Identification of suspicious invasive placentation based on clinical MRI data using textural features and automated machine learning

Perspective in diagnostic accuracy of prenatal ultrasound and MRI for placenta accreta

PURPOSE

Placenta accreta (PA) significantly increases the risk of life-threatening maternal outcomes, and its rising prevalence, driven by the increase in cesarean deliveries, underscores the need for precise diagnostic tools to improve clinical management and outcomes. This study aims to evaluate the advanced diagnostic capabilities of prenatal ultrasound and magnetic resonance imaging (MRI) in the detection of PA, a severe obstetric complication characterized by abnormal adherence of the placenta to the myometrium.

MATERIALS AND METHODS

The study utilized a review of current literature and clinical studies to assess the diagnostic accuracy and clinical utility of ultrasound and MRI in identifying PA. Both imaging modalities were evaluated for their ability to assess the depth and extent of placental invasion, as well as their complementary roles in prenatal diagnosis. The experimental system included detailed imaging protocols for ultrasound and MRI, focusing on placental and uterine structures, and their application in real-world clinical settings.

RESULTS

The findings demonstrate that ultrasound and MRI are highly effective in diagnosing PA, with each modality offering unique advantages. Ultrasound is widely accessible and serves as the first-line diagnostic tool, providing detailed visualization of placental adherence and vascular patterns. MRI, on the other hand, offers superior soft tissue contrast and is particularly valuable in complex cases or when ultrasound findings are inconclusive. Together, these imaging techniques enable accurate evaluation of placental invasion, facilitating timely and targeted prenatal interventions. The study also highlights the potential for improved maternal and fetal outcomes through early diagnosis and optimized pregnancy management.

CONCLUSIONS

Prenatal ultrasound and MRI are indispensable tools in the diagnosis and management of placenta accreta, offering complementary insights that enhance diagnostic precision. Their combined use allows for detailed assessment of placental and uterine structures, guiding clinical decision-making and improving outcomes for both mothers and infants. Future advancements in imaging technology and research hold promise for further enhancing diagnostic accuracy and expanding clinical applications, ultimately contributing to safer and more effective care for patients with PA.

Predicting placenta accreta spectrum and high postpartum hemorrhage risk using radiomics from T2-weighted MRI

BACKGROUND

Antenatal diagnosis of placenta accreta spectrum (PAS) is of critical importance, considering that women have much better outcomes when delivery occurs at a level III or IV maternal care facility before labor initiation or bleeding, thus avoiding placental disruption. Herein, we aimed to investigate the performance of magnetic resonance imaging (MRI) in antenatal prediction of PAS and postpartum hemorrhage (PPH).

METHODS

This retrospective study included 433 women with singleton pregnancies (PAS group, n = 208; non-PAS group, n = 225; PPH-positive (PPH (+)) group, n = 80; PPH-negative (PPH (-)) group, n = 353), who were randomly divided into a training set and a test set in a 7:3 ratio. Radiomic features were extracted from T2WI (T2-weighted imaging). Features strongly correlated with PAS and PPH (p < 0.05) were selected using Pearson correlation, followed by LASSO regression for dimensionality reduction. Subsequently, radiomics models were constructed for PAS and PPH risk prediction, respectively. Regression analyses were conducted using radiomics score (R-score) and clinical factors to identify independent clinical risk factors for PAS and PPH, leading to the development of corresponding clinical models. Next, clinical-radiomics models were built by combining R-score and clinical risk factors. The predictive performance of the models was evaluated using nomograms, calibration curves, and decision curves.

RESULTS

The clinical-radiomics models and radiomics models for predicting PAS and PPH risk both outperformed their clinical models in the training and testing sets. For PAS, the AUC (Area Under the Receiver Operating Characteristic Curve) of the clinical-radiomics model, radiomics model, and clinical model in the training set are 0.918, 0.908, and 0.755, respectively, and in the testing set, the AUCs are 0.885, 0.866, and 0.771, respectively. For PPH, the AUCs of the clinical-radiomics model, radiomics model, and clinical model in the training set are 0.918, 0.884, and 0.723, respectively, and in the testing set, the AUCs are 0.905, 0.860, and 0.688, respectively. The DeLong test p-values between the clinical-radiomics models and radiomics models for predicting PAS and PPH are both less than 0.05. Additionally, in the testing set, the clinical-radiomics models perform best in predicting PAS and PPH risk, with accuracies of 82.31% and 84.61%, respectively.

CONCLUSION

This novel clinical-radiomics model has a robust performance in predicting PAS antepartum and predicting massive PPH in pregnancies.

Comparative Evaluation of Clinical-MRI, Radiomics, and Integrated Nomogram Models for Preoperative Prediction of Placenta Accreta Spectrum

RATIONALE AND OBJECTIVES

The escalating incidence of placental accreta spectrum (PAS), a pregnancy complication, underscores the need for accurate prenatal diagnosis to guide optimal management strategies. This study aims to develop, validate, and compare various prenatal PAS prediction models integrating clinical data, MRI signs, and radiomics signatures.

MATERIALS AND METHODS

A cohort comprising 111 patients (72 with PAS and 39 without, denoted as N-PAS) served as the training set, while another 47 patients (33 PAS and 14 N-PAS) constituted the validation set. Clinical features and MRI signs were subjected to univariate and multivariate analyses to construct the Clinical-MRI model. Radiomic features were extracted from MRI images and refined through the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm, thereby establishing the Radiomics model. An optimal set of radiomic features was utilized to derive the Radscore, which was then integrated with clinical features and MRI signs to formulate the Nomogram model. The performance of these models was comprehensively evaluated and compared.

RESULTS

In the validation set evaluation, the Nomogram model, which integrated Radscore, a pivotal clinical indicator, and two MRI signs, demonstrated superior performance. With an area under the curve (AUC) of 0.861 (95% CI: 0.745, 0.978), this model significantly outperformed both the clinical-MRI model (AUC = 0.796, 95% CI: 0.649, 0.943) and the radiomics model (AUC = 0.704, 95% CI: 0.531, 0.877). Specifically, the Nomogram model achieved a high sensitivity of 81.8% and a specificity of 78.6% in the prenatal diagnosis of placenta accreta spectrum (PAS), thereby offering clinicians a precise and efficient diagnostic aid.

CONCLUSION

The radiomics-derived Radscore serves as an independent predictor for prenatal PAS. Combining Radscore with clinical features and MRI signs into a Nomogram model provides a non-invasive tool with high sensitivity or specificity for PAS diagnosis, enhancing prenatal assessment and management.

Radiomics analysis of placental MRI for prenatal prediction of placenta accreta spectrum in pregnant women in the third trimester: A retrospective study of 594 patients

OBJECTIVE

To develop and validate a model based on placental MRI for the prenatal prediction of placenta accreta spectrum (PAS) in pregnant women in the third trimester.

MATERIALS AND METHODS

A total of 594 pregnant women who were suspected of having PAS and underwent placental MRI antenatally were included and were allocated into the training cohort and testing cohort at a 2:1 ratio. MRI diagnosis was determined by three experienced radiologists. Radiomic features were extracted from images of T2 weighted imaging for each patient. After a feature selection strategy, a radiomics signature and a clinical-radiomics nomogram combining radiomics score and clinical risk factors were constructed to predict PAS. The performance of each model was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and clinical utility.

RESULTS

MRI diagnosis yielded AUCs of 0.77 and 0.79 for predicting PAS in the training and testing cohorts, respectively. The AUCs of the radiomics signature used to predict PAS in both cohorts were 0.80 and 0.83, respectively. The nomogram accurately predicted PAS in both cohorts (AUC = 0.84 and 0.89), with better results than those of MRI diagnosis and radiomics signature in the training (p = 0.009 and 0.003, respectively) and testing cohorts (p = 0.010 and 0.008, respectively), decision curve analysis confirmed its best clinical utility compared to the other models.

CONCLUSION

Radiomics analysis based on placental MRI may serve as an effective tool to predict PAS in patients with possible PAS in the third trimester.

Association between deep learning radiomics based on placental MRI and preeclampsia with fetal growth restriction: A multicenter study

PURPOSE

Preeclampsia (PE) is associated with placental insufficiency and could lead to adverse pregnancy outcomes. The study aimed to develop a placental T2-weighted image-based automatic quantitative model for the identification of PE pregnancies and disease severity.

METHODS

Between July 2013 and September 2022, the retrospective multicenter study featured 420 pregnant women, including 140 cases of PE and 280 cases of normotensive pregnancies. The semi-supervised approach was used to gain an automatic segmentation for placental MRI. The radiomics, deep learning, and deep learning radiomics (DLR) models were built.

RESULTS

In PE pregnancies, 65 (46.4 %) fetuses developed PE with fetal growth restriction (FGR), and 75 (53.6 %) cases were PE without FGR. The Dice of semi-supervised placental segmentation was 0.917. The AUCs of the DLR signature for discriminating PE pregnancies from normotensive pregnancies were 0.839 (95 % CI: 0.793-0.886), 0.858 (95 % CI: 0.742-0.974), 0.888 (95 % CI: 0.783-0.992), and 0.843 (95 % CI: 0.731-1.000) in the training, test, internal validation, and external validation sets, respectively. This DLR analysis model performed well in discriminating between PE with FGR and normotensive pregnancies (AUC = 0.918, 95 % CI: 0.879-0.957) and PE without FGR (AUC = 0.742, 95 % CI: 0659-0.824).

CONCLUSION

The automatic radiomics analysis has been developed to identify PE pregnancies by determining DLR features on placental T2-weighted images, and to predict FGR exposed to PE.

Machine learning applications in placenta accreta spectrum disorders

This review examines the emerging applications of machine learning (ML) and radiomics in the diagnosis and prediction of placenta accreta spectrum (PAS) disorders, addressing a significant challenge in obstetric care. It highlights recent advancements in ML algorithms and radiomic techniques that utilize medical imaging modalities like magnetic resonance imaging (MRI) and ultrasound for effective classification and risk stratification of PAS. The review discusses the efficacy of various deep learning models, such as nnU-Net and DenseNet-PAS, which have demonstrated superior performance over traditional diagnostic methods through high AUC scores. Furthermore, it underscores the importance of integrating quantitative imaging features with clinical data to enhance diagnostic accuracy and optimize surgical planning. The potential of ML to predict surgical morbidity by analyzing demographic and obstetric factors is also explored. Emphasizing the need for standardized methodologies to ensure consistent feature extraction and model performance, this review advocates for the integration of radiomics and ML into clinical workflows, aiming to improve patient outcomes and foster a multidisciplinary approach in high-risk pregnancies. Future research should focus on larger datasets and validation of biomarkers to refine predictive models in obstetric care.

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.

Ensemble machine learning framework for predicting maternal health risk during pregnancy

Maternal health risks can cause a range of complications for women during pregnancy. High blood pressure, abnormal glucose levels, depression, anxiety, and other maternal health conditions can all lead to pregnancy complications. Proper identification and monitoring of risk factors can assist to reduce pregnancy complications. The primary goal of this research is to use real-world datasets to identify and predict Maternal Health Risk (MHR) factors. As a result, we developed and implemented the Quad-Ensemble Machine Learning framework to predict Maternal Health Risk Classification (QEML-MHRC). The methodology used a vacxsriety of Machine Learning (ML) models, which then integrated with four ensemble ML techniques to improve prediction. The dataset collected from various maternity hospitals and clinics subjected to nineteen training and testing tests. According to the exploratory data analysis, the most significant risk factors for pregnant women include high blood pressure, low blood pressure, and high blood sugar levels. The study proposed a novel approach to dealing with high-risk factors linked to maternal health. Dealing with class-specific performance elaborated further to properly understand the distinction between high, low, and medium risks. All tests yielded outstanding results when predicting the amount of risk during pregnancy. In terms of class performance, the dataset associated with the "HR" class outperformed the others, predicting 90% correctly. GBT with ensemble stacking outperformed and demonstrated remarkable performance for all evaluation measure (0.86) across all classes in the dataset. The key success of the models used in this work is the ability to measure model performance using a class-wise distribution. The proposed approach can help medical experts assess maternal health risks, saving lives and preventing complications throughout pregnancy. The prediction approach presented in this study can detect high-risk pregnancies early on, allowing for timely intervention and treatment. This study's development and findings have the potential to raise public awareness of maternal health issues.

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.

Artificial Intelligence in Obstetrics and Gynecology: Transforming Care and Outcomes

The integration of artificial intelligence (AI) in obstetrics and gynecology (OB/GYN) is revolutionizing the landscape of women's healthcare. This review article explores the transformative impact of AI technologies on the diagnosis, treatment, and management of obstetric and gynecological conditions. We examine key advancements in AI-driven imaging techniques, predictive analytics, and personalized medicine, highlighting their roles in enhancing prenatal care, improving maternal and fetal outcomes, and optimizing gynecological interventions. The article also addresses the challenges and ethical considerations associated with the implementation of AI in clinical practice. This paper highlights the potential of AI to greatly improve the standard of care in OB/GYN, ultimately leading to better health outcomes for women, by offering a thorough overview of present AI uses and future prospects.

Placental T2WI MRI-based radiomics-clinical nomogram predicts suspicious placenta accreta spectrum in patients with placenta previa

BACKGROUND

The incidence of placenta accreta spectrum (PAS) increases in women with placenta previa (PP). Many radiologists sometimes cannot completely and accurately diagnose PAS through the simple visual feature analysis of images, which can affect later treatment decisions. The study is to develop a T2WI MRI-based radiomics-clinical nomogram and evaluate its performance for non-invasive prediction of suspicious PAS in patients with PP.

METHODS

The preoperative MR images and related clinical data of 371 patients with PP were retrospectively collected from our hospital, and the intraoperative examination results were used as the reference standard of the PAS. Radiomics features were extracted from sagittal T2WI MR images and further selected by LASSO regression analysis. The radiomics score (Radscore) was calculated with logistic regression (LR) classifier. A nomogram integrating Radscore and selected clinical factors was also developed. The model performance was assessed with respect to discrimination, calibration and clinical usefulness.

RESULTS

A total of 6 radiomics features and 1 clinical factor were selected for model construction. The Radscore was significantly associated with suspicious PAS in both the training (p < 0.001) and validation (p < 0.001) datasets. The AUC of the nomogram was also higher than that of the Radscore in the training dataset (0.891 vs. 0.803, p < 0.001) and validation dataset (0.897 vs. 0.780, p < 0.001), respectively. The calibration was good, and the decision curve analysis demonstrated the nomogram had higher net benefit than the Radscore.

CONCLUSIONS

The T2WI MRI-based radiomics-clinical nomogram showed favorable diagnostic performance for predicting PAS in patients with PP, which could potentially facilitate the obstetricians for making clinical decisions.

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.

Standard diffusion-weighted, diffusion kurtosis and intravoxel incoherent motion in differentiating invasive placentas

PURPOSE

To investigate the diagnostic value of monoexponential, biexponential, and diffusion kurtosis MR imaging (MRI) in distinguishing invasive placentas.

METHODS

A total of 53 patients with invasive placentas and 47 patients with noninvasive placentas undergoing conventional diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI) were retrospectively enrolled. The mean, minimum, and maximum parameters including the apparent diffusion coefficient (ADC) and exponential ADC (eADC) from standard DWI, diffusion kurtosis (MK), and diffusion coefficient (MD) from DKI and pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) from IVIM were measured and compared from the volumetric analysis. Receiver operating characteristics (ROC) curve and logistic regression analyses were conducted to evaluate the diagnostic efficiency of different diffusion parameters for distinguishing invasive placentas.

RESULTS

Comparisons between accreta lesions in patients with invasive placentas (AL) and lower 1/3 part of the placenta in patients with noninvasive placentas (LP) demonstrated that MD mean, D mean, and D* mean were significantly lower while ADC max and D max were significantly higher in invasive placentas (all p < 0.05). Multivariate analysis demonstrated that D mean, D max and D* mean differed significantly among all the studied parameters for invasive placentas. A combined use of these three parameters yielded an AUC of 0.86 with sensitivity, specificity, and accuracy of 84.91%, 76.60%, and 80%, respectively.

CONCLUSION

The combined use of different IVIM parameters is helpful in distinguishing invasive placentas.

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.

Segmentation methods applied to MRI-derived radiomic analysis for the prediction of placenta accreta spectrum in patients with placenta previa

PURPOSE

To retrospectively evaluate the performance of different manual segmentation methods of placenta MR images for predicting Placenta Accreta Spectrum (PAS) disorders in patients with placenta previa (PP) using a Machine Learning (ML) Radiomics analysis.

METHODS

64 patients (n=41 with PAS and n= 23 without PAS) with PP who underwent MRI examination for suspicion of PAS were retrospectively selected. All MRI examinations were acquired on a 1.5 T using T2-weighted (T2w) sequences on axial, sagittal and coronal planes. Ten different manual segmentation methods were performed on sagittal placental T2-weighted images obtaining five sets of 2D regions of interest (ROIs) and five sets of 3D volumes of interest (VOIs) from each patient. In detail, ROIs and VOIs were positioned on the following areas: placental tissue, retroplacental myometrium, cervix, placenta with underneath myometrium, placenta with underneath myometrium and cervix. For feature stability testing, the same process was repeated on 30 randomly selected placental MRI examinations by two additional radiologists, working independently and blinded to the original segmentation. Radiomic features were extracted from all available ROIs and VOIs. 100 iterations of 5-fold cross-validation with nested feature selection, based on recursive feature elimination, were subsequently run on each ROI/VOI to identify the best-performing method to classify instances correctly.

RESULTS

Among the segmentation methods, the best performance in predicting PAS was obtained by the VOIs covering the retroplacental myometrium (Mean validation score: 0.761, standard deviation: 0.116).

CONCLUSION

Our preliminary results show that the VOI including the retroplacental myometrium using T2w images seems to be the best method when segmenting images for the development of ML radiomics predictive models to identify PAS in patients with PP.

External Validation of a Multivariable Prediction Model for Placenta Accreta Spectrum

BACKGROUND

Placenta accreta spectrum (PAS) is a disorder of abnormal placentation associated with severe postpartum hemorrhage, maternal morbidity, and mortality. Predelivery prediction of this condition is important to determine appropriate delivery location and multidisciplinary planning for operative management. This study aimed to validate a prediction model for PAS developed by Weiniger et al in 2 cohorts who delivered at 2 different United States tertiary centers.

METHODS

Cohort A (Brigham and Women's Hospital; N = 253) included patients with risk factors (prior cesarean delivery and placenta previa) and/or ultrasound features of PAS presenting to a tertiary-care hospital. Cohort B (Columbia University Irving Medical Center; N = 99) consisted of patients referred to a tertiary-care hospital specifically because of ultrasound features of PAS. Using the outcome variable of surgical and/or pathological diagnosis of PAS, discrimination (via c-statistic), calibration (via intercept, slope, and flexible calibration curve), and clinical usefulness (via decision curve analysis) were determined.

RESULTS

The model c-statistics in cohorts A and B were 0.728 (95% confidence interval [CI], 0.662-0.794) and 0.866 (95% CI, 0.754-0.977) signifying acceptable and excellent discrimination, respectively. The calibration intercept (0.537 [95% CI, 0.154-0.980] for cohort A and 3.001 [95% CI, 1.899- 4.335] for B), slopes (0.342 [95% CI, 0.170-0.532] for cohort A and 0.604 [95% CI, -0.166 to 1.221] for B), and flexible calibration curves in each cohort indicated that the model underestimated true PAS risks on average and that there was evidence of overfitting in both validation cohorts. The use of the model compared to a treat-all strategy by decision curve analysis showed a greater net benefit of the model at a threshold probability of >0.25 in cohort A. However, no net benefit of the model over the treat-all strategy was seen in cohort B at any threshold probability.

CONCLUSIONS

The performance of the Weiniger model is variable based on the case-mix of the population with regard to PAS clinical risk factors and ultrasound features, highlighting the importance of spectrum bias when applying this PAS prediction model to distinct populations. The model showed benefit for predicting PAS in populations with substantial case-mix heterogeneity at threshold probability of >25%.

Prediction of placenta accreta spectrum with nomogram combining radiomic and clinical factors: A novel developed and validated integrative model

OBJECTIVE

To develop and validate a clinicoradiomic nomogram based on sagittal T2WI images to predict placenta accreta spectrum (PAS).

METHODS

Between October 2016 and April 2022, women suspected of PAS by ultrasound were enrolled. After taking into account exclusion criteria, 132 women were retrospectively included in the study. The variance threshold SelectKBest and the least absolute shrinkage and selection operator were applied to select radiomic features, which was further used to calculate the Rad-score. Multivariable logistic regression was used to screen clinical factor.

RESULTS

Based on 13 radiomic features, five radiomic models were constructed. A clinical factor of intraplacental T2-hypointense bands was obtained by multivariate logistic regression. The area under the curve (AUC) value of the stochastic gradient descent (SGD) radiomic model was 0.82 in the training cohort and 0.78 in the test cohort. After adding clinical factors to the SGD radiomic model, the AUC value of the clinicoradiomic model was significantly increased from 0.82 and 0.78 to 0.84 in both the training and test cohorts. The nomogram of the clinicoradiomic model was constructed, which had good performance verified by calibration and a decision curve.

CONCLUSION

The presented nomogram could be useful for predicting PAS.

Influence of maternal psychological distress during COVID-19 pandemic on placental morphometry and texture

The Coronavirus Disease 2019 (COVID-19) pandemic has been accompanied by increased prenatal maternal distress (PMD). PMD is associated with adverse pregnancy outcomes which may be mediated by the placenta. However, the potential impact of the pandemic on in vivo placental development remains unknown. To examine the impact of the pandemic and PMD on in vivo structural placental development using advanced magnetic resonance imaging (MRI), acquired anatomic images of the placenta from 63 pregnant women without known COVID-19 exposure during the pandemic and 165 pre-pandemic controls. Measures of placental morphometry and texture were extracted. PMD was determined from validated questionnaires. Generalized estimating equations were utilized to compare differences in PMD placental features between COVID-era and pre-pandemic cohorts. Maternal stress and depression scores were significantly higher in the pandemic cohort. Placental volume, thickness, gray level kurtosis, skewness and run length non-uniformity were increased in the pandemic cohort, while placental elongation, mean gray level and long run emphasis were decreased. PMD was a mediator of the association between pandemic status and placental features. Altered in vivo placental structure during the pandemic suggests an underappreciated link between disturbances in maternal environment and perturbed placental development. The long-term impact on offspring is currently under investigation.

MRI-based radiomics nomogram in patients with high-risk placenta accreta spectrum: can it aid in the prenatal diagnosis of intraoperative blood loss?

PURPOSE

To develop and validate the nomogram by combining MRI-derived radiomics and clinical features for preoperatively predicting massive intraoperative blood loss (IBL) in high-risk placenta accreta spectrum (PAS) patients.

METHODS

A total of 152 high-risk PAS patients from Hospital A were enrolled and constituted the training cohort, and 64 patients from Hospital B constituted the validation cohort. Clinical features were analyzed retrospectively. Placental regions of interest were manually positioned on sagittal T₂-weighted HASTE images for each patient to extract quantitative radiomics features. Clinical model, radiomics model, and nomogram were built to predict the risk of massive IBL. The diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC) and the DeLong test. Decision curve analysis (DCA) was performed to determine the performance of the best predictive model.

RESULTS

The nomogram (AUC = 0.866 and 0.876, respectively) and radiomics model (AUC = 0.821 and 0.855, respectively) outperformed the clinical model (AUC = 0.685 and 0.619, respectively) both in the training and validation cohorts (Delong test, P < 0.05). Furthermore, the nomogram performed best with an accuracy of 0.844, sensitivity of 0.882, and specificity of 0.830 for differentiating massive IBL in the validation cohort. DCA confirmed the clinical utility of the nomogram.

CONCLUSION

The nomogram can be used to noninvasively predict massive IBL patients and guide obstetricians to make reasonable preoperative treatment plans.

Effects of randomness in the development of machine learning models in neuroimaging studies of schizophrenia

Numerous studies have used machine learning with neuroimaging data for identifying individuals with a schizophrenia diagnosis. However, inconsistent results have limited the ability of the psychiatric community to objectively judge and accept the value of this approach. One factor that has contributed to the inconsistency, but has long been ignored, is randomness in the practice of machine learning. This is manifest when executing the same machine learning pipeline multiple times on the same dataset but getting different results. In the current study, a dataset of anatomical MRI scans from 158 patients with first-episode medication-naïve schizophrenia and 166 matched controls was used to investigate the effect of randomness on classifier performance estimates under different algorithm complexity and data splitting ratios. The maximum discriminatory accuracy that could be reached was 62.6 % ± 4.7 % (43.5 %-79.3 %) obtained when using extra-trees classifiers without feature normalization. Regions contributing to discrimination were located at bilateral temporal lobes and right frontal lobe. The results show that randomness has a significant impact on the precision of model performance estimates, especially when the size of test set is small. Current neuroimaging feature engineering combined with machine learning still falls short of being able to make diagnoses in the clinical context, but has value in revealing patterns of regional brain alteration associated with the illness. The current results indicate that effects of randomness on model performance should be reported and considered in interpreting model utility and it is necessary to evaluate models on large test sets to obtain valid estimates of model performance.

Use of artificial intelligence in obstetrics: not quite ready for prime time

Artificial intelligence is finding several applications in healthcare settings. This study aimed to report evidence on the effectiveness of artificial intelligence application in obstetrics. Through a narrative review of literature, we described artificial intelligence use in different obstetrical areas as follows: prenatal diagnosis, fetal heart monitoring, prediction and management of pregnancy-related complications (preeclampsia, preterm birth, gestational diabetes mellitus, and placenta accreta spectrum), and labor. Artificial intelligence seems to be a promising tool to help clinicians in daily clinical activity. The main advantages that emerged from this review are related to the reduction of inter- and intraoperator variability, time reduction of procedures, and improvement of overall diagnostic performance. However, nowadays, the diffusion of these systems in routine clinical practice raises several issues. Reported evidence is still very limited, and further studies are needed to confirm the clinical applicability of artificial intelligence. Moreover, better training of clinicians designed to use these systems should be ensured, and evidence-based guidelines regarding this topic should be produced to enhance the strengths of artificial systems and minimize their limits.

Texture analysis of sonographic image of placenta in pregnancies with normal and adverse outcomes, a pilot study

INTRODUCTION

Many maternal and fetal morbidity and mortality from complications of pregnancy have been attributed to placenta abnormality. Placenta assessment in developing countries is mainly through ultrasonography which is subjective and prone to error. Objective assessment of placental abnormalities through texture analysis has been frequently done using magnetic resonance images with little done on ultrasound generated images, thus, the need for this study. The study is aimed at using statistical texture analysis in characterizing placenta tissue into normal and abnormal placenta as well as testing the accuracy of different texture analysis algorithms in differentiating placenta into normal and abnormal placental tissues.

METHODS

This longitudinal study involved 500 ultrasound-generated placenta images from patients screened for adverse pregnancy outcomes in a private hospital in Enugu. These images were loaded onto an HP laptop for viewing. Two regions of interest were selected from the placenta tissue where texture features were extracted and were classified into normal and abnormal placentas using MaZda® software version 47 while the accuracy of the classification descriptors was assessed using WEKA classification algorithms.

RESULTS

Co-occurrence matrix, run length matrix and histogram parameters differentiated normal placenta tissue from abnormal placental tissues (p-value <0.05) while variance is the only absolute gradient parameter that can differentiate normal placenta tissue from abnormal placenta tissues. All feature descriptors show high classification accuracy using KNN and ANN algorithms.

CONCLUSION

Texture analysis can differentiate normal placenta tissues from abnormal placenta tissue which will reduce the errors associated with subjective assessment of the placenta echogenicity.

IMPLICATIONS FOR PRACTICE

Integrating these computer-aided algorithms into our ultrasound machines will lead to early detection of abnormal placenta tissues as early management results in better pregnancy outcomes.

Applications of artificial intelligence in obstetrics

Artificial intelligence, which has been applied as an innovative technology in multiple fields of healthcare, analyzes large amounts of data to assist in disease prediction, prevention, and diagnosis, as well as in patient monitoring. In obstetrics, artificial intelligence has been actively applied and integrated into our daily medical practice. This review provides an overview of artificial intelligence systems currently used for obstetric diagnostic purposes, such as fetal cardiotocography, ultrasonography, and magnetic resonance imaging, and demonstrates how these methods have been developed and clinically applied.

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

PURPOSE

To achieve prenatal prediction of placenta accreta spectrum (PAS) by combining clinical model, radiomics model, and deep learning model using T2-weighted images (T2WI), and to objectively evaluate the performance of the prediction through multicenter validation.

METHODS

A total of 407 pregnant women from two centers undergoing preoperative magnetic resonance imaging (MRI) were retrospectively recruited. The patients from institution I were divided into a training cohort (n = 298) and a validation cohort (n = 75), while patients from institution II served as the external test cohort (n = 34). In this study, we built a clinical prediction model using patient clinical data, a radiomics model based on selected key features, and a deep learning model by mining deep semantic features. Based on this, we developed a combined model by ensembling the prediction results of the three models mentioned above to achieve prenatal prediction of PAS. The performance of these predictive models was evaluated with respect to discrimination, calibration, and clinical usefulness.

RESULTS

The combined model achieved AUCs of 0.872 (95% confidence interval, 0.843 to 0.908) in the validation cohort and 0.857 (0.808 to 0.894) in the external test cohort, both of which outperformed the other models. The calibration curves demonstrated excellent consistency in the validation cohort and the external test cohort, and the decision curves indicated high clinical usefulness.

CONCLUSION

By using preoperative clinical information and MRI images, the combined model can accurately predict PAS by ensembling clinical model, radiomics model, and deep learning model.

Diffusion and perfusion MRI parameters in the evaluation of placenta accreta spectrum disorders in patients with placenta previa

OBJECTIVES

To evaluate the placental function by monoexponential, biexponential, and diffusion kurtosis MR imaging (MRI) in patients with placenta previa.

METHODS

A total of 62 patients with placenta accreta spectrum (PAS) disorders and 11 patients with normal placentas were retrospectively enrolled, who underwent conventional diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI). The apparent diffusion coefficient (ADC) and exponential ADC (eADC) from standard DWI, mean kurtosis (MK), and diffusion coefficient (MD) from DKI, and pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) from IVIM were measured and compared from the volumetric analysis.

RESULTS

Comparisons between patients with PAS disorders and patients with normal placentas demonstrated that MD mean, D mean, and D* mean values in patients with PAS disorders were significantly higher than those in patients with normal placentas (p < 0.05). Comparisons between patients with accreta, increta, and percreta, and patients with normal placentas showed that the D mean was significantly higher in patients with placenta increta and percreta than in patients with normal placentas (p < 0.05).

CONCLUSION

The accreta lesions in PAS disorders had deceased cellularity and increased blood movement. The alteration of placental cellularity was more prominent in placenta increta and percreta.

MRI-radiomics-clinical-based nomogram for prenatal prediction of the placenta accreta spectrum disorders

OBJECTIVES

To investigate whether an MRI-radiomics-clinical-based nomogram can be used to prenatal predict the placenta accreta spectrum (PAS) disorders.

METHODS

The pelvic MR images and clinical data of 156 pregnant women with pathologic-proved PAS (PAS group) and 115 pregnant women with no PAS (non-PAS group) identified by clinical and prenatal ultrasonic examination were retrospectively collected from two centers. These pregnancies were divided into a training (n = 133), an independent validation (n = 57), and an external validation (n = 81) cohort. Radiomic features were extracted from images of transverse oblique T2-weighted imaging. A radiomics signature was constructed. A nomogram, composed of MRI morphological findings, radiomic features, and prenatal clinical characteristics, was developed. The discrimination and calibration of the nomogram were conducted to assess its performance.

RESULTS

A radiomics signature, including three PAS-related features, was associated with the presence of PAS in the three cohorts (p < 0.001 to p = 0.001). An MRI-radiomics-clinical nomogram incorporating radiomics signature, two prenatal clinical features, and two MRI morphological findings was developed, yielding a higher area under the curve (AUC) than that of the MRI morphological-determined PAS in the training cohort (0.89 vs. 0.78; p < 0.001) and external validation cohort (0.87 vs. 0.75; p = 0.003), while a comparable AUC value in the validation cohort (0.91 vs. 0.81; p = 0.09). The calibration was good.

CONCLUSIONS

An MRI-radiomics-clinical nomogram had a robust performance in antenatal predicting the PAS in pregnancies.

KEY POINTS

• An MRI-radiomics-clinical-based nomogram might serve as an adjunctive approach for the treatment decision-making in pregnancies suspicious of PAS. • The radiomic score provides a mathematical formula that predicts the possibility of PAS by using the MRI data, and pregnant women with PAS had higher radiomic scores than those without PAS.

Radiomics analysis of T2 -weighted images for differentiating invasive placentas in women at high risks

PURPOSE

To develop and validate an MRI-based radiomics model for differentiating invasive placentas in patients with high risks.

METHODS

A total of 181 pregnant women suspected of placenta accreta spectrum (PAS) disorders and who underwent MRI for placenta evaluation were retrospectively enrolled. The data set was randomly divided into the training (n = 125; invasive = 63, noninvasive = 62) and test (n = 56; invasive = 28, noninvasive = 28) groups. Radiomics features were extracted from half-Fourier acquisition single-shot turbo spin echo (HASTE) and sagittal true fast imaging in steady-state precession (TRUFISP) sequences independently and mainly selected based on their correlations with invasive placentas to construct two radiomics signatures including HASTE-Radscore and TRUFISP-Radscore. Then, the predictive performance of radiomic signatures, clinical features, radiographic features, and their combination were evaluated. The model with the best predictive performance was validated with its discrimination ability, calibration, and clinical usefulness.

RESULTS

Five radiomics features from HASTE and three radiomics features from TRUFISP were retained, respectively, for predicting invasive placentas. The combination of radiomics signatures and clinical features including prior cesarean delivery, placenta previa, and radiographic feature, the placental thickness resulted in the best discrimination ability, with area under the curve of 0.898 (95% confidence interval [CI] 0.844-0.9522) and 0.858 (95% confidence interval 0.7514-0.9655) in the training and test cohort, respectively. The combined model showed a significantly better area under the curve performance and clinical usefulness than independent clinical or radiographic model according to DeLong test (p < .05), net reclassification improvement and integrated discrimination improvement analysis (positive improvement) and decision curve analysis (higher net benefit).

CONCLUSIONS

The T₂ -weighted imaging MRI radiomics model could serve as a potential prenatal diagnosis tool for identifying invasive placentas in patients with high risks.

Placenta Accreta Spectrum Disorders and Radiomics: Systematic review and quality appraisal

PURPOSE

Ultrasound and magnetic resonance imaging are the imaging modalities of choice for placenta accrete spectrum (PAS) disorders assessment. Radiomics could further increase the value of medical images and allow to overcome the limitations linked to their visual assessment. Aim of this systematic review was to identify and appraise the methodological quality of radiomics studies focused PAS disorders applications.

METHOD

Three online databases (PubMed, Scopus and Web of Science) were searched to identify original research articles on human subjects published in English. For the qualitative synthesis of results, data regarding study design (e.g., retrospective or prospective), purpose, patient population (e.g., sample size), imaging modalities and radiomics pipelines (e.g., segmentation and feature extraction strategy) were collected. The appraisal of methodological quality was performed using the Radiomics Quality Score (RQS).

RESULTS

10 articles were finally included and analyzed. All were retrospective and MRI-powered. The majority included more than 100 patients (6/10). Four were prognostic (focused on either the prediction of bleeding volume or the prediction of needed management) while six diagnostic (PAS vs not PAS classification) studies. The median RQS was 8, with maximum and minimum respectively equal to 17/36 and - 6/36. Major methodological concerns were the lack of feature stability to multiple segmentation testing and poor data openness.

CONCLUSIONS

Radiomics studies focused on PAS disorders showed a heterogeneous methodological quality, overall lower than desirable. Furthermore, many relevant research questions remain unexplored. More robust investigations are needed to foster advancements in the field and possibly clinical translation.

Texture analysis of myometrium-derived T2WI in the evaluation of placenta increta: An observational retrospective study

INTRODUCTION

MRI has demonstrated its potential in the diagnosis of placenta percreta. Texture analysis is a novel technique to quantify tissue heterogeneity. The study aimed to evaluate the feasibility of using texture analysis based on myometrium-derived T2WI to differentiate placenta accreta from increta.

METHODS

Participants with MRI and clinical or histopathological diagnosis of placenta increta were retrospectively enrolled. Texture analysis of T2WI was implemented on normal myometrium and placenta increta by MaZda software. With the Fisher discriminant method, parameter selection and reduction were done automatically. Multivariate analysis was used for the comparison of response variables between two groups. The contours of multivariable average vectors were compared using profile analysis. Two-step clustering was performed to assess the importance of parameters.

RESULTS

There were a total of 23 participants (median age 29 years, range 22-43 years). The pixel intensity distribution was narrow and wide in two first-order histograms taken from normal myometrium and placenta increta, respectively. Multivariate analysis showed nine second-order parameters derived from the histogram were statistically significant (P < 0.05). The results of two-step clustering indicated that three second-order parameters (Mean, Percentile 90%, and Percentile 99%) were important (predictor importance > 0.8). Multivariate analysis of three second-order parameters further showed they were different between normal myometrium and placenta increta.

DISCUSSION

Texture analysis based on myometrium-derived T2WI may be a useful add-on to MRI in diagnosing placenta increta.

TRIAL REGISTRATION

Registration number: ChiCTR2000038604 and name of registry: Evaluation of diagnostic accuracy of MRI multi-parameter imaging combined with texture analysis for placenta accreta spectrum disorders (PAD).

Automated machine learning for differentiation of hepatocellular carcinoma from intrahepatic cholangiocarcinoma on multiphasic MRI

With modern management of primary liver cancer shifting towards non-invasive diagnostics, accurate tumor classification on medical imaging is increasingly critical for disease surveillance and appropriate targeting of therapy. Recent advancements in machine learning raise the possibility of automated tools that can accelerate workflow, enhance performance, and increase the accessibility of artificial intelligence to clinical researchers. We explore the use of an automated Tree-Based Optimization Tool that leverages a genetic programming algorithm for differentiation of the two common primary liver cancers on multiphasic MRI. Manual and automated analyses were performed to select an optimal machine learning model, with an accuracy of 73–75% (95% CI 0.59–0.85), sensitivity of 70–75% (95% CI 0.48–0.89), and specificity of 71–79% (95% CI 0.52–0.90) on manual optimization, and an accuracy of 73–75% (95% CI 0.59–0.85), sensitivity of 65–75% (95% CI 0.43–0.89) and specificity of 75–79% (95% CI 0.56–0.90) for automated machine learning. We found that automated machine learning performance was similar to that of manual optimization, and it could classify hepatocellular carcinoma and intrahepatic cholangiocarcinoma with an sensitivity and specificity comparable to that of radiologists. However, automated machine learning performance was poor on a subset of scans that met LI-RADS criteria for LR-M. Exploration of additional feature selection and classifier methods with automated machine learning to improve performance on LR-M cases as well as prospective validation in the clinical setting are needed prior to implementation.

Monoexponential, biexponential and diffusion kurtosis MR imaging models: quantitative biomarkers in the diagnosis of placenta accreta spectrum disorders

BACKGROUND

To investigate the diagnostic value of monoexponential, biexponential, and diffusion kurtosis MR imaging (MRI) in differentiating placenta accreta spectrum (PAS) disorders.

METHODS

A total of 65 patients with PAS disorders and 27 patients with normal placentas undergoing conventional DWI, IVIM, and DKI were retrospectively reviewed. The mean, minimum, and maximum parameters including the apparent diffusion coefficient (ADC) and exponential ADC (eADC) from standard DWI, diffusion kurtosis (MK), and mean diffusion coefficient (MD) from DKI and pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) from IVIM were measured from the volumetric analysis and compared between patients with PAS disorders and patients with normal placentas. Univariate and multivariated logistic regression analyses were used to evaluate the value of the above parameters for differentiating PAS disorders. Receiver operating characteristics (ROC) curve analyses were used to evaluate the diagnostic efficiency of different diffusion parameters for predicting PAS disorders.

RESULTS

Multivariate analysis demonstrated that only D mean and D max differed significantly among all the studied parameters for differentiating PAS disorders when comparisons between accreta lesions in patients with PAS (AP) and whole placentas in patients with normal placentas (WP-normal) were performed (all p < 0.05). For discriminating PAS disorders, a combined use of these two parameters yielded an AUC of 0.93 with sensitivity, specificity, and accuracy of 83.08, 88.89, and 83.70%, respectively.

CONCLUSION

The diagnostic performance of the parameters from accreta lesions was better than that of the whole placenta. D mean and D max were associated with PAS disorders.

Active Management of Labor Process under Smart Medical Model Improves Vaginal Delivery Outcomes of Pregnant Women with Preeclampsia

Background

In a global environment of increasing cesarean delivery rate, promoting vaginal delivery, reducing the rate of first cesarean section, and the incidence of vaginal delivery complications are the objectives of obstetric medical quality and safety in China. As a common obstetric complication, preeclampsia affects the safety of many pregnant women. It is the obstetrician's great responsibility to promote vaginal delivery and improve delivery outcomes in preeclampsia. To this end, we explored the roles of active labor management under the smart medical model in improving the outcomes of vaginal delivery for pregnant women with preeclampsia.

Methods

The clinical data of 219 cases of preeclampsia pregnant women who delivered vaginally in our hospital from January 2017 to December 2020 were retrospectively analyzed. According to different labor process management, they were divided into study group (active labor process management group) and control group (normal labor process management group). Active labor process management methods included intrapartum ultrasound, central fetal heart rate monitoring, Doula delivery, labor analgesia, and quality of life care. The differences in delivery process, delivery outcome, bleeding causes, and hemostatic measures were compared between the two groups.

Results

(1) The incidence of preeclampsia in our hospital showed an increasing trend in recent four years; (2) in smart hospitals, the active management of labor process reduced the probability of transferring to the cesarean section in preeclampsia pregnant women with vaginal trial failure; and (3) active labor process management reduced the rate of lateral episiotomy, decreased the postpartum hemorrhage volume within two hours, and improved the vaginal delivery outcome of preeclampsia pregnant women.

Conclusions

In the era of the rapid development of the Internet, vigorously promoting the construction of smart hospitals and actively managing the delivery process can reduce the failure rate of vaginal trial delivery and improve the outcomes of vaginal delivery in preeclampsia women.

Machine learning analysis for the noninvasive prediction of lymphovascular invasion in gastric cancer using PET/CT and enhanced CT-based radiomics and clinical variables

PURPOSE

Lymphovascular invasion (LVI) is associated with metastasis and poor survival in patients with gastric cancer, yet the noninvasive diagnosis of LVI is difficult. This study aims to develop predictive models using different machine learning (ML) classifiers based on both enhanced CT and PET/CT images and clinical variables for preoperatively predicting lymphovascular invasion (LVI) status of gastric cancer.

METHODS

A total of 101 patients with gastric cancer who underwent surgery were retrospectively recruited, and the LVI status was confirmed by pathological analysis. Patients were randomly divided into a training dataset (n = 76) and a validation dataset (n = 25). By 3D manual segmentation, radiomics features were extracted from the PET and venous phase CT images. Image models, clinical models, and combined models were constructed by selected enhanced CT-based and PET-based radiomics features, clinical factors, and a combination of both, respectively. Three ML classifiers including adaptive boosting (AdaBoost), linear discriminant analysis (LDA), and logistic regression (LR) were used for model development. The performance of these predictive models was evaluated with respect to discrimination, calibration, and clinical usefulness.

RESULTS

Ten radiomics features and eight clinical factors were selected for the development of predictive models. In the validation dataset, the area under curve (AUC) values of clinical models using AdaBoost, LDA, and LR classifiers were 0.742, 0.706, and 0.690, respectively. The image models using AdaBoost, LDA, and LR classifiers achieved an AUC of 0.849, 0.778, and 0.810, respectively. The combined models showed improved performance than the image models and the clinical models, with the AUC values of AdaBoost, LDA, and LR classifier yielding 0.944, 0.929, and 0.921, respectively. The combined models also showed good calibration and clinical usefulness for LVI prediction.

CONCLUSION

ML-based models integrating PET/CT and enhanced CT radiomics features and clinical factors have good discrimination capability, which could serve as a noninvasive, preoperative tool for the prediction of LVI and assist surgical treatment decisions in patients with gastric cancer.

A dual-path neural network fusing dual-sequence magnetic resonance image features for detection of placenta accrete spectrum (PAS) disorder

With the increase of various risk factors such as cesarean section and abortion, placenta accrete spectrum (PAS) disorder is happening more frequently year by year. Therefore, prenatal prediction of PAS is of crucial practical significance. Magnetic resonance imaging (MRI) quality will not be affected by fetal position, maternal size, amniotic fluid volume, etc., which has gradually become an important means for prenatal diagnosis of PAS. In clinical practice, T2-weighted imaging (T2WI) magnetic resonance (MR) images are used to reflect the placental signal and T1-weighted imaging (T1WI) MR images are used to reflect bleeding, both plays a key role in the diagnosis of PAS. However, it is difficult for traditional MR image analysis methods to extract multi-sequence MR image features simultaneously and assign corresponding weights to predict PAS according to their importance. To address this problem, we propose a dual-path neural network fused with a multi-head attention module to detect PAS. The model first uses a dual-path neural network to extract T2WI and T1WI MR image features separately, and then combines these features. The multi-head attention module learns multiple different attention weights to focus on different aspects of the placental image to generate highly discriminative final features. The experimental results on the dataset we constructed demonstrate a superior performance of the proposed method over state-of-the-art techniques in prenatal diagnosis of PAS. Specifically, the model we trained achieves 88.6% accuracy and 89.9% F1-score on the independent validation set, which shows a clear advantage over methods that only use a single sequence of MR images.

MRI-Based Radiomics Analysis for Intraoperative Risk Assessment in Gravid Patients at High Risk with Placenta Accreta Spectrum

Background: Gravid patients at high risk with placenta accreta spectrum (PAS) face life-threatening risk at delivery. Intraoperative risk assessment for patients is currently insufficient. We aimed to develop an assessment system of intraoperative risks through MRI-based radiomics. Methods: A total of 131 patients enrolled were randomly grouped according to a ratio of 7:3. Clinical data were analyzed retrospectively. Radiomic features were extracted from sagittal Fast Imaging Employing State-sate Acquisition images. Univariate and multivariate regression analyses were performed to build models using R software. A receiver operating characteristic curve and decision curve analysis (DCA) were performed to determine the predictive performance of models. Results: Six radiomic features and two clinical variables were used to construct the combined model for selection of removal protocols of the placenta, with an area under the curve (AUC) of 0.90 and 0.91 in the training and test cohorts, respectively. Nine radiomic features and two clinical variables were obtained to establish the combined model for prediction of intraoperative blood loss, with an AUC of 0.90 and 0.88 in the both cohorts, respectively. The DCA confirmed the clinical utility of the combined model. Conclusion: The analysis of combined MRI-based radiomics with clinics could be clinically beneficial for patients.

Using Machine Learning to Predict Complications in Pregnancy: A Systematic Review

Introduction: Artificial intelligence is widely used in medical field, and machine learning has been increasingly used in health care, prediction, and diagnosis and as a method of determining priority. Machine learning methods have been features of several tools in the fields of obstetrics and childcare. This present review aims to summarize the machine learning techniques to predict perinatal complications.

Objective: To identify the applicability and performance of machine learning methods used to identify pregnancy complications.

Methods: A total of 98 articles were obtained with the keywords “machine learning,” “deep learning,” “artificial intelligence,” and accordingly as they related to perinatal complications (“complications in pregnancy,” “pregnancy complications”) from three scientific databases: PubMed, Scopus, and Web of Science. These were managed on the Mendeley platform and classified using the PRISMA method.

Results: A total of 31 articles were selected after elimination according to inclusion and exclusion criteria. The features used to predict perinatal complications were primarily electronic medical records (48%), medical images (29%), and biological markers (19%), while 4% were based on other types of features, such as sensors and fetal heart rate. The main perinatal complications considered in the application of machine learning thus far are pre-eclampsia and prematurity. In the 31 studies, a total of sixteen complications were predicted. The main precision metric used is the AUC. The machine learning methods with the best results were the prediction of prematurity from medical images using the support vector machine technique, with an accuracy of 95.7%, and the prediction of neonatal mortality with the XGBoost technique, with 99.7% accuracy.

Conclusion: It is important to continue promoting this area of research and promote solutions with multicenter clinical applicability through machine learning to reduce perinatal complications. This systematic review contributes significantly to the specialized literature on artificial intelligence and women’s health.

Qualitative evaluation of MR images for assessing placenta accreta spectrum disorders in patients with placenta previa: A pilot validation study

PURPOSE

To validate a qualitative imaging method using magnetic resonance (MR) for predicting placental accreta spectrum (PAS) in patients with placenta previa (PP).

METHOD

Two MR imaging methods built in our previous experience was tested in an external comparable group of sixty-five patients with PP; these methods consisted of presence of at least one (Method 1) or two (Method 2) of the following abnormal MR imaging signs: intraplacental dark bands, focal interruption of myometrial border and abnormal placental vascularity. Three groups of radiologists with different level of expertise evaluated MR images: at least 5 years of experience in body imaging (Group 1); at least 10 (Group 2) or 20 (Group 3) years of experience in genito-urinary MR. While radiologists of Group 1 routinely evaluated MR images, those of Groups 2 and 3 used both Methods 1 and 2.

RESULTS

A significant (p < 0.005) difference was found between the diagnostic accuracy values of imaging evaluation performed by Group 3 using Method 1 (63%) and Method 2 (89%); of note, the accuracy of Method 2 by Group 3 was also significantly (p < 0.005) higher compared to that of both Methods 1 (46%) and 2 (63%) by Group 2 as well as to that of the routine evaluation by Group 1 (60%).

CONCLUSIONS

The qualitative identification of at least two abnormal MR signs (Method 2) represents an accurate method for predicting PAS in patients with PP particularly when this method was used by more experienced radiologists; thus, imaging expertise and methodology is required for this purpose.

Prediction of placenta accreta spectrum using texture analysis on coronal and sagittal T2-weighted imaging

PURPOSE

To separately perform visual and texture analyses of the axial, coronal, and sagittal planes of T2-weighted images and identify the optimal method for differentiating between the normal placenta and placenta accreta spectrum (PAS).

METHODS

Eighty consecutive patients (normal group, n = 50; PAS group, n = 30) underwent preoperative MRI. A scoring system (0-2) was used to evaluate the degree of abnormality observed in visual analysis (bulging, abnormal vascularity, T2 dark band, placental heterogeneity). The axial, coronal, and sagittal planes were manually segmented separately to obtain texture features, and seven combinations were obtained: axial; coronal; sagittal; axial and coronal; axial and sagittal; coronal and sagittal; and axial, coronal, and sagittal. Feature selection using the least absolute shrinkage and selection operator method and model construction using a support vector machine algorithm with k-fold cross-validation were performed. AUC was used to evaluate diagnostic performance.

RESULTS

The AUC of visual analysis was 0.75. The model 'coronal and sagittal' had the highest AUC (0.98) amongst the seven combinations. The fivefold cross-validation for the model 'coronal and sagittal' showed AUCs of 0.85 and 0.97 in training and validation sets, respectively. The AUC of the model 'coronal and sagittal' for all subjects was significantly higher than that of visual analysis (0.98 vs. 0.75; p < 0.0001).

CONCLUSION

The model 'coronal and sagittal' can accurately differentiate between the normal placenta and PAS, with a significantly better diagnostic performance than visual analysis. Texture analysis is an optimal method for differentiating between the normal placenta and PAS.

Emerging placental biomarkers of health and disease through advanced magnetic resonance imaging (MRI)

Placental dysfunction is a major cause of fetal demise, fetal growth restriction, and preterm birth, as well as significant maternal morbidity and mortality. Infant survivors of placental dysfunction are at elevatedrisk for lifelong neuropsychiatric morbidity. However, despite the significant consequences of placental disease, there are no clinical tools to directly and non-invasively assess and measure placental function in pregnancy. In this work, we will review advanced MRI techniques applied to the study of the in vivo human placenta in order to better detail placental structure, architecture, and function. We will discuss the potential of these measures to serve as optimal biomarkers of placental dysfunction and review the evidence of these tools in the discrimination of health and disease in pregnancy. Efforts to advance our understanding of in vivo placental development are necessary if we are to optimize healthy pregnancy outcomes and prevent brain injury in successive generations. Current management of many high-risk pregnancies cannot address placental maldevelopment or injury, given the standard tools available to clinicians. Once accurate biomarkers of placental development and function are constructed, the subsequent steps will be to introduce maternal and fetal therapeutics targeting at optimizing placental function. Applying these biomarkers in future studies will allow for real-time assessments of safety and efficacy of novel interventions aimed at improving maternal-fetal well-being.

Deep learning and radiomics analysis for prediction of placenta invasion based on T2WI

The purpose of this study was to explore whether the Nomogram, which was constructed by combining the Deep learning and Radiomic features of T2-weighted MR images with Clinical factors (NDRC), could accurately predict placenta invasion. This retrospective study included 72 pregnant women with pathologically confirmed placenta invasion and 40 pregnant women with normal placenta. After 24 gestational weeks, all participants underwent magnetic resonance imaging. The uterus and placenta regions were segmented in magnetic resonance images on sagittal T2WI. Ninety-three radiomics features were extracted from the placenta region, and 128 deep features were extracted from the uterus region using a deep neural network. The least absolute shrinkage and selection operator (LASSO) algorithm was used to filter these 221 features and to form the combined signature. Then the combined signature (CS) and clinical factors were combined to construct a nomogram. The accuracy, sensitivity, specificity and AUC of the nomogram were compared with four machine learning methods. The model NDRC was trained on the dataset of 78 pregnant women in the training cohort. Finally, the model NDRC was compared with four machine learning methods on the independent validation cohort of 34 pregnant women. The results showed that the prediction accuracy, sensitivity, specificity and AUC of the NDRC model were 0.941, 0.952, 0.923 and 0.985 respectively, which outperforms the traditional machine learning methods which rely on radiomics features and deep learning features alone.

Diffusion MRI Based Myometrium Tractography for Detection of Placenta Accreta Spectrum Disorder

BACKGROUND

Prenatal diagnosis of placenta accreta spectrum (PAS) disorders is difficult. Magnetic resonance imaging (MRI) has been shown to be a useful supplementary method to ultrasound.

PURPOSE

To investigate diffusion MRI (dMRI) based tractography as a tool for detecting PAS disorders, and to evaluate its performance compared with anatomical MRI.

STUDY TYPE

Prospective.

POPULATION

Forty-seven pregnant women in the third trimester with risk factors for PAS.

FIELD STRENGTH/SEQUENCE

Using fast imaging employing steady-state acquisition and high-angular resolution dMRI at 1.5 Tesla.

ASSESSMENT

Diagnosis of PAS was performed by three radiologists based on the dMRI-based feature of myometrial fiber discontinuity and on commonly used anatomical features including presence of dark band, discontinuous myometrium and bladder wall interruption. We evaluated the sensitivity, specificity, accuracy, and area-under-the-curve (AUC) of the individual features and established an integrated model with random forest analysis.

STATISTICAL TESTS

Maternal age and gestational age at scan were compared between PAS and control group using a t-test, and childbearing history was compared using a chi-squared test. The random forest model was employed to combine the anatomical and dMRI features with 5-fold cross-validation, and the weight of each feature was normalized to evaluate its importance in predicting PAS.

RESULTS

Based on surgical pathology reports, 16 out of 47 patients had confirmed PAS. The anatomical feature of dark bands and tractography marker achieved the highest AUC of 0.842 for predicting PAS, and the integrated anatomical and tractography features further improved the AUC of 0.880 with an accuracy of 87.2%. The tractography feature contributed most (30.1%) to the integrated model.

DATA CONCLUSION

Myometrial tractography demonstrated superior performance in detecting PAS. Moreover, the combination of dMRI-based tractography and anatomical MRI could potentially improve the diagnosis of PAS disorders in clinical practice.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY STAGE: 2.

Prenatal prediction and typing of placental invasion using MRI deep and radiomic features

BACKGROUND

To predict placental invasion (PI) and determine the subtype according to the degree of implantation, and to help physicians develop appropriate therapeutic measures, a prenatal prediction and typing of placental invasion method using MRI deep and radiomic features were proposed.

METHODS

The placental tissue of abdominal magnetic resonance (MR) image was segmented to form the regions of interest (ROI) using U-net. The radiomic features were subsequently extracted from ROI. Simultaneously, a deep dynamic convolution neural network (DDCNN) with codec structure was established, which was trained by an autoencoder model to extract the deep features from ROI. Finally, combining the radiomic features and deep features, a classifier based on the multi-layer perceptron model was designed. The classifier was trained to predict prenatal placental invasion as well as determine the invasion subtype.

RESULTS

The experimental results show that the average accuracy, sensitivity, and specificity of the proposed method are 0.877, 0.857, and 0.954 respectively, and the area under the ROC curve (AUC) is 0.904, which outperforms the traditional radiomic based auxiliary diagnostic methods.

CONCLUSIONS

This work not only labeled the placental tissue of MR image in pregnant women automatically but also realized the objective evaluation of placental invasion, thus providing a new approach for the prenatal diagnosis of placental invasion.

Development and validation of magnetic resonance imaging-based radiomics models for preoperative prediction of microsatellite instability in rectal cancer

BACKGROUND

Microsatellite instability (MSI) is a predictive biomarker for response to chemotherapy and a prognostic biomarker for clinical outcomes of rectal cancer. The purpose of this study was to develop and validate radiomics models for preoperative prediction of the MSI status of rectal cancer based on magnetic resonance (MR) images.

METHODS

This study retrospectively recruited 491 rectal cancer patients with pathologically confirmed MSI status. Patients were randomly divided into a training cohort (n=327) and a validation cohort (n=164). The most predictive radiomics features were selected using intraclass correlation coefficient (ICC) analysis, the two-sample t test, and the least absolute shrinkage and selection operator (LASSO) method. XGBoost models were constructed in the training cohort to discriminate the MSI status using clinical factors, radiomics features, or a combined model incorporating both the radiomics signature and independent clinical characteristics. The diagnostic performance of these three models was evaluated in the validation cohort based on their area under the curve (AUC), sensitivity, specificity, and accuracy.

RESULTS

Among the 491 rectal cancer patients, the prevalence of MSI was 10.39% (51/491). Following ICC analysis, two-sample t test, and LASSO regression, six radiomics features were selected for subsequent analysis. The combined model, which incorporated both the clinical factors and radiomics features achieved an AUC of 0.895 [95% confidence interval (CI), 0.838-0.938] in the validation cohort, and showed better performance in predicting MSI status than the other two models using either clinical factors (P=0.015) or radiomics features (P=0.204) alone.

CONCLUSIONS

Radiomics features based on preoperative T2-weighted MR imaging (MRI) are associated with the MSI status of rectal cancer. Combinational analysis of clinical factors and radiomics features may improve predictive performance and potentially contribute to noninvasive personalized therapy selection.

Functional diagnosis of placenta accreta by intravoxel incoherent motion model diffusion-weighted imaging

OBJECTIVES

To investigate the diagnostic value of intravoxel incoherent motion (IVIM) DWI for placenta accreta by comparing diffusion and perfusion characteristics of placentas with accreta lesions (APs) with those of normal placentas (NPs).

METHODS

Twenty-five pregnant women with AP and 24 with NP underwent 3-T magnetic resonance examinations with IVIM-DWI. The perfusion percentage (f), pseudo-diffusion coefficient (D*), and diffusion coefficient (D) values were calculated from different ROIs: the entire-plane of the AP (AP-ROI) and NP (NP-ROI) and the implanted (IR-ROI) and non-implanted region (NIR-ROI) of the AP. The AP-ROIs and NP-ROIs were compared using covariance analysis; the IR-ROIs and NIR-ROIs were compared using the Wilcoxon signed-rank test. ROC curves were produced to evaluate the parameters for predicting placenta accreta.

RESULTS

The f and D* values for the AP-ROIs ([45.0 ± 7.63]%, [11.64 ± 2.15]mm²/s) were significantly higher than those for the NP-ROIs ([31.85 ± 5.96]%, [9.04 ± 3.13]mm²/s) (both p < 0.05); the IR-ROIs (54.8%, 14.03 mm²/s) were also significantly higher than the NIR-ROIs (37.4%, 11.4 mm²/s) (both p < 0.05). No significant differences were found between the D values of the AP-ROIs and NP-ROIs (p > 0.05) or of the IR-ROIs and NIR-ROIs (p > 0.05). The areas under the curve for f and D* of the ROC curves were 0.93 and 0.79, respectively.

CONCLUSIONS

These results suggest that the IVIM parameters f and D* can be used to quantitatively evaluate the higher perfusion of AP when compared with NP. Furthermore, IVIM may be a useful functional diagnostic technique to predict placenta accreta.

KEY POINTS

• Intravoxel incoherent motion (IVIM) may be a useful diagnostic technique to quantitatively estimate the perfusion of the placenta. • The perfusion percentage (f) and pseudo-diffusion coefficient (D*) values differed significantly between placentas with accreta lesions and normal placentas. • ROC curves showed that perfusion percentage (f) and pseudo-diffusion coefficient (D*) values could accurately predict placenta accreta.

Grade Prediction of Bleeding Volume in Cesarean Section of Patients With Pernicious Placenta Previa Based on Deep Learning

In order to predict the amount of bleeding in the cesarean section of the patients with Pernicious Placenta Previa (PPP), this study proposed an automatic blood loss prediction method based on Magnetic Resonance Imaging (MRI) uterus image. Firstly, the DeepLab-V3 + network was used to segment the original MRI abdominal image to obtain the uterine region image. Then, the uterine region image and the corresponding blood loss data were trained by Visual Geometry Group Network-16 (VGGNet-16) network. The classification model of blood loss level was obtained. Using a dataset of 82 positive samples and 128 negative samples, the proposed method achieved accuracy, sensitivity and specificity of 75.61, 73.75, and 77.46% respectively. The experimental results showed that this method can not only automatically identify the uterine region of pregnant women, but also objectively determine the level of intraoperative bleeding. Therefore, this method has the potential to reduce the workload of the attending physician and improve the accuracy of experts' judgment on the level of bleeding during cesarean section, so as to select the corresponding hemostasis measures.