Position-Sensing Graph Neural Networks: Proactively Learning Nodes Relative Positions.

Most existing graph neural networks (GNNs) learn node embeddings using the framework of message passing and aggregation. Such GNNs are incapable of learning relative positions between graph nodes within a graph. To empower GNNs with the awareness of node positions, some nodes are set as anchors. Then, using the distances from a node to the anchors, GNNs can infer relative positions between nodes. However, position-aware GNNs (P-GNNs) arbitrarily select anchors, leading to compromising position awareness and feature extraction. To eliminate this compromise, we demonstrate that selecting evenly distributed and asymmetric anchors is essential. On the other hand, we show that choosing anchors that can aggregate embeddings of all the nodes within a graph is NP-complete. Therefore, devising efficient optimal algorithms in a deterministic approach is practically not feasible. To ensure position awareness and bypass NP-completeness, we propose position-sensing GNNs (PSGNNs), learning how to choose anchors in a backpropagatable fashion. Experiments verify the effectiveness of PSGNNs against state-of-the-art GNNs, substantially improving performance on various synthetic and real-world graph datasets while enjoying stable scalability. Specifically, PSGNNs on average boost area under the curve (AUC) more than 14% for pairwise node classification and 18% for link prediction over the existing state-of-the-art position-aware methods. Our source code is publicly available at: https://github.com/ZhenyueQin/PSGNN.

Disentangling Noise from Images: A Flow-Based Image Denoising Neural Network.

The prevalent convolutional neural network (CNN)-based image denoising methods extract features of images to restore the clean ground truth, achieving high denoising accuracy. However, these methods may ignore the underlying distribution of clean images, inducing distortions or artifacts in denoising results. This paper proposes a new perspective to treat image denoising as a distribution learning and disentangling task. Since the noisy image distribution can be viewed as a joint distribution of clean images and noise, the denoised images can be obtained via manipulating the latent representations to the clean counterpart. This paper also provides a distribution-learning-based denoising framework. Following this framework, we present an invertible denoising network, FDN, without any assumptions on either clean or noise distributions, as well as a distribution disentanglement method. FDN learns the distribution of noisy images, which is different from the previous CNN-based discriminative mapping. Experimental results demonstrate FDN's capacity to remove synthetic additive white Gaussian noise (AWGN) on both category-specific and remote sensing images. Furthermore, the performance of FDN surpasses that of previously published methods in real image denoising with fewer parameters and faster speed.

Application of modified subtotal resection of adenomyosis combined with LNG-IUS and GnRH-a sequential therapy in severe adenomyosis: A case series.

Background and Objective: Adenomyosis focus resection has always been the main surgical method for patients with uterine preservation, but its curative effect and surgical method are still controversial. We improved this method on the basis of the "double-flap method" and combined it with the levonorgestrel intrauterine delivery system (LNG-IUS) and gonadotropin-releasing hormone agonist (GnRH-a) sequential treatment to determine the clinical effect and feasibility of this scheme in the treatment of severe adenomyosis. Methods: This is a retrospective review. A total of 64 patients with severe adenomyosis were treated in the Department of Gynecology of Changzhou Second People's Hospital, which is affiliated to Nanjing Medical University, from December 2017 to September 2021. The transabdominal approach and laparoscopic approach were adopted for the purposes of treatment in this study. Hence, the patients were subdivided into the transabdominal approach subgroup and the laparoscopic approach subgroup. The hemoglobin, visual analog score (VAS) score, menstruation score, and other indices of each patient before and after treatment were observed, recorded, and analyzed. Results: All 64 patients underwent the operation successfully. After the completion of sequential treatment, the CA125 decreased significantly 1 month after the operation, the average uterine volume significantly reduced, the hemoglobin value increased to a certain extent 3 months after the operation, and the menstrual score and dysmenorrhea during the first menstruation were significantly lower than they were before the operation. After the treatment, the therapeutic results of the transabdominal approach subgroup and endoscopic approach subgroup were compared on the basis of the observed indices, and no significant difference was observed (P > 0.05). Only one patient had a downward movement of the LNG-IUS, and the vaginal ultrasound showed that the upper end of the LNG-IUS was approximately 1.5 cm from the bottom of the uterine cavity. The average follow-up period was 24.02 ± 11.77 months, and no lesion progression was found in any patients. Conclusion: For patients suffering from severe adenomyosis who have no pregnancy plans and require uterine preservation, transabdominal or laparoscopic subtotal resection of the focus of adenomyosis, combined with the LNG-IUS + GnRH-a sequential treatment, may be a safe and effective alternative when conservative treatments such as drugs fail.

Fusing Higher-Order Features in Graph Neural Networks for Skeleton-Based Action Recognition.

Skeleton sequences are lightweight and compact and thus are ideal candidates for action recognition on edge devices. Recent skeleton-based action recognition methods extract features from 3-D joint coordinates as spatial-temporal cues, using these representations in a graph neural network for feature fusion to boost recognition performance. The use of first-and second-order features, that is, joint and bone representations, has led to high accuracy. Nonetheless, many models are still confused by actions that have similar motion trajectories. To address these issues, we propose fusing higher-order features in the form of angular encoding (AGE) into modern architectures to robustly capture the relationships between joints and body parts. This simple fusion with popular spatial-temporal graph neural networks achieves new state-of-the-art accuracy in two large benchmarks, including NTU60 and NTU120, while employing fewer parameters and reduced run time. Our source code is publicly available at: https://github.com/ZhenyueQin/Angular-Skeleton-Encoding.

A preliminary clinical report of transvaginal natural orifice transluminal endoscopic Sacrospinous Ligament Fixation in the treatment of moderate and severe pelvic organ prolapse.

Objective: To study the efficacy and safety of transvaginal natural orifice transluminal endoscopic Sacrospinous Ligament Fixation in the treatment of moderate and severe pelvic organ prolapse. Design: Patients were selected into this study on a voluntary basis to evaluate the short-term efficacy of this surgery by comparing the OP-Q scores before the operation, three months after the operation, and six months after the operation. Setting and Patients: Evaluate the clinical efficacy and safety by a retrospective analysis of the clinical data of the 18 patients with POP-Q grade III-IV pelvic organ prolapse treated by the Department of Gynecology of Nanjing Medical University Affiliated Changzhou No.2 People's Hospital from April 2020 to November 2020, and their post-operation follow-ups. Interventions: Patients with postoperative follow-ups found no obvious relapse without intervention measures. Measurements and Main Results: The transvaginal natural orifice transluminal endoscopic Sacrospinous Ligament Fixation was performed successfully, and the anterior and posterior walls of vagina and/or trans-vaginal hysterectomy were repaired as appropriate. Except the total vaginal length (TVL), the P values of numerical analysis for all points before, three months after, and six months after the operation were all <0.05, being statistically significant. Conclusion: This method is effective in the treatment of moderate and severe pelvic organ prolapse with few complications, but more cases and longer-term follow-up data are needed to determine the long-term effect of this procedure. For the selection of puncture sites, more anatomical data are needed to get more accurate result.

Preliminary Analysis of Safety and Feasibility of a Single-Hole Laparoscopic Myomectomy via an Abdominal Scar Approach.

Purpose: This paper aims to explore the safety and feasibility of a single-hole laparoscopic myomectomy through an abdominal scar approach. Method: The clinical data of seven patients who underwent the single-hole laparoscopic myomectomy via the abdominal scar approach from January to November 2021 in the Department of Gynecology, the Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University, were studied retrospectively. The duration of operation, the intraoperative blood loss, the decrease of postoperative hemoglobin, and the postoperative visual analogue score (0 points: no pain, 10 points: maximum pain) were recorded. Results: All seven patients received the operation successfully, without changing to the conventional laparoscopic operation or open appendectomy. The average blood loss was 101.42 ± 7.89 ml, the average length of hospital stay was 5 ± 0.53 days, the average operation duration was 130 ± 26.86 min, and the 24-h pain score was 1.57 ± 0.53. The seven patients had no intraoperative or postoperative complications and no damage to the ureter or bladder. All patients could urinate spontaneously without urinary retention or urinary tract infection after catheter removal. No analgesic drugs were used after the operation. Conclusion: The single-hole laparoscopic myomectomy via the abdominal scar approach is a more aesthetic and feasible option for eligible patients, but more cases and studies are needed for further confirmation.

A Preliminary Study on the Effects of Black Cohosh Preparations on Bone Metabolism of Rat Models With GnRH-a-Induced Peri-Menopausal Symptoms.

Background: Endometriosis (EMS) is a relapsing and estrogen-dependent disease. For endometriosis such as deep endometriosis and ovarian endometrioid cysts, surgery is the most effective treatment. Long-term follow-up showed that the recurrence rate of endometriosis after surgical treatment was high, so postoperative drugs were needed to reduce recurrence, and Gonadotropin-releasing hormone agonists (GnRH-a) were the most commonly used drug for postoperative management.GnRH-a may reduce the post-treatment endometriosis relapses by lowering the hormone levels in the body. However, the use of GnRH-a can give rise to perimenopausal symptoms, especially osteoporosis, bone loss, and bone pain, for which reason GnRH-a use is often limited. The add-back therapy is often used to alleviate the untoward effects caused by GnRH-a. However, long-term use of hormone drugs may lead to EMS recurrence, thrombosis, and breast cancer. Therefore, a safer and more effective drug is urgently needed to alleviate the untoward effects caused by GnRH-a. In recent years, scholars at home and abroad have found that isopropanolic Cimicifuga racemosa extract (ICR), as a plant extract, can better relieve the symptoms of perimenopausal women. At the same time, some studies have initially confirmed that black cohosh preparations can relieve the perimenopausal symptoms caused by GnRH-a treatment in EMS patients. Objective: To investigate the effect of black cohosh preparations on the bone metabolism of rat models with GnRH-a-induced perimenopausal symptoms. Methods: The rat models of perimenopausal symptoms were established by GnRH-a injection. and normal saline (NS injection) was used as the control. According to the modeling method and drug intervention, the rats were randomly divided into four groups: GnRH-a injection + saline intervention group (GnRH-a + NS), saline injection control + saline intervention group (NS + NS), GnRH-a injection + estradiol intervention group (GnRH-a + E2), and GnRH-a injection + black cohosh preparation intervention group (GnRH-a + ICR). The rat models were identified with the vaginal smear method, and then the corresponding drug intervention was administrated for 28 days. After the intervention, the rats were sacrificed. The rats' bone mineral density (BMD) of the distal femur was detected by a dual-energy X-ray bone density scanner. Rat tibia bone tissues were decalcified and made into slices. The pathological and morphological changes of rat tibial bones in each group were observed through HE staining. Histomorphometry parameters of rat tibial bones in each group, such as trabecular bone volume (TBV), trabecular thickness (TbTh), trabecular number (TbN), and trabecular spacing (TbSp), were detected and analyzed by using an automatic image analysis system. Results: (1) The BMD level of the distal femur in the GnRH-a + NS group was significantly lower than the NS + NS, GnRH-a + E2, and GnRH-a + ICR groups (P<0.01), the BMD levels in GnRH-a + E2 and GnRH-a + ICR groups were slightly lower than the NS + NS group, but there was no significant difference among the three groups (P>0.05). (2) The pathological changes of the tibia bones under the microscope in different groups were as follows: The tibia bone trabecular structure was normal in the NS + NS group, without trabecular thinning or fracture, and the arch structure was normal. In the GnRH-a + NS group, some trabecular structures tapered, the arch structure disappeared, but no obvious bone fracture was observed in the trabecula. In the GnRH-a + E2 and GnRH-a + ICR groups, the trabecular structures were normal, without trabecular bone thinning or fracture, and the arch structures were normal. (3) The TBV level of the GnRH-a + INS group was significantly lower than that of the NS + NS, GnRH-a + E2 and GnRH-a + ICR groups (P<0.01, P<0.05, P<0.01), while there was no significant difference among NS + NS, GnRH-a + E2 and GnRH-a + ICR groups (P>0.05). (4) The TbTh levels in the four groups had no significant difference (P>0.05). Compared with the NS + NS group, the TbTh levels in the GnRH-a + NS, GnRH-a + E2, and GnRH-a + ICR groups showed a descending tendency, while the TbTh levels in the GnRH-a + E2 and GnRH-a + ICR groups were slightly higher than that of the GnRH-a + NS group. However, such differences were not significant statistically (P>0.05). (5) Compared with the NS + NS group, the TbN levels in the GnRH-a + NS, GnRH-a + E2, and GnRH-a + ICR groups decreased remarkably (P<0.05). Compared with the GnRH-a + NS group, the TbN levels in the GnRH-a + E2 and GnRH-a + ICR groups showed a mild descending tendency, but such differences were not significant statistically (P>0.05). (6) The TbSp level of the GnRH-a + NS group was significantly higher than that of the NS + NS, GnRH-a + E2, and GnRH-a + ICR groups (P<0.01), while there was no significant difference among NS + NS, GnRH-a + E2 and GnRH-a + ICR groups (P>0.05). Conclusion: The GnRH-a injection could achieve the desired effect. GnRH-a injection may lead to the loss of bone mass in rats. Black cohosh preparations, like estrogen, may have a protective effect on bone mass loss caused by GnRH-a injection.

Safety and Feasibility of Vaginal Delivery in Full-Term Pregnancy After Transvaginal-Natural Orifice Transluminal Endoscopic Surgery: A Case Series.

Study Objective: The aim was to investigate the outcome of vaginal delivery of full-term pregnancies in patients after transvaginal-natural orifice transluminal endoscopic surgery (vNOTES) treatment for gynecological disorders. Design: A case series report. Setting: A medical university hospital. Patients: 12 cases of successful delivery after transvaginal-natural orifice transluminal endoscopic surgery. Interventions: Long-term follow-up of patients with fertility needs after transvaginal-natural orifice transluminal endoscopic surgery. Measurements and Main Results: From 2018 to 2021, 163 cases of gynecological diseases were treated by vNOTES. One hundred forty-seven patients were followed up, with a follow-up rate of 90.1%. The average follow-up time was 28 (15-47) months, including 66 cases with fertility requirements. Among these 66 patients, 12 patients successfully got pregnant and completed delivery, including 10 cases of vaginal delivery and 2 cases of cesarean section, with no adverse pregnancy outcomes associated with vNOTES arising. Conclusion: Vaginal delivery of a full-term pregnancy after transvaginal-natural orifice transluminal endoscopic surgery appears to be safe and feasible and would not be one of the bases for elective cesarean delivery.

Resolving Anomalies in the Behaviour of a Modularity-Inducing Problem Domain with Distributional Fitness Evaluation.

Discrete gene regulatory networks (GRNs) play a vital role in the study of robustness and modularity. A common method of evaluating the robustness of GRNs is to measure their ability to regulate a set of perturbed gene activation patterns back to their unperturbed forms. Usually, perturbations are obtained by collecting random samples produced by a predefined distribution of gene activation patterns. This sampling method introduces stochasticity, in turn inducing dynamicity. This dynamicity is imposed on top of an already complex fitness landscape. So where sampling is used, it is important to understand which effects arise from the structure of the fitness landscape, and which arise from the dynamicity imposed on it. Stochasticity of the fitness function also causes difficulties in reproducibility and in post-experimental analyses. We develop a deterministic distributional fitness evaluation by considering the complete distribution of gene activity patterns, so as to avoid stochasticity in fitness assessment. This fitness evaluation facilitates repeatability. Its determinism permits us to ascertain theoretical bounds on the fitness, and thus to identify whether the algorithm has reached a global optimum. It enables us to differentiate the effects of the problem domain from those of the noisy fitness evaluation, and thus to resolve two remaining anomalies in the behaviour of the problem domain of Espinosa-Soto and A. Wagner (2010). We also reveal some properties of solution GRNs that lead them to be robust and modular, leading to a deeper understanding of the nature of the problem domain. We conclude by discussing potential directions toward simulating and understanding the emergence of modularity in larger, more complex domains, which is key both to generating more useful modular solutions, and to understanding the ubiquity of modularity in biological systems.

Mature cystic extragonadal teratoma in Douglas' pouch: Case report and literature review.

Teratomas often occur in the gonads, while Extragonadal mature cystic teratomas are reported occasionally, with the most common site being the omentum. Teratoma in the Douglas sac is extremely rare. we report a rare case of mature cystic Teratoma in the Douglas sac in a 71-year-old woman who underwent laparoscopic surgery. A cyst with a diameter of approximately 6 cm from Douglas was found during surgery, and the mass was separated from both ovaries. Microscopically, the cyst was a mature cystic teratoma that did not originate from the ovary.

GnRH-a-Induced Perimenopausal Rat Modeling and Black Cohosh Preparations' Effect on Rat's Reproductive Endocrine.

Background: Endometriosis (EMS) is an estrogen-dependent disease, which easily recurs after operation. Gonadotropin-releasing hormone agonist (GnRH-a), an estrogen-inhibiting drug, can effectively inhibit the secretion of gonadotropin by pituitary gland, so as to significantly decrease the ovarian hormone level and facilitate the atrophy of ectopic endometrium, playing a positive role in preventing postoperative recurrence. The application of GnRH-a can lead to the secondary low estrogen symptoms, namely the perimenopausal symptoms, and is a main reason for patients to give up further treatment. The add-back therapy based on sex hormones can well address the perimenopausal symptoms, but long-term use of hormones may cause the recurrence of EMS, as well as liver function damage, venous embolism, breast cancer and other risks, which has long been a heated topic in the industry. Therefore, it is necessary to find effective and safe anti-additive drugs soon. Studies at home and abroad show that, as a plant extract, isopropanolic extract of cimicifuga racemosa (ICR) can well relieve the perimenopausal symptoms caused by natural menopause. Some studies have preliminarily confirmed that black cohosh preparations can antagonize perimenopausal symptoms of EMS patients treated with GnRH-a after operation. Objective: To establish a rat model of perimenopausal symptoms induced by GnRH-a injection, for the purposes of laying a foundation for further research and preliminarily exploring the effect of black cohosh preparations on reproductive endocrine of the rat model. Method: The rat model of perimenopausal symptoms was established by GnRH-a injection, and normal saline (NS injection) was used as the control. The rats were randomly divided into four groups according to different modeling methods and drug intervention schemes. GnRH-a injection + normal saline intervention group (GnRH-a + NS), normal saline injection control + normal saline intervention group (NS + NS), GnRH-a injection + estradiol intervention group (GnRH-a + E2), and GnRH-a injection + black cohosh preparations intervention group (GnRH-a + ICR). After modelling was assessed to be successful with the vaginal smear method, the corresponding drugs were given for intervention for 28d. In the process of rat modeling and drug intervention, the skin temperature and anus temperature of the rat tails were measured every other day, the body weights of the rats were measured every other day, and the dosage was adjusted according to the body weight. After the intervention was over, the serum sex hormone level, the uterine weight, the uterine index, and the endometrial histomorphology changes, as well as the ovarian weight, the ovarian index, and the morphological changes of ovarian tissues of each group were measured. Results: (1) The vaginal cell smears of the control group (NS + NS) showed estrous cycle changes, while other model rats had no estrous cycle of vaginal cells. (2) The body weight gains of the GnRH-a + NS, GnRH-a + E2 and GnRH-a + ICR groups were significantly higher than that of the NS + NS control group. The intervention with E2 and ICR could delay the weight gain trend of rats induced by GnRH-A. (3) After GnRH-a injection, the temperature of the tail and anus of rats showed an overall upward trend, and the intervention with E2 and ICR could effectively improve such temperature change. (4) The E2, FSH, and LH levels in the GnRH-a + NS, GnRH-a + E2, and GnRH-a + ICR groups were significantly lower than those in the NS + NS group (P < 0.01). The E2 level was significantly higher and the LH level was significantly lower in the GnRH-a + E2 group than those in the GnRH-a + NS and GnRH-a + ICR groups (P < 0.05). Compared with those of the GnRH-a + NS and GnRH-a + ICR groups, the FSH level of the GnRH-a + E2 group showed a slight downward trend, but the difference was not statistically significant (P > 0.05). There was no significant difference in the levels of sex hormones between the GnRH-a + NS group and GnRH-a + ICR group (P > 0.05). (5) Compared with those of the NS + NS group, the uterine weight and uterine index of the GnRH-a + NS, GnRH-a + E2 and GnRH-a + ICR groups significantly decreased (P < 0.01). In a comparison between the groups, the uterine weight and uterine index in the GnRH-a + NS and GnRH-a + ICR groups were significantly lower than those in the GnRH-a + E2 group (P < 0.01). There was a statistical difference in the uterine weight and uterine index between the GnRH-a + NS group and GnRH-a + ICR group (P > 0.05). (6) Compared with those of the NS + NS group, the ovarian weight and ovarian index of the GnRH-a + NS, GnRH-a + E2 and GnRH-a + ICR groups significantly decreased (P < 0.01). There was no statistical difference in the ovarian weight and ovarian index among the GnRH-a + E2, GnRH-a + NS and GnRH-a + ICR groups (P > 0.05). (7) Compared with those in the NS + NS group, the number of primordial follicles increased significantly, while the number of growing follicles and mature follicles decreased significantly in the GnRH-a + NS, GnRH-a + E2, and GnRH-a + ICR groups (P < 0.01), but there was a statistical difference in the total number of follicles among the four groups (P > 0.05). Conclusions: The GnRH-a injection could achieve the desired effect. The animal model successfully achieved a significant decrease in the E2, FSH, and LH levels in rats, and could cause the rats to have rising body surface temperature similar to hot flashes in the perimenopausal period. The intervention with E2 and ICR could effectively relieve such "perimenopausal symptoms", and ICR had no obvious effect on the serum sex hormone level in rats.