Prediction of placenta accreta spectrum by combining deep learning and radiomics using T2WI: a multicenter study

MR Imaging of Placenta Accreta Spectrum: A Comprehensive Literature Review of the Most Recent Advancements

This article delves into the latest MR imaging developments dedicated to diagnosing placenta accreta spectrum (PAS). PAS, characterized by abnormal placental adherence to the uterine wall, is of paramount concern owing to its association with maternal morbidity and mortality, particularly in high-risk pregnancies featuring placenta previa and prior cesarean sections. Although ultrasound (US) remains the primary screening modality, limitations have prompted heightened emphasis on MR imaging. This review underscores the utility of quantitative MR imaging, especially where US findings prove inconclusive or when maternal body habitus poses challenges, acknowledging, however, that interpreting placenta MR imaging demands specialized training for radiologists.

Looking back to look forward: Has the time arrived for active management of obstetricians in placenta accreta spectrum?

Placenta accreta spectrum (PAS) is a relatively new obstetric condition which, until recently, was poorly understood. The true incidence is unknown because of the poor quality and heterogeneous diagnostic criteria. Classification systems have attempted to provide clarity on how to grade and diagnose PAS, but these are no longer reflective of our current understanding of PAS. This is particularly true for placenta percreta, which referred to extrauterine disease, as recent studies have demonstrated that placental villi associated with PAS have minimal potential to invade beyond the uterine serosa. It is accepted that PAS is a direct consequence of previous iatrogenic uterine injury, most commonly a previous cesarean section. Here, we "look back to look forwards"-starting with the primary predisposing factor for PAS, an iatrogenic uterine injury and subsequent wound healing. We then consider the evolution of definitions and diagnostic criteria of PAS from its first description over a century ago to current classifications. Finally, we discuss why modifications to the current classifications are needed to allow accurate diagnosis of this rare but life-threatening complication, while avoiding overdiagnosis and potential patient harm.

Deep Learning Radiomic Analysis of MRI Combined with Clinical Characteristics Diagnoses Placenta Accreta Spectrum and its Subtypes

BACKGROUND

Different placenta accreta spectrum (PAS) subtypes pose varying surgical risks to the parturient. Machine learning model has the potential to diagnose PAS disorder.

PURPOSE

To develop a cascaded deep semantic-radiomic-clinical (DRC) model for diagnosing PAS and its subtypes based on T2-weighted MRI.

STUDY TYPE

Retrospective.

POPULATION

361 pregnant women (mean age: 33.10 ± 4.37 years), suspected of PAS, divided into segment training cohort (N = 40), internal training cohort (N = 139), internal testing cohort (N = 60), and external testing cohort (N = 122).

FIELD STRENGTH/SEQUENCE

Coronal T2-weighted sequence at 1.5 T and 3.0 T.

ASSESSMENT

Clinical characteristics such as history of uterine surgery and the presence of placenta previa, complete placenta previa and dangerous placenta previa were extracted from clinical records. The DRC model (incorporating radiomics, deep semantic features, and clinical characteristics), a cumulative radiological score method performed by radiologists, and other models (including a radiomics and clinical, the clinical, radiomics and deep learning models) were developed for PAS disorder diagnosing (existence of PAS and its subtypes).

STATISTICAL TESTS

AUC, ACC, Student's t-test, the Mann-Whitney U test, chi-squared test, dice coefficient, intraclass correlation coefficients, least absolute shrinkage and selection operator regression, receiver operating characteristic curve, calibration curve with the Hosmer-Lemeshow test, decision curve analysis, DeLong test, and McNemar test. P < 0.05 indicated a significant difference.

RESULTS

In PAS diagnosis, the DRC-1 outperformed than other models (AUC = 0.850 and 0.841 in internal and external testing cohorts, respectively). In PAS subtype classification (abnormal adherent placenta and abnormal invasive placenta), DRC-2 model performed similarly with radiologists (P = 0.773 and 0.579 in the internal testing cohort and P = 0.429 and 0.874 in the external testing cohort, respectively).

DATA CONCLUSION

The DRC model offers efficiency and high diagnostic sensitivity in diagnosis, aiding in surgical planning.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

Prenatal Diagnosis of Placenta Accreta Spectrum Disorders: Deep Learning Radiomics of Pelvic MRI

BACKGROUND

Diagnostic performance of placenta accreta spectrum (PAS) by prenatal MRI is unsatisfactory. Deep learning radiomics (DLR) has the potential to quantify the MRI features of PAS.

PURPOSE

To explore whether DLR from MRI can be used to identify pregnancies with PAS.

STUDY TYPE

Retrospective.

POPULATION

324 pregnant women (mean age, 33.3 years) suspected PAS (170 training and 72 validation from institution 1, 82 external validation from institution 2) with clinicopathologically proved PAS (206 PAS, 118 non-PAS).

FIELD STRENGTH/SEQUENCE

3-T, turbo spin-echo T2-weighted images.

ASSESSMENT

The DLR features were extracted using the MedicalNet. An MRI-based DLR model incorporating DLR signature, clinical model (different clinical characteristics between PAS and non-PAS groups), and MRI morphologic model (radiologists' binary assessment for the PAS diagnosis) was developed. These models were constructed in the training dataset and then validated in the validation datasets.

STATISTICAL TESTS

The Student t-test or Mann-Whitney U, χ2 or Fisher exact test, Kappa, dice similarity coefficient, intraclass correlation coefficients, least absolute shrinkage and selection operator logistic regression, multivariate logistic regression, receiver operating characteristic (ROC) curve, DeLong test, net reclassification improvement (NRI) and integrated discrimination improvement (IDI), calibration curve with Hosmer-Lemeshow test, decision curve analysis (DCA). P < 0.05 indicated a significant difference.

RESULTS

The MRI-based DLR model had a higher area under the curve than the clinical model in three datasets (0.880 vs. 0.741, 0.861 vs. 0.772, 0.852 vs. 0.675, respectively) or MRI morphologic model in training and independent validation datasets (0.880 vs. 0.760, 0.861, vs. 0.781, respectively). The NRI and IDI were 0.123 and 0.104, respectively. The Hosmer-Lemeshow test had nonsignificant statistics (P = 0.296 to 0.590). The DCA offered a net benefit at any threshold probability.

DATA CONCLUSION

An MRI-based DLR model may show better performance in diagnosing PAS than a clinical or MRI morphologic model.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

Deep Learning Model Based on Multisequence MRI Images for Assessing Adverse Pregnancy Outcome in Placenta Accreta

BACKGROUND

Preoperative assessment of adverse outcomes risk in placenta accreta spectrum (PAS) disorders is of high clinical relevance for perioperative management and prognosis.

PURPOSE

To investigate the association of preoperative MRI multisequence images and adverse pregnancy outcomes by establishing a deep learning model in patients with PAS.

STUDY TYPE

Retrospective.

POPULATION

323 pregnant women (age from 20 to 46, the median age is 33), suspected of PAS, underwent MRI to assess the PAS, divided into the training (N = 227) and validation datasets (N = 96).

FIELD STRENGTH/SEQUENCE

1.5T scanner/fast imaging employing steady-state acquisition sequence and single shot fast spin echo sequence.

ASSESSMENT

Different deep learning models (i.e., with single MRI input sequence/two sequences/multisequence) were compared to assess the risk of adverse pregnancy outcomes, which defined as intraoperative bleeding ≥1500 mL and/or hysterectomy. Net reclassification improvement (NRI) was used for quantitative comparison of assessing adverse pregnancy outcome between different models.

STATISTICAL TESTS

The AUC, sensitivity, specificity, and accuracy were used for evaluation. The Shapiro-Wilk test and t-test were used. A P value of <0.05 was considered statistically significant.

RESULTS

215 cases were invasive placenta accreta (67.44% of them with adverse outcomes) and 108 cases were non-invasive placenta accreta (9.25% of them with adverse outcomes). The model with four sequences assessed adverse pregnancy outcomes with AUC of 0.8792 (95% CI, 0.8645-0.8939), with ACC of 85.93% (95%, 84.43%-87.43%), with SEN of 86.24% (95% CI, 82.46%-90.02%), and with SPC of 85.62% (95%, 82.00%-89.23%) on the test cohort. The performance of model with four sequences improved above 0.10 comparing with that of model with two sequences and above 0.20 comparing with that of model with single sequence in terms of NRI.

DATA CONCLUSION

The proposed model showed good diagnostic performance for assessing adverse pregnancy outcomes.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

Enhancing Automatic Placenta Analysis through Distributional Feature Recomposition in Vision-Language Contrastive Learning

The placenta is a valuable organ that can aid in understanding adverse events during pregnancy and predicting issues post-birth. Manual pathological examination and report generation, however, are laborious and resource-intensive. Limitations in diagnostic accuracy and model efficiency have impeded previous attempts to automate placenta analysis. This study presents a novel framework for the automatic analysis of placenta images that aims to improve accuracy and efficiency. Building on previous vision-language contrastive learning (VLC) methods, we propose two enhancements, namely Pathology Report Feature Recomposition and Distributional Feature Recomposition, which increase representation robustness and mitigate feature suppression. In addition, we employ efficient neural networks as image encoders to achieve model compression and inference acceleration. Experiments validate that the proposed approach outperforms prior work in both performance and efficiency by significant margins. The benefits of our method, including enhanced efficacy and deployability, may have significant implications for reproductive healthcare, particularly in rural areas or low- and middle-income countries.

Radiomics-based prediction of FIGO grade for placenta accreta spectrum

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.

Machine learning and disease prediction in obstetrics

Machine learning technologies and translation of artificial intelligence tools to enhance the patient experience are changing obstetric and maternity care. An increasing number of predictive tools have been developed with data sourced from electronic health records, diagnostic imaging and digital devices. In this review, we explore the latest tools of machine learning, the algorithms to establish prediction models and the challenges to assess fetal well-being, predict and diagnose obstetric diseases such as gestational diabetes, pre-eclampsia, preterm birth and fetal growth restriction. We discuss the rapid growth of machine learning approaches and intelligent tools for automated diagnostic imaging of fetal anomalies and to asses fetoplacental and cervix function using ultrasound and magnetic resonance imaging. In prenatal diagnosis, we discuss intelligent tools for magnetic resonance imaging sequencing of the fetus, placenta and cervix to reduce the risk of preterm birth. Finally, the use of machine learning to improve safety standards in intrapartum care and early detection of complications will be discussed. The demand for technologies to enhance diagnosis and treatment in obstetrics and maternity should improve frameworks for patient safety and enhance clinical practice.

The prenatal diagnostic indicators of placenta accreta spectrum disorders

Placenta accreta spectrum (PAS) disorders refers to a heterogeneous group of anomalies distinguished by abnormal adhesion or invasion of chorionic villi through the myometrium and uterine serosa. PAS frequently results in life-threatening complications, including postpartum hemorrhage and hysterotomy. The incidence of PAS has increased recently as a result of rising cesarean section rates. Consequently, prenatal screening for PAS is essential. Despite the need to increase specificity, ultrasound is still considered a primary adjunct. Given the dangers and adverse effects of PAS, it is necessary to identify pertinent markers and validate indicators to improve prenatal diagnosis. This article summarizes the predictors regarding biomarkers, ultrasound indicators, and magnetic resonance imaging (MRI) features. In addition, we discuss the effectiveness of joint diagnosis and the most recent research on PAS. In particular, we focus on (a) posterior placental implantation and (b) accreta after in vitro fertilization-embryo transfer, both of which have low diagnostic rates. At last, we graphically display the prenatal diagnostic indicators and each diagnostic performance.

Normalization Strategies in Multi-Center Radiomics Abdominal MRI: Systematic Review and Meta-Analyses

Goal: Artificial intelligence applied to medical image analysis has been extensively used to develop non-invasive diagnostic and prognostic signatures. However, these imaging biomarkers should be largely validated on multi-center datasets to prove their robustness before they can be introduced into clinical practice. The main challenge is represented by the great and unavoidable image variability which is usually addressed using different pre-processing techniques including spatial, intensity and feature normalization. The purpose of this study is to systematically summarize normalization methods and to evaluate their correlation with the radiomics model performances through meta-analyses. This review is carried out according to the PRISMA statement: 4777 papers were collected, but only 74 were included. Two meta-analyses were carried out according to two clinical aims: characterization and prediction of response. Findings of this review demonstrated that there are some commonly used normalization approaches, but not a commonly agreed pipeline that can allow to improve performance and to bridge the gap between bench and bedside.