Dataset Information

Radiomics-based prediction of FIGO grade for placenta accreta spectrum.

ABSTRACT:

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.

SUBMITTER: Bartels HC

PROVIDER: S-EPMC10509122 | biostudies-literature | 2023 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Radiomics-based prediction of FIGO grade for placenta accreta spectrum.

Bartels Helena C HC O'Doherty Jim J Wolsztynski Eric E Brophy David P DP MacDermott Roisin R Atallah David D Saliba Souha S Young Constance C Downey Paul P Donnelly Jennifer J Geoghegan Tony T Brennan Donal J DJ Curran Kathleen M KM

European radiology experimental 20230920 1

<h4>Background</h4>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.<h4>Methods</h4>This was a bi-centre retrospective analysis of a prospective cohort study conducted between 2018 and 2022. Women w ...[more]

PMID: 37726591

Similar Datasets

Project description:ObjectivesWe previously demonstrated the potential of radiomics for the prediction of severe histological placenta accreta spectrum (PAS) subtypes using T2-weighted MRI. We aim to validate our model using an additional dataset. Secondly, we explore whether the performance is improved using a new approach to develop a new multivariate radiomics model.MethodsMulti-centre retrospective analysis was conducted between 2018 and 2023. Inclusion criteria: MRI performed for suspicion of PAS from ultrasound, clinical findings of PAS at laparotomy and/or histopathological confirmation. Radiomic features were extracted from T2-weighted MRI. The previous multivariate model was validated. Secondly, a 5-radiomic feature random forest classifier was selected from a randomized feature selection scheme to predict invasive placenta increta PAS cases. Prediction performance was assessed based on several metrics including area under the curve (AUC) of the receiver operating characteristic curve (ROC), sensitivity, and specificity.ResultsWe present 100 women [mean age 34.6 (±3.9) with PAS], 64 of whom had placenta increta. Firstly, we validated the previous multivariate model and found that a support vector machine classifier had a sensitivity of 0.620 (95% CI: 0.068; 1.0), specificity of 0.619 (95% CI: 0.059; 1.0), an AUC of 0.671 (95% CI: 0.440; 0.922), and accuracy of 0.602 (95% CI: 0.353; 0.817) for predicting placenta increta. From the new multivariate model, the best 5-feature subset was selected via the random subset feature selection scheme comprised of 4 radiomic features and 1 clinical variable (number of previous caesareans). This clinical-radiomic model achieved an AUC of 0.713 (95% CI: 0.551; 0.854), accuracy of 0.695 (95% CI 0.563; 0.793), sensitivity of 0.843 (95% CI 0.682; 0.990), and specificity of 0.447 (95% CI 0.167; 0.667).ConclusionWe validated our previous model and present a new multivariate radiomic model for the prediction of severe placenta increta from a well-defined, cohort of PAS cases.Advances in knowledgeRadiomic features demonstrate good predictive potential for identifying placenta increta. This suggests radiomics may be a useful adjunct to clinicians caring for women with this high-risk pregnancy condition.

Project description:BackgroundPlacenta accreta spectrum (PAS) is one of the most dangerous pregnancy-related conditions. This study aims to conduct a systematic review of current research on the ultrasound scoring systems used in PAS patients with a comprehensive summarization of researches and comparison of prenatal ultrasound scoring in evaluating postpartum outcomes.MethodsThis systematic review was conducted and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies were retrieved from four databases (PubMed, Cochrane Library, Embase and Web of Science) up to December 2023. Original studies involving women diagnosed with PAS using ultrasound scoring for diagnosis or outcome evaluation were screened based on predefined inclusion and exclusion criteria. The primary outcome was the diagnostic performance of ultrasound scoring systems and their effectiveness in predicting labor outcomes. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was utilized to assess the study quality. The study has been registered on the PROSPERO website with the registration number CRD42024507311.ResultsSixteen studies met the inclusion criteria. 3930 patients were included in the review, with 1693 participants in the control group. Included studies had a high risk of bias in patient selection, with a low risk in flow and timing and reference standards. Existing studies have reported several indicators that can be included in ultrasound scoring systems and tested their effectiveness in diagnosis and severity evaluation. The pool sensitivity, specificity, and area under the curve (AUC) of ultrasound scoring system in diagnosing PAS were 0.89 (95% confidence interval [CI]: 0.82-0.94), 0.85(95%CI: 0.80-0.90) and 0.93 (95%CI:0.91-0.95). Ultrasound scores are associated with PAS outcomes including intraoperative haemorrhage, postpartum haemorrhage, hysterectomy, length of hospital stay, and neonatal prognosis. Limitation is that the existing studies are wide-ranging but have low replication and association.ConclusionUltrasound scoring systems play a role in the prenatal diagnosis, management, and prediction of postnatal complications in PAS. Nevertheless, additional research is required to further evaluate the performance between different scoring systems to develop a unified consensus.

Project description:Background Uterine incarceration is a rare obstetric complication that is always associated with retroversion and prone to misdiagnosis. Pelvic examination and imaging methods including ultrasound and magnetic resonance imaging (MRI) are used as the primary diagnostic tool. We present an asymptomatic anterior uterine incarceration complicated by placenta previa and placenta accreta spectrum (PAS) disorder, which could be diagnosed during the pregnancy periods, but was first diagnosed during the cesarean section (CS) and got the surgeons into trouble. Case Description A 28-year-old woman, gravidity 4, parity 1, was hospitalized due to placenta previa and PAS disorder diagnosed by ultrasound and MRI at 35.6 weeks of gestation. She had not experienced any discomfort. Given her history of a previous CS, she underwent a well-prepared cesarean delivery for the termination of the pregnancy. The patient had a series of periodical ultrasound and MRI examinations in which placenta previa and placenta accreta were described in disregard of the abnormal location of cervix; consequently, interior uterine incarceration was first diagnosed during the surgery, which caused significant difficulties in the operation which lasted 3 hours and 21 minutes. The patient developed severe hemorrhaging and lost approximately 5,000 mL of blood. Fortunately, she delivered a health male infant weighing 3,440 grams with quite good Apgar scores. During the follow-up, maternal and child health was confirmed. Conclusions With regard to patients who have undergone previous pelvic surgery, doctors need to pay close attention to the position of the cervix, the pelvic adhesion situation during the prenatal examination, and be on alert for uterine incarceration. Uterine incarceration can be accurately recognized by periodic sonography and MRI if the radiologist is aware of this unusual condition. We do suggest that special are must be taken to avoid unnecessary trauma by misdiagnosis because of lack of awareness.

Project description:ImportancePlacenta previa is widely acknowledged as a risk factor for placenta accreta spectrum (PAS) disorders, which are severe maternal complications; however, data are limited regarding whether placenta previa is associated with a higher risk of worse maternal outcomes among patients with PAS disorders.ObjectiveTo examine the association between placenta previa and the risk of severe maternal morbidities (SMMs) and higher resource use among patients with PAS disorders.Design, setting, and participantsThis retrospective cohort study extracted records of 3793 patients with PAS diagnosis and delivery indicators between October 1, 2015, and December 31, 2019, from the US National Inpatient Sample database.ExposuresPlacenta previa.Main outcomes and measuresData on 21 Centers for Disease Control and Prevention-defined SMMs and 25 study-defined surgical morbidities associated with PAS were extracted. Six surgical procedures (cystoscopy, intra-arterial balloon occlusion, cesarean delivery, hysterectomy, cystectomy, and oophorectomy), hospital length of stay, and inpatient costs were compared. Multivariable Poisson regression models built in the generalized estimating equation framework were used.ResultsAmong 3793 patients with PAS (median [IQR] age at admission, 33 [29-37] years), 621 women (16.4%) were Black, 765 (20.2%) were Hispanic, 1779 (46.9%) were White, 441 (11.6%) were of other races and/or ethnicities (47 [1.2%] were American Indian, 220 [5.8%] were Asian or Pacific Islander, and 174 [4.6%] were of multiple or other races and/or ethnicities), and 187 (4.9%) were of unknown race and ethnicity. A total of 1323 patients (34.9%) had placenta previa and 2470 patients (65.1%) did not; of those with placenta previa, 405 patients (30.6%) had invasive PAS. Patients with vs without placenta previa had a significantly higher rate and risk of any SMM (935 women [70.7%] vs 1087 women [44.0%]; P < .001; adjusted risk ratio [aRR], 1.19; 95% CI, 1.12-1.27) and any surgical morbidity (1170 women [88.4%] vs 1667 women [67.5%]; P < .001; aRR, 1.18; 95% CI, 1.13-1.23). With regard to specific outcomes, those with vs without placenta previa had a significantly higher rate of peripartum hemorrhage (878 patients [66.4%] vs 1217 patients [49.3%]; P < .001), blood product transfusion (413 patients [31.2%] vs 610 patients [24.7%]; P < .001), shock (83 patients [6.3%] vs 108 patients [4.4%]; P = .01), disseminated intravascular coagulation or other coagulopathy (77 patients [5.8%] vs 105 patients [4.3%]; P = .04), and urinary tract injury (44 patients [3.3%] vs 41 patients [1.7%]; P = .002). Patients with vs without placenta previa were more likely to undergo cesarean delivery (1292 patients [97.7%] vs 1787 patients [72.3%]; P < .001), hysterectomy (786 patients [59.4%] vs 689 patients [27.9%]; P < .001), cystoscopy (301 patients [22.8%] vs 203 patients [8.2%]; P < .001), cystectomy (157 patients [11.9%] vs 98 patients [4.0%]; P < .001), and intra-arterial balloon occlusion (121 patients [9.1%] vs 77 patients [3.1%]; P < .001) and to have significantly longer hospital length of stay (median [IQR], 5 [4-11] days vs 3 [3-5] days; P < .001) and total inpatient costs (median [IQR], $17 496 [$10 863-$30 619] vs $9728 [$6130-$16 790]; P < .001). Hypertensive disorder of pregnancy was associated with a decreased risk of placenta previa (aRR, 0.67; 95% CI, 0.46-0.96) among patients with PAS.Conclusions and relevanceIn this study, placenta previa was associated with an increased risk of maternal and surgical morbidities and higher resource use among women with PAS. These findings suggest that interventions to alleviate maternal and surgical morbidities are especially needed for patients with placenta previa-complicated PAS disorders.

Dataset Information

Radiomics-based prediction of FIGO grade for placenta accreta spectrum.

Background

Methods

Results

Conclusion

Relevance statement

Key points

Publications

Radiomics-based prediction of FIGO grade for placenta accreta spectrum.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets