ABSTRACT
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.
ABSTRACT
Objective: The pregnancy patterns and other factors of live birth for patients with intrauterine adhesions (IUAs) were identified by analyzing the clinical features of pre-, intra-, and post-hysteroscopic adhesiolysis (HA). Design: A total of 742 patients with IUAs who wanted to become pregnant underwent HA from January 2017 to May 2018 at the Third Xiangya Hospital of Central South University. The patient follow-up period was 2 years post-HA. A logistic regression was performed to analyze the clinical characteristics associated with a live birth for patients with IUAs. Pre-operative clinical indicators included age, gravidity, parity, abortion, IUA recurrence, menstrual patterns, disease course. Intraoperative clinical features assessed in the last operation were uterine cavity length, IUA appearance, IUA area, number of visible uterine cornua, number of visible tubal ostia, AFS scores. Pregnancy patterns were post-hysteroscopic adhesiolysis features. Results: Among the 742 IUA patients, 348 (46.9%) had a live birth and 394 (53.1%) did not. A bivariate and binary logistic regression analysis showed that IUA patients' pregnancy patterns, age, number of visible tubal ostia noted by a second-look hysteroscopy, and American Fertility Society (AFS) scores were significantly related to the live birth rate (P < 0.05). Conclusions: Pregnancy patterns, age, number of visible tubal ostia, and AFS scores were significantly related to the live birth rate and may be considered potential predictors of the live birth rate in IUA patients. The indications of assisted reproductive technology (ART) might be a better choice for patients with recurrent IUAs.
ABSTRACT
BACKGROUND: Intrauterine adhesions (IUAs) can be dissected using hysteroscopic scissors (cold scissors) or other methods, but there is no consensus on which hysteroscopic method is preferable. There is also no consensus on the method of how to deal with the scar tissue on the surface of the intrauterine cavity. METHODS: From January 2016 to October 2017, 179 patients who had HA met the enrollment criteria (see the text below), and their data were analyzed retrospectively. In addition, all patients were divided into three groups according to the surgical techniques used. The groups were the ploughing group (PG) (using cold scissors to dissect the adhesion and cut the scar tissue using a ploughing technique) (n=81), the traditional group (TG) (using cold scissors to dissect the adhesion, but not deal with the scar tissue) (n=42), and the electrosurgical group (EG) (using a resectoscope to dissect the adhesion with an energy L-hook electrode, and not deal with the scar tissue) (n=56). Safety (surgical complications), feasibility (surgical technique replacement rate), and postoperative efficacy (reduction of AFS score, pregnancy, and live birth rate), were each evaluated between groups. RESULTS: No statistically significant differences between the groups were observed in basic preoperative information (P>0.05), while there were significant differences between PG and TG, as well as PG and EG in postoperative AFS scores (PG vs. TG: P=0.007; PG vs. EG: P<0.001) and pregnancy outcome (PG vs. TG: P=0.039; PG vs. EG: P<0.001). No patients had surgical complications such as uterine perforations, moderate or severe fluid overload, heavy uterine bleeding, nor any surgical technique replacements (for example, transfer to use a resectoscope). CONCLUSIONS: Cold scissors ploughing technique in HA is effective, feasible, and safe, and thus worthy of further study.
ABSTRACT
BACKGROUND: Cervical insufficiency (CI) with concomitant intrauterine adhesions (IUAs) is a common clinical phenomenon among CI patients. But there are neither published reports regarding the difference in diagnosis and treatment of such patients compared to those with CI only, nor any report about their prognosis. This study aimed to preliminary the alteration in diagnostic and curative aspects of these patients, so as to provide a certain reference for the clinical management of such conditions. METHODS: Ten patients with CI combined with moderate to severe IUAs were diagnosed, treated and followed up at the Third Xiangya Hospital of Central South University from September 2017 to August 2019, their medical records and the pregnancy outcomes were retrospectively analyzed. RESULTS: All 10 patients had a previous history of typical painless cervical dilatation during the second trimester. All patients were moderate to severer IUAs, and the mean AFS score of IUAs was 9.80±1.08 (range, 8 to 12). Preoperatively, in 6 patients, the No. 7 Hegar dilator was able to pass through the internal cervical os before surgery without resistance. In the other 4 patients, the Hegar dilator could not be inserted before surgery due to the adhesions of the cervical canal and the lower uterine segment; the diagnoses of these patients were further confirmed at 3 months after hysteroscopic adhesiolysis (HA) when the No. 7 Hegar dilator was able to pass through the internal cervical os without resistance. There were 9 patients underwent pre-pregnancy laparoscopic cervical cerclage after HA. The remaining 1 patient exceptionally underwent laparoscopic cervical cerclage prior to HA, as the cervix was too loose to retain and be treated with an intrauterine device (IUD) or distended Foley's catheter balloon; which essentially prevent postoperative adhesion reformation. The patients were followed-up for 3 months to 2 years. The pregnancy rate was 60%, and the live birth rate was 100%. CONCLUSIONS: In patients with CI and concomitant cervical or lower uterine segment IUAs, it is necessary to separate the adhesion prior to evaluating the cervical competency with the No. 7 Hegar dilator, to confirm the diagnosis. However, when the cervix is too loose, laparoscopic cervical cerclage is exceptionally carried out first and then IUAs is treated. Pre-pregnancy laparoscopic cervical cerclage has a good prognosis in patients with CI complicated by moderate to severe IUAs.
ABSTRACT
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.
