AI-PLAX: AI-based placental assessment and examination using photos

Assessment of an AI-based tool for population-wide collection of placental morphological data

OBJECTIVES

Automated placental assessment could allow accurate and timely morphological/pathological measurements at scale. We undertook a pilot study using an artificial intelligence-based assessment system (AI-PLAX) to ascertain the potential of a state-wide rollout as part of Generation Victoria, assessing the impact of time post-delivery, user, and technology used for image capture, on a range of derived placental data.

STUDY DESIGN

Ten placentas were imaged by three different users and imaging technologies (iPad, iPhone, Samsung) at (0 h), 24 h, and 48 h post-delivery. Using AI-PLAX, disc size (short and long length, perimeter, area), shape (normal, abnormal), cord insertion type (central, eccentric), cord coiling, abruption (retroplacental hematoma), and meconium staining were determined.

RESULTS

When analysing the maternal surface of the placenta, time in cold storage post-delivery had modest effects on placental dimensions, with decreases in the short length (24-48 h: -3.7 %), disc area (0-24 h: 4.7 % and 0-48 h: -7.4 %), and perimeter (0-48 h: -3.8 %) observed. There was marginal impact on placental dimensions when the placenta was imaged by different users, including long length (+1.9 %), disc area (+2.9 %), and perimeter (+2.0 %). Measures of placental size were not impacted by the type of technology used to capture the images. When analysing the fetal surface of the placenta, more variance in placental size measures were observed between users. Abruption detection was not affected by any parameter. Time between delivery and imaging impacted apparent meconium staining - likely reflecting changes in fetal surface colour over time. Meconium staining was not affected by technology or user.

CONCLUSIONS

This study supports the feasibility of the collection of placenta images for later morphological analysis by AI-PLAX, with measures obtained minimally influenced by time in cold storage, user imaging the placenta, or technology to capture the images.

Placental pathology reports: A qualitative study in a US university hospital setting on perceived clinical utility and areas for improvement

OBJECTIVE

To explore how placental pathology is currently used by clinicians and what placental information would be most useful in the immediate hours after delivery.

STUDY DESIGN

We used a qualitative study design to conduct in-depth, semi-structured interviews with obstetric and neonatal clinicians who provide delivery or postpartum care at an academic medical center in the US (n = 19). Interviews were transcribed and analyzed using descriptive content analysis.

RESULTS

Clinicians valued placental pathology information yet cited multiple barriers that prevent the consistent use of pathology. Four main themes were identified. First, the placenta is sent to pathology for consistent reasons, however, the pathology report is accessed by clinicians inconsistently due to key barriers: difficult to find in the electronic medical record, understand, and get quickly. Second, clinicians value placental pathology for explanatory capability as well as for contributions to current and future care, particularly when there is fetal growth restriction, stillbirth, or antibiotic use. Third, a rapid placental exam (specifically including placental weight, infection, infarction, and overall assessment) would be helpful in providing clinical care. Fourth, placental pathology reports that connect clinically relevant findings (similar to radiology) and that are written with plain, standardized language and that non-pathologists can more readily understand are preferred.

CONCLUSION

Placental pathology is important to clinicians that care for mothers and newborns (particularly those that are critically ill) after birth, yet many problems stand in the way of its usefulness. Hospital administrators, perinatal pathologists, and clinicians should work together to improve access to and contents of reports. Support for new methods to provide quick placenta information is warranted.

Vision for Improving Pregnancy Health: Innovation and the Future of Pregnancy Research

Understanding, predicting, and preventing pregnancy disorders have been a major research target. Nonetheless, the lack of progress is illustrated by research results related to preeclampsia and other hypertensive pregnancy disorders. These remain a major cause of maternal and infant mortality worldwide. There is a general consensus that the rate of progress toward understanding pregnancy disorders lags behind progress in other aspects of human health. In this presentation, we advance an explanation for this failure and suggest solutions. We propose that progress has been impeded by narrowly focused research training and limited imagination and innovation, resulting in the failure to think beyond conventional research approaches and analytical strategies. Investigations have been largely limited to hypothesis-generating approaches constrained by attempts to force poorly defined complex disorders into a single "unifying" hypothesis. Future progress could be accelerated by rethinking this approach. We advise taking advantage of innovative approaches that will generate new research strategies for investigating pregnancy abnormalities. Studies should begin before conception, assessing pregnancy longitudinally, before, during, and after pregnancy. Pregnancy disorders should be defined by pathophysiology rather than phenotype, and state of the art agnostic assessment of data should be adopted to generate new ideas. Taking advantage of new approaches mandates emphasizing innovation, inclusion of large datasets, and use of state of the art experimental and analytical techniques. A revolution in understanding pregnancy-associated disorders will depend on networks of scientists who are driven by an intense biological curiosity, a team spirit, and the tools to make new discoveries.

Proposing a care practice bundle for neonatal encephalopathy during therapeutic hypothermia

Neonates with neonatal encephalopathy (NE) often present with multi-organ dysfunction that requires multidisciplinary specialized management. Care of the neonate with NE is thus complex with interaction between the brain and various organ systems. Illness severity during the first days of birth, and not only during the initial hypoxia-ischemia event, is a significant predictor of adverse outcomes in neonates with NE treated with therapeutic hypothermia (TH). We thus propose a care practice bundle dedicated to support the injured neonatal brain that is based on the current best evidence for each organ system. The impact of using such bundle on outcomes in NE remains to be demonstrated.

Placental contribution to neonatal encephalopathy

Placental assessment, although currently underused, can inform our understanding of the etiology and timing of Neonatal Encephalopathy (NE). We review our current understanding of the links between placental dysfunction and NE and how this information may inform clinical decisions, now and in the future, emphasizing the four major placental lesions associated with NE. In addition, we discuss maternal and fetal factors that are hypothesized to contribute to specific placental pathologies, especially innate or acquired thrombophilias. We outline the importance of assessing placenta across trimesters and after delivery. As this field continues to evolve, currently available placental histopathological examination methods may need to be combined with advanced prenatal molecular and imaging assessments of placenta and be applied in well-designed studies in large representative populations to better define the links between placental dysfunction and NE.

Perinatal asphyxia from the obstetric standpoint

Perinatal asphyxia remains one of the major causes of morbidity and mortality for term newborns. Though access to health care and birth attendants have decreased the rate, Neonatal encephalopathy (NE) has not been eliminated. Worldwide, women at socioeconomic disadvantage have the highest risk of delivering a neonate with NE. Neonates that will experience perinatal asphyxia cannot be easily identified prospectively and the intrapartum testing available is not specific enough to clearly indicate the best course of action in most cases. Despite this, training programs that aim to decrease morbidity and mortality from all causes appear to be associated with fewer cases of perinatal asphyxia. The current best approach is to support education and communication for all people involved in the care of birthing women. Ideally, new technology will address identification of the fetus likely to be affected or the fetus who is beginning to experience injury in advance of delivery.

Influence of the placental cord insertion site on the placental mass and the birth weight in dichorionic diamniotic twins

The aim of the study was to determine the effect of the displacement of the umbilical cord insertion site from the centre of the placenta on the placental mass and the birth weight of dichorionic diamniotic twins and to consider the importance of the direction of the displacement, as well as to assess the influence of the umbilical cord displacement on the placental mass and the birth weight of dichorionic diamniotic twins taking into account the direction of displacement. Material and methods. The study was performed on 135 dichorionic diamniotic pairs: 68 opposite-sex, 35 same-sex males, and 32 same-sex females. The impact of an absolute cord displacement from the centroid of the placental disc and the direction of its shifting were compared with the birth weight and the placental mass. Results. In the investigated group, a central insertion was revealed in 6 (2.2 %), eccentric – in 224 (83.0 %), marginal – in 31 (11.5 %), and velamentous – in 9 (3.3 %) cases. The first two types of cord insertion are considered to be normal, the third and the fourth are seen as abnormal. The placental mass was in a strong positive correlation with the birth weight (r=0.71, p<0.0001). The placentas with an eccentric cord insertion had a smaller surface area. A negative correlation was established between the displacement of the cord insertion site and the placental mass (r=-0.4284, p<0.0001) as well as the birth weight (r=-0.6115, p<0.0001). The shift along the long axis was of greater importance than in relation to the shorter one. The placental mass and the birth weight were higher in the new-borns with a normal cord insertion site. In the abnormal cord insertion group, 32.5 % of the infants were under the 10th birth weight percentile, in the normal cord insertion group – only 8.3 %. Conclusions. The birth weight of dichorionic diamniotic twins and their placental mass are negatively correlated with the distance of the umbilical cord insertion site from the placental centre. The insertion site displacement along the long axis has a stronger negative effect on the birth weight and the placental mass in comparison with the shifting along the short axis. The placentas with an eccentric cord insertion have a smaller surface area.

Chorioamnionitis and Risk for Maternal and Neonatal Sepsis: A Systematic Review and Meta-analysis

OBJECTIVE

To estimate the risk of maternal and neonatal sepsis associated with chorioamnionitis.

DATA SOURCES

PubMed, BIOSIS, and ClinicalTrials.gov databases were systematically searched for full-text articles in English from inception until May 11, 2020.

METHODS OF STUDY SELECTION

We screened 1,251 studies. Randomized controlled trials, case-control, or cohort studies quantifying a relationship between chorioamnionitis and sepsis in mothers (postpartum) or neonates born at greater than 22 weeks of gestation were eligible. Studies were grouped for meta-analyses according to exposures of histologic or clinical chorioamnionitis and outcomes of maternal or neonatal sepsis.

TABULATION, INTEGRATION, AND RESULTS

One hundred three studies were included, and 55 met criteria for meta-analysis (39 studies of preterm neonates, 10 studies of general populations of preterm and term neonates, and six studies of late preterm and term neonates). Study details and quantitative data were abstracted. Random-effects models were used to generate pooled odds ratios (ORs); most studies only reported unadjusted results. Histologic chorioamnionitis was associated with confirmed and any early-onset neonatal sepsis (unadjusted pooled ORs 4.42 [95% CI 2.68-7.29] and 5.88 [95% CI 3.68-9.41], respectively). Clinical chorioamnionitis was also associated with confirmed and any early-onset neonatal sepsis (unadjusted pooled ORs 6.82 [95% CI 4.93-9.45] and 3.90 [95% CI 2.74-5.55], respectively). Additionally, histologic and clinical chorioamnionitis were each associated with higher odds of late-onset sepsis in preterm neonates. Confirmed sepsis incidence was 7% (early-onset) and 22% (late-onset) for histologic and 6% (early-onset) and 26% (late-onset) for clinical chorioamnionitis-exposed neonates. Three studies evaluated chorioamnionitis and maternal sepsis and were inconclusive.

CONCLUSION

Both histologic and clinical chorioamnionitis were associated with early- and late-onset sepsis in neonates. Overall, our findings support current guidelines for preventative neonatal care. There was insufficient evidence to determine the association between chorioamnionitis and maternal sepsis.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO, CRD42020156812.

How often is the placenta included in human pregnancy research? A rapid systematic review of the literature

Background: The placenta is a complex organ that plays a vital role not only in nutrient transfer but also in directing maternal and fetal physiological processes across pregnancy. Due to its multi-functionality, assessing the placenta can provide critical information about maternal and child health and risks of adverse outcomes. Objective: We aimed to quantify the percentage of human pregnancy studies that include placenta data. Methods: We conducted a rapid review of pregnancy studies conducted in the US that were published as original research in PubMed in 2018. Human studies conducted during the second trimester, third trimester, or labor and/or delivery were eligible. The systematic search produced 1,448 publications. After screening and full article review, 290 studies met all eligibility criteria. We then extracted data on study design, reporting of placenta data, time and type of data collection, and study objective categorization. Results: In total, 32% of studies were randomized controlled trials; the remaining were observational studies. Only 14% included placenta data of any kind. A total of 10% included placenta data during pregnancy and 7% included data after delivery; only 2% included both. Most data during pregnancy were collected by ultrasound and most data on the delivered placenta were from pathology exams. Study objectives were focused on maternal and/or infant outcomes (99.7%), while only one study had a placenta outcome. Conclusion: Based on this rapid review, a small proportion of pregnancy studies use placenta data in research. The placenta, an essential component of understanding healthy or adverse outcomes, deserves much more attention in pregnancy research.

Multi-region saliency-aware learning for cross-domain placenta image segmentation

We propose a multi-region saliency-aware learning (MSL) method for cross-domain placenta image segmentation. Unlike most existing image-level transfer learning methods that fail to preserve the semantics of paired regions, our MSL incorporates the attention mechanism and a saliency constraint into the adversarial translation process, which can realize multi-region mappings in the semantic level. Specifically, the built-in attention module serves to detect the most discriminative semantic regions that the generator should focus on. Then we use the attention consistency as another guidance for retaining semantics after translation. Furthermore, we exploit the specially designed saliency-consistent constraint to enforce the semantic consistency by requiring the saliency regions unchanged. We conduct experiments using two real-world placenta datasets we have collected. We examine the efficacy of this approach in (1) segmentation and (2) prediction of the placental diagnoses of fetal and maternal inflammatory response (FIR, MIR). Experimental results show the superiority of the proposed approach over the state of the art.