Understanding why that pairing of accuracy and smallness is rare requires understanding what CTG demands of both people and machines. In a typical examination, an external transducer picks up the fetal heartbeat through the mother’s abdominal wall while a second sensor records contraction pressure, producing a continuous, two-channel record that runs across extended stretches of labor. Any model that reads such a record must therefore handle a long physiological time series, not a snapshot. Obstetricians scan it for a layered set of cues: the baseline heart rate, its beat-to-beat variability, accelerations that signal a responsive fetus, and decelerations that can warn of oxygen deprivation. Frameworks such as the 2008 NICHD workshop report on electronic fetal monitoring and the FIGO consensus guidelines on intrapartum fetal monitoring published in 2015 codify this reading, yet the evidence base for doing it reliably has long been uneasy. Interpretation varies between observers, and computerized analysis has been tested at scale — most prominently in the INFANT randomized controlled trial of computerized fetal heart rate interpretation during labour, reported in The Lancet in 2017. Deep learning has since raised the stakes considerably; models such as DeepFHR, described in BMC Medical Informatics and Decision Making in 2019, showed that convolutional networks could predict fetal acidemia from heart-rate signals. Those systems, however, exposed a different bottleneck: they were built for servers, not delivery rooms.

The most powerful sequence models of the past decade, the transformer architectures introduced in the 2017 landmark paper “Attention Is All You Need,” learn long-range dependencies by letting every point in a series attend to every other point. That power carries a quadratic price: as the sequence lengthens, computation and memory balloon with the square of its length. Transformers have swept through medicine — surveys of the field chart their rapid adoption — and various “efficient transformer” designs have been proposed to trim the cost, but those remedies reduce rather than abolish the scaling problem. A CTG trace is effectively a long physiological sentence, and models adapted from large vision transformers are far too heavy for the constrained hardware of a maternity ward: bedside monitors, portable units and the low-power edge devices where fetal monitoring actually happens. The alternative CMA-Net builds on is a newer family of architectures known as state space models. Structured state space models, described by Albert Gu, Karan Goel and Christopher Ré in 2021, and Mamba, introduced by Gu and Tri Dao in 2023, compress the history of a sequence into a compact hidden state that updates as each new step arrives. Because the state, not the entire past, carries the memory, processing time grows linearly with sequence length. Mamba adds a “selective” mechanism, letting the model decide, step by step, which inputs to remember and which to discard — a property that matters when much of a clinical trace is noise.

CMA-Net stitches these ingredients into a pipeline tailored to the peculiar geometry of a CTG printout. The recording is treated as a two-dimensional image, and a DenseNet-121 backbone — the densely connected convolutional architecture introduced by Huang and colleagues in 2017 — extracts multi-scale spatial features, capturing both the broad contour of the trace and its fine texture. A Coordinate Attention module, adapted from an efficient mobile-network design published in 2021, then infuses those feature maps with positional information, so the network knows not only what a feature is but where it sits along the twin axes of time and signal amplitude. The pivotal structural move follows. Because a standard CTG image stacks the fetal heart rate band above the uterine contraction band, the researchers vertically decouple the feature maps into two independent streams, one devoted to the fetal heart rate, the other to contractions. Separating them lets each stream specialize in a physiologically distinct signal before the two are reunited, much as a clinician reads each channel on its own terms before integrating them into a single judgment.

Down each narrowed channel, parallel Mamba blocks take over, having first passed through an hourglass-shaped bottleneck that compresses the features and strips away redundancy. Running the two streams through parallel blocks preserves their separation through the deepest stage of the network, so contraction dynamics and heart-rate dynamics are modeled on their own terms before being fused. Each Mamba block sweeps its stream in linear time, using input-dependent selective state space dynamics to track how the signal evolves — the slow drift of a baseline, the fast oscillations of variability, the ramp of a contraction. The design’s signature mechanism is what the authors call cross-scale attention, which aligns the macroscopic layer of the trace, its baseline trends, with the microscopic layer, the moment-to-moment fluctuations riding on top of it. That pairing mirrors how clinicians actually read a cardiotocogram, where a normal baseline combined with absent variability tells a very different story from the same baseline paired with vigorous fluctuations. By forcing the model to relate the two scales within each stream, CMA-Net bakes a piece of obstetric reasoning directly into its architecture rather than hoping a generic network will discover it on its own.

Robustness to artifacts is enforced at the end of the pipeline. Once the two streams are concatenated, a module the authors named Selective Channel Adaptive Regulation, or SCAR, re-weights the combined feature channels, damping responses that originate in sensor corruption rather than physiology. The idea descends from channel-attention schemes such as squeeze-and-excitation networks and the convolutional block attention module, but here its assignment is closer to janitorial: artifacts — dropped signals, spurious spikes, interference — tend to activate particular channels in characteristic ways, and SCAR learns to quiet them. The payoff is a network whose confidence is anchored in the shape of the physiological signal instead of the noise riding on it, which is precisely what a monitor in a busy, imperfect ward requires. It is a small module with a large responsibility, and it is the reason the authors frame the system as robust rather than merely accurate.

The evaluation was built around clinical realism. The team trained and tested the network on 3,036 authentic clinical recordings — real hospital traces rather than simulations — assembled into a balanced dataset and assessed with five-fold cross-validation, so that no recording the model was graded on had been seen during training. Balance matters because imbalanced medical datasets notoriously tempt classifiers into ignoring the rare, dangerous class, a pitfall documented extensively in the machine learning literature; on skewed data, plain accuracy can flatter a model that misses exactly the cases that count, which is why evaluation on imbalanced problems leans on precision-recall analysis rather than accuracy alone. On this test bed, CMA-Net reached an AUC of 94.68 percent and an F1-score of 88.13 percent, with a sensitivity of 89.59 percent. Set against the generic visual models the authors benchmarked, the decisive feature was not raw accuracy but error geometry: CMA-Net effectively minimized false negatives — true cases of fetal compromise that slip past undetected — without adding computational burden. That asymmetry is deliberate and clinically vital, because in fetal surveillance a false alarm costs an extra examination while a missed warning can cost a baby its oxygen supply.

The engineering trade-off is the study’s quiet headline. Vision backbones commonly deployed in medical image analysis often carry tens or hundreds of millions of parameters; at 8.37 million, CMA-Net lands in territory that clinical edge devices can realistically host, raising the prospect of anomaly detection running locally on the monitor itself rather than on a distant server. The authors argue that this balance — high sensitivity, suppressed false negatives and modest compute — can widen the safety margin of fetal monitoring precisely where resources are tightest. The study behind the numbers was retrospective: the team analyzed anonymized clinical data under ethical approval from the institutional Ethics Committee of a tertiary Grade-A hospital in China, with the research conducted in accordance with the Declaration of Helsinki and informed consent waived because the records were historical and de-identified. The datasets cannot be shared publicly under patient-privacy and institutional data-protection policies, though the authors state they are available from the corresponding author on reasonable request with explicit Ethics Committee permission. The CMA-Net source code, they add, will be made publicly available upon acceptance of the manuscript. The work drew funding from the Scientific Research Special Project of the Sichuan Administration of Traditional Chinese Medicine, the Scientific Research Project of the Sichuan Maternal and Child Health Association and the Sichuan Provincial Innovation and Entrepreneurship Training Program for College Students.

None of this means an algorithm will be watching over every delivery room tomorrow. The results come from cross-validated analysis of a retrospective cohort, and the jump to prospective, real-time deployment — where artifacts arrive live and unscripted, and where a wrong answer carries immediate consequence — is the hurdle every clinical AI must clear. The paper nevertheless lands at a telling moment. State space models are spreading quickly through medicine, offering transformer-grade sequence understanding at a fraction of the computational bill, while the field confronts an awkward truth: the most accurate models are worthless if they cannot run where patients actually are. CMA-Net’s wager is that in fetal monitoring the decisive variables are not benchmark scores alone but the triad of sensitivity, robustness to noise and size, a combination that maps directly onto the reality of a labor ward. If that wager holds in prospective studies, the humble two-line trace that has watched over births for decades may gain a tireless, noise-resistant reader — one small enough to sit quietly beside every bed.

One key study presented during the conference involves the integration of artificial intelligence (AI) into the detection of congenital heart defects during prenatal ultrasounds. Researchers have long understood that congenital heart defects remain a leading cause of morbidity and mortality among infants. Unfortunately, the prenatal detection rates for these conditions remain alarmingly low despite advancements in ultrasound technology. This research emphasizes the role that cutting-edge AI systems can play in improving detection accuracy, thereby increasing the chances of timely interventions. Garnering attention is Jennifer Lam-Rachlin, MD, who will present findings that indicate AI significantly enhances the ability of both obstetricians and maternal-fetal medicine specialists to identify anomalies during routine ultrasounds.

As the research unfolds, attention is drawn to another intriguing study that investigates the relationship between adverse childhood experiences and spontaneous preterm birth among Black pregnant individuals. The Philadelphia Urban ACE survey serves as a pivotal tool for this investigation, expanding upon the traditional Kaiser ACE survey by addressing community-level factors affecting health outcomes. Researchers, led by Chelsea A. DeBolt, MD, MSCR, aim to establish essential links between childhood adversity and adverse pregnancy outcomes. This work stands as a testament to the broader social determinants of health that can influence maternal and fetal well-being, highlighting an urgent need for interdisciplinary approaches in prenatal care.

Mental health is another significant area of concern during complex pregnancies, and researchers are keenly exploring the emotional ramifications of therapeutic abortion. A study presented by Daniel Kuhr, MD, delves into the psychological health of patients undergoing procedures after 22 weeks of gestation. Addressing the mental health implications of such significant medical decisions is crucial for ensuring comprehensive prenatal care. The screening for conditions such as depression, anxiety, and post-traumatic stress highlights the need for ongoing support for patients navigating these challenging experiences.

The multi-faceted nature of predicting spontaneous preterm birth is also under scrutiny. While cervical length has previously been an established parameter, a team from Mount Sinai spearheaded a study examining neighborhood-level indices in predicting preterm births. Daniel Kuhr and co-author Nicola Tavella, MPH, aim to explore how broader environmental factors impact gestational outcomes in ways unmeasured by conventional metrics. Such investigations reinforce the notion that individual health is often inextricably linked to community health and social structures.

Highlighting the interplay between metabolic factors and gestational health, researchers are examining the impact of lipid profiles on perinatal morbidity. The pilot study led by Nicola Tavella seeks to position maternal lipid levels as critical biomarkers for assessing risks associated with complications during pregnancy. Lipids are integral to metabolic health, and understanding their influence could lead to more personalized care approaches.

The implications of antiretroviral therapy for HIV patients during pregnancy are critical as well. A significant body of research presented by Sara Edwards, MD, investigates the effects of various classes of antiretroviral medications on fetal biometric measurements. The findings aim to illuminate whether these medications, known to pose certain risks in adults, similarly affect the developing fetus. Ensuring that effective maternal HIV treatments do not translate into adverse outcomes for infants represents a complex challenge warranting careful examination.

As part of their ongoing exploration within the realms of ultrasound and genetics, Mount Sinai researchers are harnessing AI systems to enhance the detection and diagnosis of major congenital heart defects. An oral concurrent session will feature groundbreaking findings that indicate AI systems can significantly augment clinical ability in diagnosing concerning findings during fetal ultrasounds. The integration of technology here is aimed at amplifying early detection capabilities, which can make all the difference in planning appropriate interventions.

The meeting will also tackle the topic of surgical techniques and outcomes in scheduled cesarean deliveries. An important study looking at the effectiveness of barbed sutures versus traditional vicryl sutures aims to determine if surgical choices can impact blood loss and resultant complications. By investigating intraoperative factors, researchers seek to contribute to a body of knowledge that could help reduce delivery-related morbidity for mothers.

As researchers delve into the associations between unplanned cesarean deliveries and postpartum outcomes, one study led by Alexandra N. Mills, MD, investigates how the method of delivery influences issues such as postpartum depression and engagement in subsequent medical care. Given that postpartum depression affects a significant percentage of new mothers, understanding these associations is crucial for developing targeted interventions that can improve maternal well-being.

The repercussions of gestational diabetes remain a critical topic, especially through the lens of the built environment. Research concerning the Environmental Protection Agency’s national walkability index aims to assess whether neighborhood walkability affects weight gain during pregnancies complicated by gestational diabetes. As obesity remains a significant risk factor for adverse perinatal outcomes, these studies seek to explore the intersection of community design, lifestyle factors, and pregnancy-related complications.

During the conference, the effect of dietary and lifestyle factors on hypertensive disorders in pregnancy will be discussed. By examining periconceptional diets and sleep quality, researchers aim to build a clearer understanding of how lifestyle management can help mitigate the incidence of hypertension during pregnancy. With the rates of hypertensive disorders rising nationally, effective intervention strategies grounded in sound research are critical.

In summary, the breadth of research being presented at this year’s SMFM meeting is impressive, touching on a wide array of topics central to maternal-fetal medicine. By integrating diverse fields such as artificial intelligence, public health, mental health, and community dynamics into their work, the Mount Sinai researchers underscore the complexity of addressing threats to maternal and fetal health. Their rigorous investigations reflect the ongoing commitment to improving prenatal outcomes and enhancing the quality of care for expectant mothers across varied contexts.

This gathering of experts not only advances the scientific community’s understanding of key issues surrounding high-risk pregnancies but also emphasizes the importance of collaborative approaches in response to the diverse needs of mothers and their babies. It illustrates a vital ongoing dialogue aimed at understanding the multi-dimensional nature of pregnancy and childbirth in a rapidly evolving healthcare landscape.

Subject of Research: The impact of AI and community factors on prenatal care and health outcomes in high-risk pregnancies.

Article Title: Innovative Research in Maternal-Fetal Medicine: Advancements from Mount Sinai at the 2025 SMFM Conference

News Publication Date: TBD

Web References: TBD

References: TBD

Image Credits: TBD

Keywords: maternal-fetal medicine, high-risk pregnancy, artificial intelligence, mental health, community health, gestational diabetes, congenital heart defects, preterm birth, obstetrics.