarian endometrioma and ovarian dermoid cyst is of clinical significance. However, the ultrasound appearance of these two diseases is variable, occasionally causing confusion and overlap with each other. This study aimed to develop a diagnostic classification model based on ultrasound radiomics to intelligently distinguish and diagnose the two diseases. Methods We collected ovarian ultrasound images from participants diagnosed as patients with ovarian endometrioma or ovarian dermoid cyst. Feature extraction and selection were performed using the Mann-Whitney U -test, Spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO) regression. We then input the final features into the machine learning classifiers for model construction. A nomogram was established by combining the radiomic signature and clinical signature. Results A total of 407 participants with 407 lesions were included and categorized into the ovarian endometriomas group ( n = 200) and the dermoid cyst group ( n = 207). In the test cohort, Logistic Regression (LR) achieved the highest area under curve (AUC) value (0.981, 95% CI: 0.963−1.000), the highest accuracy (94.8%), and the highest sensitivity (95.5%), while LightGBM achieved the highest specificity (97.1%). A nomogram incorporating both clinical features and radiomic features achieved the highest level of performance (AUC: 0.987, 95% CI: 0.967−1.000, accuracy: 95.1%, sensitivity: 88.0%, specificity: 100.0%, PPV: 100.0%, NPV: 88.0%, precision: 93.6%). No statistical difference in diagnostic performance was observed between the radiomic model and the nomogram ( P > 0.05). The diagnostic indexes of radiomic model were comparable to that of senior radiologists and superior to that of junior radiologist. The diagnostic performance of junior radiologists significantly improved with the assistance of the model. Conclusion This ultrasound radiomics-based model demonstrated superior diagnostic performance compared to those of junior radiologists and comparable diagnostic performance to those of senior radiologists, and it has the potential to enhance the diagnostic performance of junior radiologists.

medicine, general & internal
OC01.05: Ultrasound radiomics for differentiating borderline from epithelial ovarian cancer, compared with intraoperative frozen section

J. Qin

DOI: https://doi.org/10.1002/uog.27725

2024-09-11

Ultrasound in Obstetrics and Gynecology

Abstract:Ultrasound in Obstetrics &Gynecology, Volume 64, Issue S1, Page 2-3, September 2024.

radiology, nuclear medicine & medical imaging,obstetrics & gynecology,acoustics
Three-dimensional imaging and quantification of mouse ovarian follicles via optical coherence tomography

Marcello Magri Amaral,Aixia Sun,Yilin Li,Chao Ren,Anh Blue Truong,Saumya Nigam,Zexu Jiao,Ping Wang,Chao Zhou

DOI: https://doi.org/10.1364/boe.489618

IF: 3.4

2023-06-12

Biomedical Optics Express

Abstract:Ovarian tissue cryopreservation has been successfully applied worldwide for fertility preservation. Correctly selecting the ovarian tissue with high follicle loading for freezing and reimplantation increases the likelihood of restoring ovarian function, but it is a challenging process. In this work, we explore the use of three-dimensional spectral-domain optical coherence tomography (SD-OCT) to identify different follicular stages, compare the identifications with H&E images, and measure the size and age-related follicular density distribution differences in mice ovaries. We use the thickness of the layers of granulosa cells to differentiate primordial and primary follicles from secondary follicles. The measured dimensions and age-related follicular distribution agree well with histological images and physiological aging. Finally, we apply attenuation coefficient map analyses to significantly improve the image contrast and the contrast-to-noise ratio (p < 0.001), facilitating follicle identification and quantification. We conclude that SD-OCT is a promising method to noninvasively evaluate ovarian follicles for ovarian tissue cryopreservation.

radiology, nuclear medicine & medical imaging,biochemical research methods,optics
The Comparison of Evaluative Effectiveness Between Antral Follicle Count/age Ratio and Ovarian Response Prediction Index for the Ovarian Reserve and Response Functions in Infertile Women

Shan-Jie Zhou,Ming-Jia Zhao,Cui Li,Xing Su

DOI: https://doi.org/10.1097/md.0000000000021979

IF: 1.6

2020-01-01

Medicine

Abstract:Abstract Aim of the present study was to explore the evaluative effectiveness of age, ovarian volume (OV), antral follicle count (AFC), serum follicle-stimulating hormone (FSH), anti-Müllerian hormone (AMH), AFC/Age ratio, AMH/Age ratio, FSH/luteinizing hormone (LH) ratio, and ovarian response prediction index (ORPI) to determine which could more advantageously assess ovarian reserve and response. This research enrolled 319 consecutive infertile women who had undergone in vitro fertilization-ET/intracytoplasmic sperm injection (IVF-ET/ICSI) treatments. Abovementioned variables were measured and calculated. Receiver operating characteristic (ROC) curve analysis was used to analyze the predictive accuracy of variables and to calculate cut-off values and corresponding sensitivity and specificity. Our study revealed that the significant variables for evaluating a decline in ovarian reserve include age, OV, FSH, AFC/Age ratio, AMH/Age ratio, and ORPI. Moreover, the area under the curve (AUC) of AFC/Age ratio was higher than other 5 variables (AUC = 1.000), and the cut-off value of AFC/Age ratio was 0.111 (sensitivity 100.00%, specificity 100.00%). The significant variables forecasting excessive ovarian response were age, AFC, AMH, FSH, AFC/Age ratio, AMH/Age ratio, FSH/LH ratio, and ORPI, and the significant variables forecasting poor ovarian response were AMH, LH, OV, AFC/Age ratio, AMH/Age ratio, and FSH/LH ratio. When ORPI was used to predict excessive response, the cut-off value of ORPI was 0.880 (sensitivity 84.72%, specificity 67.32%) and ORPI presented better effectiveness. When used to predict poor response, the evaluative effectiveness of 6 variables was almost similar, although the AUC of AFC/Age ratio presented the largest value. Regarding the infertile women, AFC/Age ratio performed better than did the other variables in evaluating ovarian reserve, and it offered excellent effectiveness in predicting poor ovarian response, however, ORPI presented better effectiveness in predicting excessive ovarian response.
An Online Tool for Predicting Ovarian Responses in Unselected Patients Using Dynamic Inhibin B and Basal Antimüllerian Hormone Levels

Congcong Ma,Huiyu Xu,Haiyan Wang,Guoshuang Feng,Yong Han,Kannan Alpadi,Rong Li,Jie Qiao

DOI: https://doi.org/10.3389/fendo.2023.1074347

IF: 6.055

2023-01-01

Frontiers in Endocrinology

Abstract:BackgroundReliable predictive models for predicting excessive and poor ovarian response in controlled ovarian stimulation (COS) is currently lacking. The dynamic (Δ) inhibin B, which refers to increment of inhibin B responding to exogenous gonadotropin, has been indicated as a potential predictor of ovarian response.ObjectiveTo establish mathematical models to predict ovarian response at the early phase of COS using Δinhibin B and other biomarkers.Materials and methodsProspective cohort study in a tertiary teaching hospital, including 669 cycles underwent standard gonadotropin releasing hormone (GnRH) antagonist ovarian stimulation between April 2020 and September 2020. Early Δinhibin B was defined as an increment in inhibin B from menstrual day 2 to day 6 through to the day of COS. Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression with 5-fold cross-validation was applied to construct ovarian response prediction models. The area under the receiver operating characteristic curve (AUC), prevalence, sensitivity, and specificity were used for evaluating model performance.ResultsEarly Δinhibin B and basal antimüllerian hormone (AMH) levels were the best measures in building models for predicting ovarian hypo- or hyper-responses, with AUCs and ranges of 0.948 (0.887–0.976) and 0.904 (0.836–0.945) in the validation set, respectively. The contribution of the early Δinhibin B was 67.7% in the poor response prediction model and 56.4% in the excessive response prediction model. The basal AMH level contributed 16.0% in the poor response prediction model and 25.0% in the excessive response prediction model. An online website-based tool (http://121.43.113.123:8001/) has been developed to make these complex algorithms available in clinical practice.ConclusionEarly Δinhibin B might be a novel biomarker for predicting ovarian response in IVF cycles. Limiting the two prediction models to the high and the very-low risk groups would achieve satisfactory performances and clinical significance. These novel models might help in counseling patients on their estimated ovarian response and reduce iatrogenic poor or excessive ovarian responses.
Ultrasound Features Improve Diagnostic Performance of Ovarian Cancer Predictors in Distinguishing Benign and Malignant Ovarian Tumors

Yong-ning Chen,Fei Ma,Ya-di Zhang,Li Chen,Chan-yuan Li,Shi-peng Gong

DOI: https://doi.org/10.1007/s11596-020-2163-7

2020-02-01

Current Medical Science

Abstract:<h3 class="a-plus-plus">Summary</h3><p class="a-plus-plus">To determine whether ultrasound features can improve the diagnostic performance of tumor markers in distinguishing ovarian tumors, we enrolled 719 patients diagnosed as having ovarian tumors at Nanfang Hospital from September 2014 to November 2016. Age, menopausal status, histopathology, the International Federation of Gynecology and Obstetrics (FIGO) stages, tumor biomarker levels, and detailed ultrasound reports of patients were collected. The area under the curve (AUC), sensitivity, and specificity of the bellow-mentioned predictors were analyzed using the receiver operating characteristic curve. Of the 719 patients, 531 had benign lesions, 119 had epithelial ovarian cancers (EOC), 44 had borderline ovarian tumors (BOT), and 25 had non-EOC. AUCs and the sensitivity of cancer antigen 125 (CA125), human epididymis-specific protein 4 (HE4), Risk of Ovarian Malignancy Algorithm (ROMA), Risk of Malignancy Index (RMI1), HE4 model, and Rajavithi-Ovarian Cancer Predictive Score (R-OPS) in the overall population were 0.792, 0.854, 0.856, 0.872, 0.893, 0.852, and 70.2%, 56.9%, 69.1%, 60.6%, 77.1%, 71.3%, respectively. For distinguishing EOC from benign tumors, the AUCs and sensitivity of the above mentioned predictors were 0.888, 0.946, 0.947, 0.949, 0.967, 0.966, and 84.0%, 79.8%, 87.4%, 84.9%, 90.8%, 89.1%, respectively. Their specificity in predicting benign diseases was 72.9%, 94.4%, 87.6%, 95.9%, 86.3%, 90.8%, respectively. Therefore, we consider biomarkers in combination with ultrasound features may improve the diagnostic performance in distinguishing malignant from benign ovarian tumors.</p>

medicine, research & experimental
Machine Learning-Based Modeling of Ovarian Response and the Quantitative Evaluation of Comprehensive Impact Features

Liu Liu,Fujin Shen,Hua Liang,Zhe Yang,Jing Yang,Jiao Chen

DOI: https://doi.org/10.3390/diagnostics12020492

IF: 3.6

2022-02-14

Diagnostics

Abstract:Appropriate ovarian responses to the controlled ovarian stimulation strategy is the premise for a good outcome of the in vitro fertilization cycle. With the booming of artificial intelligence, machine learning is becoming a popular and promising approach for tailoring a controlled ovarian stimulation strategy. Nowadays, most machine learning-based tailoring strategies aim to generally classify the controlled ovarian stimulation outcome, lacking the capacity to precisely predict the outcome and evaluate the impact features. Based on a clinical cohort composed of 1365 women and two machine learning methods of artificial neural network and supporting vector regression, a regression prediction model of the number of oocytes retrieved is trained, validated, and selected. Given the proposed model, an index called the normalized mean impact value is defined and calculated to reflect the importance of each impact feature. The proposed models can estimate the number of oocytes retrieved with high precision, with the regression coefficient being 0.882% and 89.84% of the instances having the prediction number ≤ 5. Among the impact features, the antral follicle count has the highest importance, followed by the E2 level on the human chorionic gonadotropin day, the age, and the Anti-Müllerian hormone, with their normalized mean impact value > 0.3. Based on the proposed model, the prognostic results for ovarian response can be predicted, which enables scientific clinical decision support for the customized controlled ovarian stimulation strategies for women, and eventually helps yield better in vitro fertilization outcomes.

medicine, general & internal
OP06.02: Deep learning radiomics based on contrast‐enhanced ultrasound images for accurate diagnosis of adnexal lesions: a retrospective multicentre study

M. Wu,H. Yang,Y. Wang,J. Hu,X. Zhang

DOI: https://doi.org/10.1002/uog.27917

2024-09-11

Ultrasound in Obstetrics and Gynecology

Abstract:Ultrasound in Obstetrics &Gynecology, Volume 64, Issue S1, Page 74-74, September 2024.

radiology, nuclear medicine & medical imaging,obstetrics & gynecology,acoustics
Quantification of follicles in human ovarian tissue using image processing software and trained artificial intelligence

Gabrielle M Blevins,Colleen L Flanagan,Sridula S Kallakuri,Owen M Meyer,Likitha Nimmagadda,James D Hatch,Sydney A Shea,Vasantha Padmanabhan,Ariella Shikanov

DOI: https://doi.org/10.1093/biolre/ioae048

2024-03-27

Biology of Reproduction

Abstract:Abstract Cancer survival rates in prepubertal girls and young women have risen in recent decades due to increasingly efficient treatments. However, many such treatments are gonadotoxic, causing premature ovarian insufficiency (POI), loss of fertility and ovarian endocrine function. Implantation of donor ovarian tissue encapsulated in immune-isolating capsules is a promising method to restore physiological endocrine function without immunosuppression or risk of reintroducing cancer cells harbored by the tissue. The success of this approach is largely determined by follicle density (FD) in the implanted ovarian tissue, which is analyzed manually from histologic sections and necessitates specialized, time-consuming labor. To address this limitation, we developed a fully automated method to quantify FD that does not require additional coding. We first analyzed ovarian tissue from 12 human donors between 16 to 37 years old using semi-automated image processing with manual follicle annotation and then trained artificial intelligence program based on follicle identification and object classification. One operator manually analyzed 102 whole slide images (WSIs) from serial histologic sections. Of those, 77 images were assessed by a second manual operator, followed with an automated method utilizing artificial intelligence (AI). Of the 1181 follicles the control operator counted, the comparison operator counted 1178, and the AI counted 927 follicles with 80% of those being correctly identified as follicles. The three-stage AI pipeline finished 33% faster than manual annotation. Collectively, this report supports the use of artificial intelligence and automation to select tissue donors and grafts with the greatest FD to ensure graft longevity for POI treatment.

reproductive biology
An online tool for predicting ovarian reserve based on AMH level and age: A retrospective cohort study

Yong Han,Huiyu Xu,Guoshuang Feng,Haiyan Wang,Kannan Alpadi,Lixue Chen,Mengqian Zhang,Rong Li

DOI: https://doi.org/10.3389/fendo.2022.946123

IF: 6.055

2022-07-22

Frontiers in Endocrinology

Abstract:Purpose To establish a more convenient ovarian reserve model with anti-Müllerian hormone (AMH) level and age (the AA model), with blood samples taken at any time in the menstrual cycle. Methods We have established this AA model for predicting ovarian reserve using the AMH level and age. The outcome variable was defined as poor ovarian response (POR) with <5 oocytes retrieved during assisted reproductive technology treatment cycles. Least Absolute Shrinkage and Selection Operator logistic regression with 5-fold cross validation methods was applied to construct the model, and that with the lowest scaled log-likelihood was selected as the final one. Results The areas under the receiver operating characteristic curve for the training, inner, and external validation sets were 0.862, 0.843, and 0.854 respectively. The main effects of AMH level and age contributing to the prediction of POR were 95.3% and 1.8%, respectively. The incidences of POR increased with its predicted probability in both the model building and in external validation datasets, indicating its stability. An online website-based tool for assessing the score of ovarian reserve ( http://121.43.113.123:9999 ) has been developed. Conclusions Based on external validation data, the AA model performed well in predicting POR, and was more cost-effective and convenient than our previous published models.

endocrinology & metabolism
An ultrasound-based deep learning radiomic model combined with clinical data to predict the success of frozen embryo transfer: a pilot cohort study

Xiaowen Liang,Jianchong He,Lu He,Yan Lin,Yuewei Li,Kuan Cai,Jun Wei,Yao Lu,Zhiyi Chen

DOI: https://doi.org/10.1016/j.rbmo.2023.03.015

IF: 4.567

2023-03-29

Reproductive BioMedicine Online

Abstract:Research question Can a multi-modal fusion model based on ultrasound (US) -based deep learning (DL) radiomics combined with clinical parameters provide well personalized evaluation of endometrial receptivity and predict the outcome of frozen embryo transfer (FET)? Design This prospective cohort study enrolled 326 patients who underwent FET from August 2019 to December 2021. Input quantitative variables (included hormone concentration and multiple modes ultrasound parameters) and input image data for radiomic feature extraction (original endometrial two-dimensional ultrasound images) were collected to establish a multi-modal fusion prediction model. An additional independent dataset of 453 US endometrial images was used to establish the segmentation model to determine the endometrial region on US images for analysis. The performance of different algorithms and different input data for prediction of FET outcome were compared. Results The proposed multi-modal fusion model performed significantly better than the single use of image or quantitative variable to predict the success of FET. Its AUC, accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the proposed model were 0.825, 72.5%, 96.2%, 58.3%, 72.3%, and 89.5%, respectively. The Dice coefficient of the multi-task endometrial US segmentation model was 0.89. The use of the endometrial segmentation features significantly improved the prediction performance of the model ( p = 0.041). Conclusions The proposed multi-modal fusion model based on US-based DL radiomics combined with clinical quantitative variables offers a favorable and rapid noninvasive approach for personalized prediction of FET outcome.

obstetrics & gynecology,reproductive biology

OC11.03: Prediction of ovarian reserve using ultrasound imaging‐based radiomics and deep learning: comparison with antral follicle count and anti‐Müllerian hormone

A Clinician-Friendly Machine Learning System to Predict Ovarian Response and Deploy Individualized Ovarian Stimulation Strategies in IVF

Anti-Müllerian hormone for prediction of ovarian response in Chinese infertile women undergoing IVF/ICSI cycles: a prospective, multi-centre, observational study

Evaluation of oocyte maturity using artificial intelligence quantification of follicle volume biomarker by three-dimensional ultrasound

Assessment and quantification of ovarian reserve on the basis of machine learning models

Identification of Biomarkers for Predicting Ovarian Reserve of Primordial Follicle Via Transcriptomic Analysis

Prediction of ovarian cancer prognosis using statistical radiomic features of ultrasound images

An Ovarian Reserve Assessment Model Based on Anti-Müllerian Hormone Levels, Follicle-Stimulating Hormone Levels, and Age: Retrospective Cohort Study.

OC04.02: Prediction of non‐resectability in ovarian cancer using the predictive index value score: a prospective multicentre study using imaging (ISAAC study)

Ultrasound radiomics-based artificial intelligence model to assist in the differential diagnosis of ovarian endometrioma and ovarian dermoid cyst

OC01.05: Ultrasound radiomics for differentiating borderline from epithelial ovarian cancer, compared with intraoperative frozen section

Three-dimensional imaging and quantification of mouse ovarian follicles via optical coherence tomography

The Comparison of Evaluative Effectiveness Between Antral Follicle Count/age Ratio and Ovarian Response Prediction Index for the Ovarian Reserve and Response Functions in Infertile Women

An Online Tool for Predicting Ovarian Responses in Unselected Patients Using Dynamic Inhibin B and Basal Antimüllerian Hormone Levels

Ultrasound Features Improve Diagnostic Performance of Ovarian Cancer Predictors in Distinguishing Benign and Malignant Ovarian Tumors

Machine Learning-Based Modeling of Ovarian Response and the Quantitative Evaluation of Comprehensive Impact Features

OP06.02: Deep learning radiomics based on contrast‐enhanced ultrasound images for accurate diagnosis of adnexal lesions: a retrospective multicentre study

Quantification of follicles in human ovarian tissue using image processing software and trained artificial intelligence

An online tool for predicting ovarian reserve based on AMH level and age: A retrospective cohort study

An ultrasound-based deep learning radiomic model combined with clinical data to predict the success of frozen embryo transfer: a pilot cohort study