Validation of a computerized algorithm to quantify fetal heart rate deceleration area

Short-term variation of the fetal heart rate as a marker of intraamniotic infection in pregnancies with preterm prelabor rupture of membranes: a historical cohort study

INTRODUCTION

Intraamniotic infection (IAI) and subsequent early-onset neonatal sepsis (EONS) are among the main complications associated with preterm prelabor rupture of membranes (PPROM). Currently used diagnostic tools have been shown to have poor diagnostic performance for IAI. This study aimed to investigate whether the exposure to IAI before delivery is associated with short-term variation of the fetal heart rate in pregnancies with PPROM.

METHODS

Observational cohort study of 678 pregnancies with PPROM, delivering between 24 + 0 and 33 + 6 gestational weeks from 2012 to 2019 in five labor units in Stockholm County, Sweden. Electronic medical records were examined to obtain background and exposure data. For the exposure IAI, we used the later diagnosis of EONS in the offspring as a proxy. EONS is strongly associated to IAI and was considered a better proxy for IAI than the histological diagnosis of acute chorioamnionitis, since acute chorioamnionitis can be observed in the absence of both positive microbiology and biochemical markers for inflammation. Cardiotocography traces were analyzed by a computerized algorithm for short-term variation of the fetal heart rate, which was the main outcome measure.

RESULTS

Twenty-seven pregnancies were categorized as having an IAI, based on the proxy diagnosis of EONS after birth. Fetuses exposed to IAI had significantly lower short-term variation values in the last cardiotocography trace before birth than fetuses who were not exposed (5.25 vs 6.62 ms; unadjusted difference: -1.37, p = 0.009). After adjustment for smoking and diabetes, this difference remained significant. IAI with a later positive blood culture in the neonate (n = 12) showed an even larger absolute difference in STV (-1.65; p = 0.034), with a relative decrease of 23.5%.

CONCLUSION

In pregnancies with PPROM, fetuses exposed to IAI with EONS as a proxy have lower short-term variation of the fetal heart rate than fetuses who are not exposed. Short-term variation might be useful as adjunct surveillance in pregnancies with PPROM.

The Correlation between Fetal Monitor Decelerations Area and Neonatal Cord pH among Preeclamptic parturient: an Automated Approach to a Clinical Challenge

OBJECTIVE

 To examine whether a correlation exists between electronic fetal monitoring (EFM) in the final 120 minute prior to delivery and neonatal cord blood pH among parturients with preeclampsia.

STUDY DESIGN

 A retrospective cohort study of parturients with term and near-term preeclampsia who delivered at Soroka University Medical Center between 2013 and 2020 was conducted. EFM tracings, demographical data, and obstetrical outcomes were retrieved from the patients' electronic medical records. Using MATLAB program, area under the curve (AUC) measurements were applied. Correlation was calculated using the Spearman's correlation coefficient.

RESULTS

 EFM of 88 parturients with preeclampsia were assessed. A significant negative correlation was demonstrated between cord blood pH and the decelerations AUC, total decelerations duration, and deceleration depth. No correlation was demonstrated between area over the curve (acceleration) and cord blood pH. In a multivariate linear regression, total deceleration AUC was found independently negatively associated with cord blood pH (beta = -0.264, p = 0.011) controlling for second stage of labor duration and induction of labor.

CONCLUSION

 Among parturients with preeclampsia, EFM indices may assist in the diagnosis of fetal acidemia. Further studies are needed to strengthen and establish the use of electronic EFM in the management of labor and delivery.

KEY POINTS

· A negative correlation was demonstrated between cord blood pH and the decelerations AUC.. · Total deceleration AUC was found independently negatively associated.. · Among parturients with preeclampsia, EFM indices may assist in the diagnosis of fetal acidemia..

Correlation between total deceleration area in CTG records and cord blood pH in pregnancies with IUGR

PURPOSE

Fetal cardiotocography is the most common method to assess fetal well-being during labor. Nevertheless, its predictive ability for acidemia is limited, both in low-risk and high-risk pregnancies (Nelson et al. in N Engl J Med 334: 613-9, 1996; Rinciples P et al. in Health and Human Development Workshop Report on Electronic Fetal Monitoring : Update on Definitions. no. 2007, 510-515, 2008), especially in high-risk pregnancies, such as those complicated by growth restriction. In this study we aim examine the association between deceleration and acceleration areas and other measure of fetal heart rate in intrapartum fetal monitoring and neonatal arterial cord blood pH in pregnancies complicated by growth restriction.

MATERIALS AND METHODS

A retrospective cohort study of 100 deliveries complicated by growth restriction, delivered during 2018, was conducted. Known major fetal anomalies, non-vertex presentation and elective cesarean deliveries were excluded. Total deceleration and acceleration areas were calculated as the sum of the areas within the deceleration and acceleration, respectively.

RESULTS

In deliveries complicated by growth restriction, cord blood pH is significantly associated with total deceleration area (p = 0.05) and correlates with cumulative duration of the decelerations (Spearman's rank -0.363, p < 0.05), and total acceleration area (-0.358, p < 0.05). By comparing the cord blood pH in deliveries with a total deceleration area that was above and below the median total deceleration area, we demonstrated a significant difference between the categories.

CONCLUSIONS

Cord blood pH significantly correlates with total deceleration area and other fetal monitoring characteristics in neonates with growth restriction. Future studies using real-time, machine-learning based techniques of fetal heart rate monitoring, may provide population specific threshold values that will support bedside clinical decision making and perhaps achieve better outcomes.

Artificial intelligence and machine learning in cardiotocography: A scoping review

INTRODUCTION

Artificial intelligence (AI) is gaining more interest in the field of medicine due to its capacity to learn patterns directly from data. This becomes interesting for the field of cardiotocography (CTG) interpretation, since it promises to remove existing biases and improve the well-known issues of inter- and intra-observer variability.

MATERIAL AND METHODS

The objective of this study was to map current knowledge in AI-assisted interpretation of CTG tracings and thus, to present different approaches with their strengths, gaps, and limitations. The search was performed on Ovid Medline and PubMed databases. The Preferred Reporting Items for Systematic Reviews and meta-Analysis for Scoping Reviews (PRISMA-ScR) guidelines were followed.

RESULTS

We summarized 40 different studies investigating at least one algorithm or system to classify CTG tracings. In addition, the Oxford Sonicaid system is presented because of its wide use in clinical practice.

CONCLUSIONS

There are several promising approaches in this area, but none of them has gained big acceptance in clinical practice. Further investigation and refinement of the algorithms and features are needed to achieve a validated decision-support system. For this purpose, larger quantities of curated and labeled data may be necessary.

Intrapartum deceleration and acceleration areas are associated with neonatal encephalopathy

OBJECTIVE

To investigate whether an association exists between deceleration and acceleration areas on continuous fetal cardiotocography (CTG) and neonatal encephalopathy (NE).

METHODS

A single center, retrospective case-control study was conducted to compare CTG characteristics of low-risk pregnancies (35 weeks of gestation or more), complicated by moderate to severe NE with two matched controls for every case. Controls were matched by gestational age and cord blood pH. We analyzed the intrapartum CTG recordings by calculation of the deceleration and acceleration areas and the ratio between the two.

RESULTS

During the period between 2013 and 2019, we identified 95 cases of low-risk pregnancies that were complicated by moderate to severe NE in our center. Thirty-three (34.7%) deliveries were excluded, mostly because of an insufficient duration of the CTG recordings. The remaining 62 cases were matched with 123 controls. We found that NE was significantly associated with an increased total deceleration area, a decreased total acceleration area, and a lower acceleration-to-deceleration ratio.

CONCLUSIONS

NE was significantly associated with increased total deceleration area, decreased total acceleration area, and a lower acceleration-to-deceleration ratio, independent of cord blood pH. Development of a computerized real-time analysis of fetal heart rate tracings may contribute to making these measurements a more valid clinical tool.

Appraisal of trimester-specific fetal heart rate and its role in gestational age prediction

INTRODUCTION

The aim of this study was to evaluate and report normal sonographic FHR values among low-risk singleton women across the three trimesters of pregnancy and determine FHR role in gestational age prediction.

METHOD

A prospective cross-sectional study of 2727 low-risk singleton pregnant women was undertaken. FHR measurements were obtained by a consultant radiologist and three experienced sonographers using transabdominal approach from January 2019 to December 2020. Two FHR measurements were taken for each participant. The fetal lie and presentation were also documented in the first trimester. Data were analysed using SPSS version 24 (IBM, Armonk, NY, USA).

RESULT

The maternal mean ± SD age was 25.8 ± 6.5 years and mean FHR for first, second and third trimesters were 151 ± 16, 145 ± 6 and 125±6 bpm respectively. The mean ± SD gestational age were 10 ± 2, 19 ± 3 and 34 ± 2 weeks for the first, second and third trimester respectively. Using ANOVA, there were statistically significant differences in FHR across the three trimesters (p ≤ 0.05). A positive correlation existed between maternal age and FHR (r = 0.57, p ≤ 0.05).

CONCLUSION

This study has established normal values for FHR in first, second and third trimester respectively. Referring physicians, radiologists, sonographers, obstetricians and gynaecologists may consider FHR of (135-167) bpm (139-151) bpm and (119-131) bpm as normal FHR ranges for the first, second and third trimester respectively. This study has also revealed the possibility of gestational age prediction using FHR with the equation [Gestational Age = 87.8 - (0.47) FHR].

IMPLICATIONS FOR PRACTICE

This paper provides the most up-to-date sonographic FHR recommendations for fetal management. More importantly, findings from this study also suggests that ultrasound practitioners can use FHR measurements as a reliable alternative for fetal dating.

On designing a biosignal-based fetal state assessment system: A systematic mapping study

BACKGROUND AND OBJECTIVE

The patterns present in biosignals, such as fetal heart rate (FHR), are valuable indicators of fetal well-being. In designing biosignal analysis systems, the variety of approaches and technology usage impairs the decision-making for the fundamental units of the systems. There is a need for an updated overview of studies encompassing the biosignal-based fetal state assessment systems. Therefore, we propose a systematic mapping study to identify and synthesize recent research regarding the building blocks that compose these systems.

METHODS

We followed well-established guidelines to perform a systematic mapping of studies regarding the building-blocks that compose the fetal state assessment systems and published between January 2016 and January 2021. A search string was determined based on the mapping questions and the PI (population and intervention) divisions. The search string was applied in digital libraries covering the fields of computer science, engineering, and medical informatics. Then, we applied the forward snowballing technique to complement the resulting set. This process resulted in 75 primary studies selected from a total of 871 papers.

RESULTS

Selected studies were classified according to the publication types, systems design stages, datasets, and predictive capabilities. The results revealed that (i) The majority of the selected studies refer to the method as a type of publication and there is a lack of validation studies; (ii) The CTU-UHB was the most frequent biosignal-based dataset and UCI-CTG was the most frequent feature-based data; (iii) The selected studies made use of the system design stages alone or in a mixed-mode.

CONCLUSION

The results indicated that the well-established classification models achieved competitive results compared with the state-of-the-art methods in data-constrained system designs. Moreover, we identified the need for validation studies in the clinical environment.

Variable deceleration features and intrapartum fetal acidemia - The role of deceleration area

OBJECTIVE

In recent years deceleration area has received increasing attention as a potential predictor of intrapartum fetal hypoxia. In several studies, the area has been manually esteemed as triangular in shape, which might introduce bias. In addition, the commonly used outcome acidemia in umbilical artery is affected by mode of delivery. We wanted to investigate the association between the variable deceleration features area, duration, depth and cumulative deceleration area (30 and 60 min) and intrapartum fetal acidemia measured as lactate concentration at fetal scalp blood sampling (FBS) in immediate connection to the decelerations.

STUDY DESIGN

In the source population of 1070 labors at Karolinska University Hospital, Sweden, with FBS performed on indication, we found 507 fetuses with predominantly variable decelerations as the indication for FBS. We examined the last 60- and 30-minutes of fetal monitoring preceding the FBS with focus on deceleration area, duration and depth. The contours of the decelerations were outlined manually but the area was calculated with a computer software program. We assessed area, duration and depth both as mean values per deceleration and as cumulative values during the time period of interest. We analyzed Pearson correlations and area under receiver operating characteristics curves (AUC). We also performed an adjusted analysis, with baseline frequency, variability, and accelerations as covariates.

RESULTS

Deceleration area and duration were the best predictors of intrapartum fetal acidemia (fetal lactate concentration >4.8 mmol/L measured with Lactate Pro™) with AUCs of 0.671 (0.682) and 0.678 (0.683) for cumulative measures during 30 (60) minutes prior to FBS, compared to deceleration depth with AUC of 0.632 (0.631). Corresponding Pearson correlations in 30-min (60-min) groups were 0.329 (0.335) and 0.358 (0.354) for deceleration area and duration and 0.212 (0.204) for deceleration depth. Using 250 beats cumulative cut-off for deceleration area during last 30 min, 71% vs. 43% were acidemic and non-acidemic, odds ratio = 3.2 (95% CI 1.7-6.1).

CONCLUSIONS

Deceleration area and duration were better predictors of intrapartum fetal acidemia than deceleration depth. Cumulative deceleration area >250 beats during 30 min was associated with three-fold higher odds of intrapartum acidemia compared to <250 beats.

The Use of Artificial Intelligence in Automation in the Fields of Gynaecology and Obstetrics - an Assessment of the State of Play

The long-awaited progress in digitalisation is generating huge amounts of medical data every day, and manual analysis and targeted, patient-oriented evaluation of this data is becoming increasingly difficult or even infeasible. This state of affairs and the associated, increasingly complex requirements for individualised precision medicine underline the need for modern software solutions and algorithms across the entire healthcare system. The utilisation of state-of-the-art equipment and techniques in almost all areas of medicine over the past few years has now indeed enabled automation processes to enter - at least in part - into routine clinical practice. Such systems utilise a wide variety of artificial intelligence (AI) techniques, the majority of which have been developed to optimise medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection and classification and, as an emerging field of research, radiogenomics. Tasks handled by AI are completed significantly faster and more precisely, clearly demonstrated by now in the annual findings of the ImageNet Large-Scale Visual Recognition Challenge (ILSVCR), first conducted in 2015, with error rates well below those of humans. This review article will discuss the potential capabilities and currently available applications of AI in gynaecological-obstetric diagnostics. The article will focus, in particular, on automated techniques in prenatal sonographic diagnostics.

Fetal heart rate short term variation during labor in relation to scalp blood lactate concentration

INTRODUCTION

Fetal heart rate short term variation (STV) decreases with severe chronic hypoxia in the antenatal period. However, only limited research has been done on STV during labor. We have tested a novel algorithm for a valid baseline estimation and calculated STV. To explore the value of STV during labor, we compared STV with fetal scalp blood (FBS) lactate concentration, an early marker in the hypoxic process.

MATERIAL AND METHODS

Software was developed which estimates baseline frequency using a novel algorithm and thereby calculates STV according to Dawes and Redman in up to four 30-minute blocks prior to each FBS. Cardiotocography traces from 1070 women in labor who had had FBS performed on 2134 occasions were analyzed.

RESULTS

In acidemic cases (lactate >4.8 mmol/L; Lactate Pro™), median STV 30 minutes prior to FBS was 7.10 milliseconds compared with 6.09 milliseconds in the preacidemic (4.2-4.8 mmol/L) and 5.23 milliseconds in the normal (<4.2 mmol/L) groups (P < .05). There was a positive correlation between lactate and STV (rho = 0.16-0.24; P < .05). Median lactate concentration in cases with STV <3.0 milliseconds (n = 160) was 2.3 mmol/L. When 2 FBS were performed within 60 minutes the change rate of lactate correlated to STV (rho = 0.33; P < .001). Cases with increasing lactate concentration had a median STV of 5.29 milliseconds vs 4.41 milliseconds in those with decreasing lactate (P < .001).

CONCLUSIONS

In the early stages of intrapartum hypoxia, STV increases, contrary to findings regarding chronic hypoxia in the antenatal period. The increase in the adrenergic surge is a likely explanation.