Background: There were a very large number of intrauterine adhesion (IUA) patients. As improving the classification of three-dimensional transvaginal ultrasound (3D-TVUS) of IUA or non-IUA images remains a clinical challenge and is needed to avoid inappropriate surgery. Our study aimed to evaluate deep learning as a method to classify 3D-TVUS of IUA or non-IUA images taken with panoramic technology. Methods: After meeting an inclusion/exclusion criteria, a total of 4,401 patients were selected for this study. This included 2,803 IUA patients and 1,598 non-IUA patients. IUA was confirmed by hysteroscopy, and each patient underwent one 3D-TVUS examination. Four well-known convolutional neural network (CNN) architectures were selected to classify the IUA images: Visual Geometry Group16 (VGG16), InceptionV3, ResNet50, and ResNet101. We used these pretrained CNNs on ImageNet by applying both TensorFlow and PyTorch. All 3D-TVUS images were normalized and mixed together. We split the data set into a training set, validation set, and test set. The performance of our classification model was evaluated according to sensitivity, precision, F1-score, and accuracy, which were determined by equations that used true-positive (TP), false-positive (FP), true-negative (TN), and false-negative (FN) numbers. Results: The overall performances of VGG16, InceptionV3, ResNet50, and ResNet101 were better in PyTorch as opposed to TensorFlow. Through PyTorch, the best CNN model was InceptionV3 with its performance measured as 94.2% sensitivity, 99.4% precision, 96.8% F1-score, and 97.3% accuracy. The area under the curve (AUC) results of VGG16, InceptionV3, ResNet50, and ResNet101 were 0.959, 0.999, 0.997, and 0.999, respectively. PyTorch also successfully transferred information from the source to the target domain where we were able to use another center's data as an external test data set. No overfitting that could have adversely affected the classification accuracy occurred. Finally, we successfully established a webpage to diagnose IUA based on the 3D-TVUS images. Conclusions: Deep learning can assist in the binary classification of 3D-TVUS images to diagnose IUA. This study lays the foundation for future research into the integration of deep learning and blockchain technology.
The septate uterus is the most common structural uterine anomalies and it is associated with the poor reproductive outcome. It is believed to be the result of the failure in resorption of the tissue connecting the 2 paramesonephric ducts prior to the 20th embryonic week. The true prevalence of uterine septum is difficult to ascertain, as many uterine septal defects are asymptomatic. The septate uterus is usually diagnosed during an infertility evaluation and affects reproductive health by impairing fertility and increasing adverse pregnancy outcomes. The variations in uterine and cervical/vaginal anomalies collectively referred to as Müllerian anomalies. No consistent gold standard for the diagnosis of Müllerian anomalies exists. The preferred diagnostic method for Müllerian anomalies is two-dimensional ultrasound, other methods such as three-dimensional ultrasound, magnetic resonance imaging, hysterosalpingo contrast sonography, hysterosalpingography, hysteroscopy, and laparoscopy are also used to improve accuracy.
Septate Uterus , Female , Humans
OBJECTIVES: The incidence of infertility is increasing, more than 30% of them having related abnormal tubal patency. Four-dimensional (4D) hysterosalpingo-contrast sonography (HyCoSy) overcomes the shortcomings of 3D HyCoSy in the diagnosis of tubal patency, showing high specificity and accuracy. In addition, 4D HyCoSy discards iodine allergy and X-ray radiation and possesses easy-operating, contributing to good acceptance in clinical practice. However, there is no research to explore the imaging standards related to the possibility of natural pregnancy after 4D HyCoSy. If a predictive model of postoperative natural pregnancy was established using the analysis of clinical data combined with imaging characteristics of 4D HyCoSy of patients with tubal factor infertility, clinical decision-making can be wisely guided in the future. This study aims to establish a predictive model of natural pregnancy after 4D HyCoSy based on clinical data and imaging characteristics of patients with tubal factor in fertility. METHODS: A retrospective study was conducted for patients who were diagnosed with tubal factor infertility in Hunan Guangxiu Hospital from February 2017 to May 2018. The patients ought to possess complete 4D HyCoSy imaging data and at least one-side-unobstructed fallopian tube. General clinical data and imaging data were collected. Pregnancy outcome was followed up till 3 months after 4D HyCoSy. According to pregnancy outcome, patients were divided into a pregnancy group and a non-pregnancy group. Binary logistic regression was used to analyze the correlation between various variables and natural pregnancy after 4D HyCoSy. The variables with significant difference (P<0.05) in single-factor logistic regression were included in the natural pregnancy probability prediction model. The classification accuracy was further verified with 10-fold cross-validation. RESULTS: A total of 1 085 patients with clinically suspected tubal factor infertility who met the requirements and followed the doctors' prescription were collected. Clinical characteristics (age and duration of infertility) and 4D HyCoSy imaging characteristics (thickness of endometrium from the 3rd to the 7th day after the end of menstruation, visualization of the left fallopian tube, the diffusion of contrast agent around the left ovary, and the diffusion of contrast agent around the right ovary) were independent predictors for natural pregnancy 3 months after 4D HyCoSy. A natural pregnancy probability prediction model was established with the area under the curve (AUC) verified by the 10-fold cross-validation all greater than 0.75, and the best AUC was 0.868. The Q value obtained by the prediction model was the probability of natural pregnancy, and the cutoff value was 0.5. When the Q value was greater than 0.5, it was recommended to attempt natural pregnancy for 3 months, and when the Q value was less than 0.5, in vitro fertilization was adviced. CONCLUSIONS: A predictive model for the evaluating probability of natural pregnancy in women with tubal factor infertility after 4D HyCoSy is successfully established based on the analysis for clinical data and imaging characteristics. This model shows a great potential in assisting clinical decision making.
Contrast Media , Humans , Female , Pregnancy , Retrospective Studies
BACKGROUND: Hysteroscopic adhesiolysis (HA) remains the mainstay of treatment for intrauterine adhesions (IUA). In cases of moderate or severe IUA, the assistance of various adjunctive aids are usually sought to improve HA's success rate. Among these, intraoperative transabdominal ultrasound (TAS) is the most common; however, it has certain limitations. Preoperative three-dimensional transvaginal ultrasound (3D-TVUS) has been accepted as a non-invasive way to provide accurate information about the uterine cavity. This prospective, non-randomized controlled study will assess the effects of pre-operative 3D-TVUS prior to HA in improving the surgeon's intraoperative judgement. METHODS: A total of 362 patients, who met the inclusion criteria, aged between 18 and 45 years and diagnosed with moderate or severe IUA underwent HA at our hospital from March 2018 to December 2018. Participants were divided into 2 groups; the study group; n=182 performed 3D-TVUS evaluation prior to HA, and the control group; n=180 underwent HA without preoperative 3D-TVUS evaluation. The following basic information were collected prospectively for both groups: age, parity, history of abortion, degree of IUA, surgical complications and number of hysteroscopic interventions. The data obtained from 3D-TVUS in the study group was carefully studied at the preoperative stage by the operator and was integrated into intraoperative findings, further assisting with intraoperative decisions. The guiding value of preoperative 3D-TVUS for HA was evaluated by comparing and analyzing the postoperative exposure rate of clearly visible tubal ostia between the groups. RESULTS: Based on the basic information (P>0.05) collected preoperatively, there were no statistically significant differences between the groups. Postoperatively, the study group had a better surgical success rate with a more significant AFS score reduction (4.71±2.05; P<0.0001) and better morphological restoration of the uterine cavity, with more adhesion-free uterine horns and more clearly visible fallopian tube ostia (P<0.0001) following HA. CONCLUSIONS: This study showed that preoperative 3D-TVUS evaluation helped the hysteroscopists with their intraoperative decision-making while carrying out HA. In comparison to those who did not perform preoperative 3D ultrasound, those who underwent 3D-TVUS evaluation had a better surgical success rate in retrieving the fallopian tube ostia and the restoration of normal uterine cavity morphology.
BACKGROUND: Cellular therapy of human embryonic stem cell-derived cardiomyocytes (hES-CMs) holds great promise for regenerating infarcted cardiac tissues. Yet, the major challenge remains as little cells survived after grafting. In this study, we examined whether treating hES-CMs with 7, 8, 3'-Trihydroxyflavone (THF) may improve hES-CMs developments both in vitro and in vivo. METHOD: HES-CMs were differentiated in vitro, and treated with 5 µ59M THF for 24 h. The control hES-CMs were treated with PBS. Possible effect of THF on hES-CM differentiation was assessed by viability and western blot assays. HES-CMs were also treated with hypoxia/reoxygenation (H/R)-condition to induce apoptosis. The effect of THF on rescuing H/R-induced hES-CM apoptosis was assessed by TUNEL and western blot assays. HES-CMs were then grafted into infarcted rat hearts. The effect of THF on promoting in vivo growth of hES-CMs was examined by immunohistochemistry. RESULTS: THF pretreatment did not alter the differentiation process of hES-CMs. Under H/R condition, THF rescued hES-CM apoptosis, activated TrkA and TrkB signaling pathways through phosphorylation and induced VEGF production. In in vivo rat model of myocardial infarction, THF induced the growth of transplanted hES-CMs by promoting Cardiac Troponin I and CD31. CONCLUSION: THF has pro-cardiac effect on hES-CMs by protecting cells from H/R induced apoptosis and promoting in vivo growth of cell transplantation in infarcted hearts. These results may help optimizing the cellular therapy of using human embryonic stem cells to benefiting patients suffered from heart attack. This article is protected by copyright. All rights reserved.
