-
Diagnosis of placenta accreta spectrum using ultrasound texture feature fusion and machine learning.Computers in Biology and Medicine Jun 2024Placenta accreta spectrum (PAS) is an obstetric disorder arising from the abnormal adherence of the placenta to the uterine wall, often leading to life-threatening...
INTRODUCTION
Placenta accreta spectrum (PAS) is an obstetric disorder arising from the abnormal adherence of the placenta to the uterine wall, often leading to life-threatening complications including postpartum hemorrhage. Despite its significance, PAS remains frequently underdiagnosed before delivery. This study delves into the realm of machine learning to enhance the precision of PAS classification. We introduce two distinct models for PAS classification employing ultrasound texture features.
METHODS
The first model leverages machine learning techniques, harnessing texture features extracted from ultrasound scans. The second model adopts a linear classifier, utilizing integrated features derived from 'weighted z-scores'. A novel aspect of our approach is the amalgamation of classical machine learning and statistical-based methods for feature selection. This, coupled with a more transparent classification model based on quantitative image features, results in superior performance compared to conventional machine learning approaches.
RESULTS
Our linear classifier and machine learning models attain test accuracies of 87 % and 92 %, and 5-fold cross validation accuracies of 88.7 (4.4) and 83.0 (5.0), respectively.
CONCLUSIONS
The proposed models illustrate the effectiveness of practical and robust tools for enhanced PAS detection, offering non-invasive and computationally-efficient diagnostic tools. As adjunct methods for prenatal diagnosis, these tools can assist clinicians by reducing the need for unnecessary interventions and enabling earlier planning of management strategies for delivery.
PubMed: 38878399
DOI: 10.1016/j.compbiomed.2024.108757 -
European Radiology Experimental Sep 2023Placenta accreta spectrum (PAS) is a rare, life-threatening complication of pregnancy. Predicting PAS severity is critical to individualise care planning for the birth....
BACKGROUND
Placenta accreta spectrum (PAS) is a rare, life-threatening complication of pregnancy. Predicting PAS severity is critical to individualise care planning for the birth. We aim to explore whether radiomic analysis of T2-weighted magnetic resonance imaging (MRI) can predict severe cases by distinguishing between histopathological subtypes antenatally.
METHODS
This was a bi-centre retrospective analysis of a prospective cohort study conducted between 2018 and 2022. Women who underwent MRI during pregnancy and had histological confirmation of PAS were included. Radiomic features were extracted from T2-weighted images. Univariate regression and multivariate analyses were performed to build predictive models to differentiate between non-invasive (International Federation of Gynecology and Obstetrics [FIGO] grade 1 or 2) and invasive (FIGO grade 3) PAS using R software. Prediction performance was assessed based on several metrics including sensitivity, specificity, accuracy and area under the curve (AUC) at receiver operating characteristic analysis.
RESULTS
Forty-one women met the inclusion criteria. At univariate analysis, 0.64 sensitivity (95% confidence interval [CI] 0.0-1.00), specificity 0.93 (0.38-1.0), 0.58 accuracy (0.37-0.78) and 0.77 AUC (0.56-.097) was achieved for predicting severe FIGO grade 3 PAS. Using a multivariate approach, a support vector machine model yielded 0.30 sensitivity (95% CI 0.18-1.0]), 0.74 specificity (0.38-1.00), 0.58 accuracy (0.40-0.82), and 0.53 AUC (0.40-0.85).
CONCLUSION
Our results demonstrate a predictive potential of this machine learning pipeline for classifying severe PAS cases.
RELEVANCE STATEMENT
This study demonstrates the potential use of radiomics from MR images to identify severe cases of placenta accreta spectrum antenatally.
KEY POINTS
• Identifying severe cases of placenta accreta spectrum from imaging is challenging. • We present a methodological approach for radiomics-based prediction of placenta accreta. • We report certain radiomic features are able to predict severe PAS subtypes. • Identifying severe PAS subtypes ensures safe and individualised care planning for birth.
Topics: Pregnancy; Humans; Female; Placenta Accreta; Prospective Studies; Retrospective Studies; Machine Learning; Research Design
PubMed: 37726591
DOI: 10.1186/s41747-023-00369-2 -
Acta Obstetricia Et Gynecologica... Jul 2023This study examined obstetric outcomes in patients diagnosed with uterine adenomyosis.
INTRODUCTION
This study examined obstetric outcomes in patients diagnosed with uterine adenomyosis.
MATERIAL AND METHODS
This historical cohort study queried the Healthcare Cost and Utilization Project's National Inpatient Sample. The study population was all hospital deliveries in women aged 15-54 years between January 2016 and December 2019. The exposure was a diagnosis of uterine adenomyosis. The main outcome measures were obstetric characteristics, including placenta previa, placenta accreta spectrum, and placental abruption. Secondary outcomes were delivery complications including severe maternal morbidity. Analytic steps to assess these outcomes included (i) a 1-to-N propensity score matching to mitigate and balance prepregnancy confounders to assess obstetric characteristics, followed by (ii) an adjusting model with preselected pregnancy and delivery factors to assess maternal morbidity. Sensitivity analyses were also performed with restricted cohorts to account for prior uterine scar, uterine myoma, and extra-uterine endometriosis.
RESULTS
After propensity score matching, 5430 patients with adenomyosis were compared to 21 720 patients without adenomyosis. Adenomyosis was associated with an increased odds of placenta accreta spectrum (adjusted-odds ratio [aOR] 3.07, 95% confidence interval [CI] 2.01-4.70), placenta abruption (aOR 3.21, 95% CI: 2.60-3.98), and placenta previa (aOR 5.08, 95% CI: 4.25-6.06). Delivery at <32 weeks of gestation (aOR 1.48, 95% CI: 1.24-1.77) and cesarean delivery (aOR 7.72, 95% CI: 7.04-8.47) were both increased in women with adenomyosis. Patients in the adenomyosis group were more likely to experience severe maternal morbidity at delivery compared to those in the nonadenomyosis group (aOR 1.86, 95% CI: 1.59-2.16). Results remained robust in the aforementioned several sensitivity analyses.
CONCLUSIONS
This national-level analysis suggests that a diagnosis of uterine adenomyosis is associated with an increased risk of placental pathology (placenta accreta spectrum, placenta abruption, and placental previa) and adverse maternal outcomes at delivery.
Topics: Pregnancy; Humans; Female; Placenta Previa; Placenta; Placenta Accreta; Cohort Studies; Risk Factors; Adenomyosis; Propensity Score; Abruptio Placentae; Retrospective Studies
PubMed: 37087741
DOI: 10.1111/aogs.14581 -
The Journal of Maternal-fetal &... Dec 2023Describe the clinical-surgical results of patients with PAS in the low-posterior cervical-trigonal space associated with fibrosis (PAS type 4) compared with PAS types in...
OBJECTIVE
Describe the clinical-surgical results of patients with PAS in the low-posterior cervical-trigonal space associated with fibrosis (PAS type 4) compared with PAS types in other locations (Types 1, upper bladder, 2 in upper parametrium) and in particular with PAS type 3, corresponding to dissectible cervical-trigonal invasion. The clinical-surgical results of using a standard hysterectomy were analyzed with a modified subtotal hysterectomy (MSTH) in patients with PAS type 4.
MATERIAL AND METHODS
A descriptive, retrospective, multicenter study included 337 patients of PAS; thirty-two corresponding to PAS type 4, from three PAS reference hospitals, CEMIC, Buenos Aires, Argentina, Fundación Valle de Lili, Cali, Colombia, and Dr. Soetomo General Hospital, Surabaya, Indonesia, between January 2015 and December 2020. PAS was diagnosed by abdominal and transvaginal ultrasound and topographically characterized by ultrafast T2 weighted MRI. In persistent macroscopic hematuria after MSTH, the surgeon performs an intentional cystotomy and uses a square compression suture to achieve the hemostasis inside the bladder wall.According to a PAS topographical classification, the patients with low-vesical cervical involvement compared with PAS located in relation with the upper blader (type1), upper parametrium (type 2 upper), and also with PAS situated in the lower vesical-trigon space (type 3). PAS 3 and 4 are located in identical area, but in type 3, group A, the vesicouterine space was dissectible, and in type 4, group B, significant fibrosis made surgical dissection extremely challenging. Furthermore, group B was divided into patients treated with total hysterectomy (HT) and those treated with a modified subtotal hysterectomy (MSTH). The surgical requirements to perform an MSHT included the availability of proximal vascular control at the aortic level (internal manual aortic compression, aortic endovascular balloon, aortic loop, or aortic cross-clamping). Then surgeon performed an upper segmental hysterotomy, avoiding the abnormal placenta invasion area; after that, the fetus was delivered, and the umbilical cord was ligated.After uterine exteriorization, the surgeon applies a continuous circular suture with number 2 polyglactin 910, taking some portions of the myometrium -to avoid unintentional slipping- around the lower uterine segment and a 3-4 cm proximal to the abnormal adhesion of the placenta. After tightening hard the circular suture, the uterine segment was circumferentially cut, three centimeters proximal to the circular hemostatic sutures. Next, the surgery follows the upper steps of conventional hysterectomy without changes. Additionally, the histological presence of fibrosis was examined in all samples.
RESULTS
Modified subtotal hysterectomy in patients with PAS type 4 (cervical-trigonal fibrosis) resulted in a significant clínico-surgical improvement over total hysterectomy. The median operative time and intraoperative bleeding were 140 min (IQR 90--240) and 1895 mL (IQR 1300-2500) in patients undergoing modified subtotal hysterectomy, and 260 min (IQR 210-287) and 2900 mL (IQR 2150-5500) in patients treated with total hysterectomy, respectively. The complication rate was 20% for MSHT and 82.3% for patients with a total hysterectomy.
CONCLUSIONS
PAS in the cervical trigonal area associated with fibrosis implies a greater risk of complications due to uncontrollable bleeding and organ damage. MSTH is associated with lower morbidity and difficulties in PAS type 4. Prenatal or intrasurgical diagnosis is essential to plan surgical alternatives to improve the results.
Topics: Pregnancy; Female; Humans; Placenta Accreta; Retrospective Studies; Uterus; Hysterectomy; Morbidity; Fibrosis; Placenta
PubMed: 37193605
DOI: 10.1080/14767058.2023.2183741 -
Scientific Reports Sep 2023The purpose of this study was to implore the association among clinical features, long-term fertility outcomes and the anatomical location of adenomyosis identified by...
The purpose of this study was to implore the association among clinical features, long-term fertility outcomes and the anatomical location of adenomyosis identified by ultrasound. We collected data of non-pregnant patients between 20 and 40 years old who had undergone surgical exploration for benign gynecological conditions at our institution between January 2010 and December 2017. A total of 158 women met the inclusion criteria and were allocated into three groups according to the ultrasound-determined adenomyosis anatomical location: anterior (Group A), posterior (Group B), both posterior and anterior (Group C). 44.3% (70/158) adenomyosis was located at the posterior side. History of miscarriage and parity were significantly higher in Group C (p = 0.036 and 0.001 respectively). Group C also had a higher concurrence rate of ovarian endometrioma (OEM) (80.4%, p = 0.002), pelvic adhesion (80.4%, P = 0.003) and the revised American Fertility Society (rAFS) Score (median64, range2-100, P < 0.001), while a significantly lower rate of concurrent peritoneal endometriosis (P = 0.01). Group B showed a relative higher rate of coexistent heavy menstrual bleeding (28.6%, p = 0.04) and oviduct obstruction (24.3%, P = 0.038). Group A had a higher proportion of coexistent leiomyoma (53.1%, P = 0.002). There were no significant differences between group A, B, and C in terms of pain symptoms, endometrial polyps, operation time, and endometriosis fertility index score and other basic characters (p > 0.05). During the follow-up, 59.2% (61/103) patients had clinical pregnancies, and 26.2% (16/61) of them experienced pregnancy loss. Total in vitro fertilization and embryo transfer pregnancy rate was 64.6% (42/65) and spontaneous pregnancy rate was 50.0% (19/38). The Kaplan-Meier curves demonstrated significant lower cumulative pregnancy rate in Group C than Group A and Group B (p = 0.01). Severe obstetric complications such as placenta previa, placenta accreta, preeclampsia, and preterm birth were only found in women with adenomyosis located in the posterior side. In conclusion, types of adenomyosis based on sonographic location had different clinical features and pregnancy outcome. Patients with adenomyosis lesion in both anterior and posterior sides had higher combination of OEM, pelvic adhesion and rAFS score.
Topics: Infant, Newborn; Pregnancy; Humans; Female; Young Adult; Adult; Pregnancy Outcome; Endometriosis; Adenomyosis; Premature Birth; Abortion, Spontaneous
PubMed: 37679426
DOI: 10.1038/s41598-023-40816-z -
AJOG Global Reports Aug 2023Management options for placenta accreta spectrum disorder are multiple, without a clear picture of which one is superior. Management guidelines describe the use of a...
BACKGROUND
Management options for placenta accreta spectrum disorder are multiple, without a clear picture of which one is superior. Management guidelines describe the use of a wide range of human and technological resources that are not always available in resource-limited settings.
OBJECTIVE
This consensus seeks agreement on general guidelines that facilitate the management of placenta accreta spectrum in low- and middle-income countries.
STUDY DESIGN
Consensus was developed using the modified Delphi methodology, incorporating 3 successive rounds in which 6 dimensions of placenta accreta spectrum treatment were discussed: pathway for placenta accreta spectrum care, roles at different levels of care, organization of the interdisciplinary teams at the reference hospitals, training interdisciplinary teams, placenta accreta spectrum surgical treatment, and management of placenta accreta spectrum patients without prenatal diagnosis.
RESULTS
Consensus was achieved on all questions on placenta accreta spectrum management. Specific low- and middle-income countries problems were addressed, trying to establish guidelines for the construction of trained placenta accreta spectrum interdisciplinary teams, as well as the rational use of the different therapeutic options available in a limited resources setting. In addition, it is highlighted the need to facilitate contact between patients affected by this disease and the interdisciplinary groups, overcoming administrative barriers typical of some health systems.
CONCLUSION
We present a consensus on the treatment of placenta accreta spectrum in a low- and middle-income countries, based on local experts' opinions. Construction of high-quality scientific evidence is essential in settings with limited resources.
PubMed: 37456142
DOI: 10.1016/j.xagr.2023.100188 -
BMC Pregnancy and Childbirth Jul 2023Placenta accreta spectrum (PAS) disorder is a major cause of postpartum hemorrhage-associated maternal and fetal death, and novel methods for PAS screening are urgently...
BACKGROUND
Placenta accreta spectrum (PAS) disorder is a major cause of postpartum hemorrhage-associated maternal and fetal death, and novel methods for PAS screening are urgently needed for clinical application.
METHODS
The purpose of this study was to develop new methods for PAS screening using serum biomarkers and clinical indicators. A total of 95 PAS cases and 137 controls were enrolled in a case-control study as cohort one, and 44 PAS cases and 35 controls in a prospective nested case-control study were enrolled as cohort two. All subjects were pregnant women of Chinese Han population. Biomarkers for PAS from maternal blood samples were screened based on high-throughput immunoassay and were further validated in three phases of cohort one. Screening models for PAS were generated using maternal serum biomarkers and clinical indicators, and were validated in two cohorts. The expression levels of biomarkers were analyzed using histopathological and immunohistochemical (IHC) techniques, and gene expression was examined by QPCR in the human placenta. Binary logistic regression models were built, and the area under the curve (AUC), sensitivity, specificity, and Youden index were calculated. Statistical analyses and model building were performed in SPSS and graphs were generated in GraphPad Prism. The independent-sample t test was used to compare numerical data between two groups. For nonparametric variables, a Mann-Whitney U test or a X test was used.
RESULTS
The results demonstrated that the serum levels of matrix metalloproteinase-1 (MMP-1), epidermal growth factor (EGF), and vascular endothelial growth factor-A (VEGF-A) were consistently higher, while the level of tissue-type plasminogen activator (tPA) was significantly lower in PAS patients compared with normal term controls and patients with pre-eclampsia (PE) and placenta previa (PP). IHC and QPCR analysis confirmed that the expression of the identified biomarkers significantly changed during the third trimester in human placenta. The generated screening model combining serum biomarkers and clinical indicators detected 87% of PAS cases with AUC of 0.94.
CONCLUSIONS
Serum biomarkers can be used for PAS screening with low expense and high clinical performance; therefore, it may help to develop a practicable method for clinical prenatal PAS screening.
Topics: Female; Humans; Pregnancy; Biomarkers; Case-Control Studies; Placenta Accreta; Prospective Studies; Vascular Endothelial Growth Factor A
PubMed: 37434137
DOI: 10.1186/s12884-023-05784-2 -
Korean Journal of Women Health Nursing Sep 2023This scoping review investigated the content and quality of YouTube videos on women's health. (Review)
Review
PURPOSE
This scoping review investigated the content and quality of YouTube videos on women's health.
METHODS
A literature search of the Cochrane Library, PubMed, Embase, CINAHL, ERIC, and RISS databases was performed using the keywords "('youtube'/exp OR youtube OR 'social media'/ exp OR 'social media' OR (('social'/exp OR social) AND ('media'/exp OR media))) AND ('female health care' OR (('female'/exp OR female) AND ('health'/exp OR health) AND ('care'/exp OR care)))" from February 21 to 27, 2023. Peer-reviewed analytic studies in English or Korean that focused on women's health using YouTube were included.
RESULTS
The review identified 21 articles that covered various themes related to women's health, such as breast cancer, urinary disease, sexual health, pelvic organ prolapse, the human papillomavirus vaccine, Papanikolaou smears, contraception, women's health information during the coronavirus disease 2019 pandemic, obstetric epidural anesthesia, and placenta accreta. However, the overall quality of the content was low, inaccurate, unreliable, and misleading.
CONCLUSION
This scoping review demonstrated that YouTube videos on women's health covered diverse topics, but the quality of the content needed improvement. More reliable and high-quality videos produced by academic institutes and healthcare professionals specializing in women's health are needed for social media to be usable as a reliable source of women's health information. The high number of views and shares received by the videos underscores the importance of providing accurate and reliable information on women's health.
Topics: Humans; Female; Social Media; COVID-19; Videotape Recording; Women's Health; Breast Neoplasms
PubMed: 37813661
DOI: 10.4069/kjwhn.2023.08.19 -
Journal of Global Health Dec 2023Although maternal age might affect pregnancy outcomes, it remains unclear whether this relationship is linear or curvilinear and if it differs between nulliparous and...
BACKGROUND
Although maternal age might affect pregnancy outcomes, it remains unclear whether this relationship is linear or curvilinear and if it differs between nulliparous and multiparous women. We aimed to characterize the relationship between maternal age and risks of pregnancy outcomes in a diverse sample of Chinese singleton pregnant women and to evaluate whether the relationship varied by parity.
METHODS
We based this prospective multicenter cohort study on data from 18 495 singleton pregnant women who participated in the University Hospital Advanced Age Pregnant Cohort Study, conducted in eight Chinese public hospitals from 2016 to 2021. We used restricted cubic splines to model nonlinear relationships between maternal age continuum and adverse outcomes, and performed multivariable log-binomial regression to estimate the adjusted relative risk (RR) and 95% confidence interval (CI).
RESULTS
Among 18 495 singleton pregnant women (mean age 35.7, standard deviation (SD) = 4.2 years), maternal age was not related to postpartum hemorrhage or small for gestational age, but showed a positive, nonlinear relationship to gestational diabetes mellitus, hypertensive disorders of pregnancy, preeclampsia, placenta accreta spectrum, placenta previa, cesarean delivery, preterm birth, large for gestational age, macrosomia, and fetal congenital anomaly, with inflection points around 35.6-40.4 years. Compared to women younger than 35 years, older women had higher risks of adverse pregnancy outcomes, except for postpartum hemorrhage and small for gestational age. The risks of placenta accreta spectrum, placenta previa, large for gestational age, and macrosomia were highest for women aged 40-44 years, and risks of gestational diabetes mellitus, hypertensive disorders of pregnancy, preeclampsia, cesarean delivery, preterm birth and congenital anomaly were highest for those aged ≥45 years. Most risks were more pronounced in nulliparous than multiparous women (P for interaction <0.02).
CONCLUSIONS
Delayed childbirth was related to increased risks of adverse pregnancy outcomes, especially for nulliparous women. Appropriate childbearing age, generally before 35 years, is recommended for optimising pregnancy outcomes.
Topics: Pregnancy; Female; Infant, Newborn; Humans; Aged; Adult; Maternal Age; Diabetes, Gestational; Premature Birth; Placenta Previa; Pre-Eclampsia; Fetal Macrosomia; Hypertension, Pregnancy-Induced; Postpartum Hemorrhage; Cohort Studies; Placenta Accreta; Prospective Studies; Pregnancy Outcome; Infant, Newborn, Diseases; Retrospective Studies
PubMed: 38038697
DOI: 10.7189/jogh.13.04161