Explainable artificial intelligence for cough-related quality of life impairment prediction in asthmatic patients

Explainable Artificial Intelligence (XAI) is becoming a disruptive trend in healthcare, allowing for transparency and interpretability of autonomous decision-making. In this study, we present an innovative application of a rule-based classification model to identify the main causes of chronic cough-related quality of life (QoL) impairment in a cohort of asthmatic patients. The proposed approach first involves the design of a suitable symptoms questionnaire and the subsequent analyses via XAI. Specifically, feature ranking, derived from statistically validated decision rules, helped in automatically identifying the main factors influencing an impaired QoL: pharynx/larynx and upper airways when asthma is under control, and asthma itself and digestive trait when asthma is not controlled. Moreover, the obtained if-then rules identified specific thresholds on the symptoms associated to the impaired QoL. These results, by finding priorities among symptoms, may prove helpful in supporting physicians in the choice of the most adequate diagnostic/therapeutic plan.

Characterization of Synthetic Health Data Using Rule-Based Artificial Intelligence Models

The aim of this study is to apply and characterize eXplainable AI (XAI) to assess the quality of synthetic health data generated using a data augmentation algorithm. In this exploratory study, several synthetic datasets are generated using various configurations of a conditional Generative Adversarial Network (GAN) from a set of 156 observations related to adult hearing screening. A rule-based native XAI algorithm, the Logic Learning Machine, is used in combination with conventional utility metrics. The classification performance in different conditions is assessed: models trained and tested on synthetic data, models trained on synthetic data and tested on real data, and models trained on real data and tested on synthetic data. The rules extracted from real and synthetic data are then compared using a rule similarity metric. The results indicate that XAI may be used to assess the quality of synthetic data by (i) the analysis of classification performance and (ii) the analysis of the rules extracted on real and synthetic data (number, covering, structure, cut-off values, and similarity). These results suggest that XAI can be used in an original way to assess synthetic health data and extract knowledge about the mechanisms underlying the generated data.

Dual-View Single-Shot Multibox Detector at Urban Intersections: Settings and Performance Evaluation

The explosion of artificial intelligence methods has paved the way for more sophisticated smart mobility solutions. In this work, we present a multi-camera video content analysis (VCA) system that exploits a single-shot multibox detector (SSD) network to detect vehicles, riders, and pedestrians and triggers alerts to drivers of public transportation vehicles approaching the surveilled area. The evaluation of the VCA system will address both detection and alert generation performance by combining visual and quantitative approaches. Starting from a SSD model trained for a single camera, we added a second one, under a different field of view (FOV) to improve the accuracy and reliability of the system. Due to real-time constraints, the complexity of the VCA system must be limited, thus calling for a simple multi-view fusion method. According to the experimental test-bed, the use of two cameras achieves a better balance between precision (68%) and recall (84%) with respect to the use of a single camera (i.e., 62% precision and 86% recall). In addition, a system evaluation in temporal terms is provided, showing that missed alerts (false negatives) and wrong alerts (false positives) are typically transitory events. Therefore, adding spatial and temporal redundancy increases the overall reliability of the VCA system.

A novel method to derive personalized minimum viable recommendations for type 2 diabetes prevention based on counterfactual explanations

Despite the growing availability of artificial intelligence models for predicting type 2 diabetes, there is still a lack of personalized approaches to quantify minimum viable changes in biomarkers that may help reduce the individual risk of developing disease. The aim of this article is to develop a new method, based on counterfactual explanations, to generate personalized recommendations to reduce the one-year risk of type 2 diabetes. Ten routinely collected biomarkers extracted from Electronic Medical Records of 2791 patients at low risk and 2791 patients at high risk of type 2 diabetes were analyzed. Two regions characterizing the two classes of patients were estimated using a Support Vector Data Description classifier. Counterfactual explanations (i.e., minimal changes in input features able to change the risk class) were generated for patients at high risk and evaluated using performance metrics (availability, validity, actionability, similarity, and discriminative power) and a qualitative survey administered to seven expert clinicians. Results showed that, on average, the requested minimum viable changes implied a significant reduction of fasting blood sugar, systolic blood pressure, and triglycerides and a significant increase of high-density lipoprotein in patients at risk of diabetes. A significant reduction in body mass index was also recommended in most of the patients at risk, except in females without hypertension. In general, greater changes were recommended in hypertensive patients compared to non-hypertensive ones. The experts were overall satisfied with the proposed approach although in some cases the proposed recommendations were deemed insufficient to reduce the risk in a clinically meaningful way. Future research will focus on a larger set of biomarkers and different comorbidities, also incorporating clinical guidelines whenever possible. Development of additional mathematical and clinical validation approaches will also be of paramount importance.

Characterization of Type 2 Diabetes Using Counterfactuals and Explainable AI

Type 2 diabetes mellitus is a metabolic disorder of glucose management, whose prevalence is increasing inexorably worldwide. Adherence to therapies, along with a healthy lifestyle can help prevent the onset of disease. This preliminary study proposes the use of explainable artificial intelligence techniques with the aim of (i) characterizing diabetic patients through a set of easily interpretable rules and (ii) providing individualized recommendations for the prevention of the onset of the disease through the generation of counterfactual explanations, based on minimal variations of biomarkers routinely collected in primary care. The results of this preliminary study parallel findings from the literature as differences in biomarkers between patients with and without diabetes are observed for fasting blood sugar, body mass index, and high-density lipoprotein levels.

Evaluation of a Novel Speech-in-Noise Test for Hearing Screening: Classification Performance and Transducers Characteristics

One of the current gaps in teleaudiology is the lack of methods for adult hearing screening viable for use in individuals of unknown language and in varying environments. We have developed a novel automated speech-in-noise test that uses stimuli viable for use in non-native listeners. The test reliability has been demonstrated in laboratory settings and in uncontrolled environmental noise settings in previous studies. The aim of this study was: (i) to evaluate the ability of the test to identify hearing loss using multivariate logistic regression classifiers in a population of 148 unscreened adults and (ii) to evaluate the ear-level sound pressure levels generated by different earphones and headphones as a function of the test volume. The multivariate classifiers had sensitivity equal to 0.79 and specificity equal to 0.79 using both the full set of features extracted from the test as well as a subset of three features (speech recognition threshold, age, and number of correct responses). The analysis of the ear-level sound pressure levels showed substantial variability across transducer types and models, with earphones levels being up to 22 dB lower than those of headphones. Overall, these results suggest that the proposed approach might be viable for hearing screening in varying environments if an option to self-adjust the test volume is included and if headphones are used. Future research is needed to assess the viability of the test for screening at a distance, for example by addressing the influence of user interface, device, and settings, on a large sample of subjects with varying hearing loss.

A Generative Adversarial Network (GAN) Technique for Internet of Medical Things Data

The application of machine learning and artificial intelligence techniques in the medical world is growing, with a range of purposes: from the identification and prediction of possible diseases to patient monitoring and clinical decision support systems. Furthermore, the widespread use of remote monitoring medical devices, under the umbrella of the "Internet of Medical Things" (IoMT), has simplified the retrieval of patient information as they allow continuous monitoring and direct access to data by healthcare providers. However, due to possible issues in real-world settings, such as loss of connectivity, irregular use, misuse, or poor adherence to a monitoring program, the data collected might not be sufficient to implement accurate algorithms. For this reason, data augmentation techniques can be used to create synthetic datasets sufficiently large to train machine learning models. In this work, we apply the concept of generative adversarial networks (GANs) to perform a data augmentation from patient data obtained through IoMT sensors for Chronic Obstructive Pulmonary Disease (COPD) monitoring. We also apply an explainable AI algorithm to demonstrate the accuracy of the synthetic data by comparing it to the real data recorded by the sensors. The results obtained demonstrate how synthetic datasets created through a well-structured GAN are comparable with a real dataset, as validated by a novel approach based on machine learning.

MQTTset, a New Dataset for Machine Learning Techniques on MQTT

IoT networks are increasingly popular nowadays to monitor critical environments of different nature, significantly increasing the amount of data exchanged. Due to the huge number of connected IoT devices, security of such networks and devices is therefore a critical issue. Detection systems assume a crucial role in the cyber-security field: based on innovative algorithms such as machine learning, they are able to identify or predict cyber-attacks, hence to protect the underlying system. Nevertheless, specific datasets are required to train detection models. In this work we present MQTTset, a dataset focused on the MQTT protocol, widely adopted in IoT networks. We present the creation of the dataset, also validating it through the definition of a hypothetical detection system, by combining the legitimate dataset with cyber-attacks against the MQTT network. Obtained results demonstrate how MQTTset can be used to train machine learning models to implement detection systems able to protect IoT contexts.

Wireless Communication Technologies for Safe Cooperative Cyber Physical Systems

Cooperative Cyber-Physical Systems (Co-CPSs) can be enabled using wireless communication technologies, which in principle should address reliability and safety challenges. Safety for Co-CPS enabled by wireless communication technologies is a crucial aspect and requires new dedicated design approaches. In this paper, we provide an overview of five Co-CPS use cases, as introduced in our SafeCOP EU project, and analyze their safety design requirements. Next, we provide a comprehensive analysis of the main existing wireless communication technologies giving details about the protocols developed within particular standardization bodies. We also investigate to what extent they address the non-functional requirements in terms of safety, security and real time, in the different application domains of each use case. Finally, we discuss general recommendations about the use of different wireless communication technologies showing their potentials in the selected real-world use cases. The discussion is provided under consideration in the 5G standardization process within 3GPP, whose current efforts are inline to current gaps in wireless communications protocols for Co-CPSs including many future use cases.

Three-dimensional transvaginal sonographic assessment of uterine volume as preoperative predictor of need to morcellate in women undergoing laparoscopic hysterectomy

OBJECTIVE

In light of recent statements from the United States Food and Drug Administration warning against the use of power morcellation of uterine leiomyomas during laparoscopy, we sought to evaluate the use of preoperative two- (2D) and three- (3D) dimensional transvaginal ultrasound (US) assessment of uterine volume to predict the need for morcellation in women undergoing laparoscopic hysterectomy (LH).

METHODS

This was a prospective observational study performed between October 2008 and November 2011 in a tertiary referral laparoscopic unit. All women scheduled to undergo LH were included and underwent detailed preoperative transvaginal US. Uterine volumes were calculated using 2D-US measurements (ellipsoid formula), and using Virtual Organ Computer-aided AnaLysis (VOCAL™) having acquired 3D-US volumes of the uterus. Age, parity, need to morcellate and final uterine dry weight at histology were recorded. The estimated uterine volumes were then incorporated into a previously published logistic regression model to predict the need to morcellate for both nulliparous and parous women. The probability threshold cut-off of 0.14 (95% sensitivity) was evaluated in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratios (LRs). The performance of the models incorporating 2D- and 3D-US calculations were compared with 2D- and 3D-US-generated volumes alone, using receiver-operating characteristics (ROC) curves.

RESULTS

Of 76 women who underwent LH during the study period, 79% (n = 60) had complete background and 3D-US data. Their mean age was 43.7 years, 91.7% were parous and 35% underwent morcellation. The greatest uterine volume that did not require morcellation was 404 mL estimated using 3D-US, which corresponded to a uterine volume of 688.8 mL using 2D-US. The smallest uterine volume that required morcellation was 118.9 mL using 3D-US, which corresponded to a uterine volume of 123.4 mL using 2D-US. The 3D-US uterine volume for parous women with a sensitivity of 95% based on ROC-curve analysis was approximately 120 mL, which equated to a predicted probability of morcellation cut-off of 0.14. For this cut-off, specificity was 55.00%, PPV was 51.35%, NPV was 95.65%, LR+ was 2.11 and LR- was 0.09. Areas under the ROC curves for the morcellation logistic regression model were 0.769 (95% CI, 0.653-0.886) and 0.586 (95% CI, 0.419-0.753) using uterine volumes obtained by 3D-US and by 2D-US, respectively, and they were 0.938 (95% CI, 0.879-0.996) and 0.815 (95% CI, 0.681-0.948) using 3D-US and 2D-US volumes alone.

CONCLUSIONS

The need to morcellate can be predicted preoperatively using 3D-US uterine volumes obtained by transvaginal US with a fair degree of accuracy. Uteri with volumes smaller than 120 mL at 3D-US are very unlikely to require morcellation. The incorporation of 3D-US-estimated uterine volume into the previously published logistic regression model does not seem to confer any significant improvement when compared with 3D-US uterine volume alone to predict the need to morcellate in women undergoing total LH. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Accuracy of ultrasound dating formulae in the late second-trimester in pregnancies conceived with in-vitro fertilization

Purpose: To investigate the accuracy of ultrasound dating formulae in the late second trimester of pregnancy.

Material and Methods: A dataset of 136 singleton pregnancies conceived by artificial reproductive techniques was studied to assess the accuracy of ultrasound dating formulae in the late second trimester, and compared with early second trimester. A total of 21 published ultrasound-dating formulae were tested.

Results: For most formulae, late second trimester scans yielded results that were marginally less accurate than the early second trimester. The best performance was obtained with dating formulae based on femur length, either alone or combined with the biparietal diameter. These formulae had mean absolute errors of 3–3.5 days. Combining two or more parameters did not result in any substantial gain in accuracy.

Conclusions: Pregnancy dating by ultrasound in the late second trimester is sufficiently accurate for routine clinical use. Formulae based on femur length appear to be at least as accurate as head measurements.

Surveying multidisciplinary aspects in real-time distributed coding for Wireless Sensor Networks

Wireless Sensor Networks (WSNs), where a multiplicity of sensors observe a physical phenomenon and transmit their measurements to one or more sinks, pertain to the class of multi-terminal source and channel coding problems of Information Theory. In this category, “real-time” coding is often encountered for WSNs, referring to the problem of finding the minimum distortion (according to a given measure), under transmission power constraints, attainable by encoding and decoding functions, with stringent limits on delay and complexity. On the other hand, the Decision Theory approach seeks to determine the optimal coding/decoding strategies or some of their structural properties. Since encoder(s) and decoder(s) possess different information, though sharing a common goal, the setting here is that of Team Decision Theory. A more pragmatic vision rooted in Signal Processing consists of fixing the form of the coding strategies (e.g., to linear functions) and, consequently, finding the corresponding optimal decoding strategies and the achievable distortion, generally by applying parametric optimization techniques. All approaches have a long history of past investigations and recent results. The goal of the present paper is to provide the taxonomy of the various formulations, a survey of the vast related literature, examples from the authors' own research, and some highlights on the inter-play of the different theories.

The K-P algorithm: a new first-trimester growth model - comparison with other formulae and correlations with recorded embryonic lengths

OBJECTIVE

To generate and validate individually fitted first-trimester growth curves using a new growth model. Secondary aims were to compare this new model with actual recorded embryonic measurements and validate its predictive accuracy.

METHODS

A prospective study of women presenting to the Early Pregnancy Unit in the first trimester. Women with viable singleton pregnancies at the end of the first trimester who had had at least two crown-rump length (CRL) measurements were selected. An individual power function of CRL was derived from serial CRL measurements. Individual curves were fitted using computer software to estimate a dating adjustment factor ('k'), and a growth coefficient ('P') for each case. The predictive accuracy of the growth curves was then tested in a validation subset of the population that had a third CRL measurement. The population average curve from the developed model was also extrapolated to day 27 menstrual age (Carnegie stage 6), day 30 menstrual age (Carnegie stage 7) and day 84 menstrual age, and values were compared to previously reported measurements.

RESULTS

326 viable pregnancies were selected for CRL growth curve development. The mean time interval between CRL measurements was 20.5 days (range 2-44). The mean value for 'P' was 2.058, and for 'k' 24.6. Testing the model on a subset of 81 cases showed that the average error in predicting a third CRL measurement was 1% (SD 9.1%).

CONCLUSIONS

These new, individually fitted growth curves for the first trimester correlate more closely with the recorded embryonic lengths than other standards.

Does chewing coca leaves influence physiology at high altitude?

Andean Indians have used coca leaves (Erythroxylon coca and related species) for centuries to enhance physical performance. The benefits and disadvantages of using coca leaf have been a subject of many political debates. The aim of this study was to investigate the effects of chewing coca leaves on biochemical and physiological parameters. Cutaneous microdialysis catheters were used to estimate systemic biochemical changes. We subjected 10 healthy adult males (local residents) in Cajamarca (Peru, altitude 2700 m) to a standardised exercise routine on a stationary cycle ergometer. The blood pressure, oxygen saturation (digital), pulse, VO2 max and ECG (Holter monitor) were recorded before the exercise. Cutaneous microdialysis catheters were introduced in the forearm. The subjects were given to chew 8 g of coca leaves with a small amount of lime. They were then placed on the cycle ergometer for 20 min. Blood pressure, oxygen saturation, pulse, ECG and VO2 max were recorded. Pyruvate, glucose, lactate, glycerol and glutamate levels were estimated. Oxygen saturation, blood pressure, and pulse rate did not show any significant changes between the two groups. Glucose levels showed hyperglycaemic response. Glycerol, Lactate and Pyruvate increased. Glutamate remained unchanged. Similar changes were not seen in the controls. These results suggest that coca leaves have blocked the glycolytic pathway of glucose oxidation resulting in accumulation of glucose and pyruvate. The energy requirement for exercise is being met with beta-oxidation of fatty acids. The glycerol released was also getting accumulated since its pathway for oxidation was blocked. These experimental findings suggest that chewing coca leaves is beneficial during exercise and that the effects are felt over a prolonged period of sustained physical activity.

Z-velocity in screening for intrauterine growth restriction

OBJECTIVES

Ultrasound scans provide the basis for detection of intrauterine growth restriction (IUGR) but often fail to distinguish IUGR from small-for-gestational age (SGA) fetuses. This study introduces the concept of Z-velocity, calculated as changes in Z-scores over time, as an additional criterion in the diagnosis of IUGR.

METHODS

A computer program simulated 50 000 fetal abdominal circumference (FAC) scans based on published growth formulae. False-positive rates were calculated to determine optimal scan time and scan intervals. Using an independent simulation of 32 500 FAC scans, the two methods were compared using receiver-operating characteristics (ROC) curve analysis.

RESULTS

ROC showed areas under the curve of > 0.74 over the complete range of scan intervals. The positive predictive value of growth arrest as the only diagnostic criterion was, however, too low to recommend it as an exclusive or the first diagnostic criterion.

CONCLUSIONS

Z-velocity can be used to decide whether further investigations for growth abnormality are required in fetuses that fall below the 10(th) percentile. The gain of combined diagnostic approaches should be calculated from large databases that include the neonatal ponderal index as the gold standard.

Neural approximation of open-loop feedback rate control in satellite networks

A resource allocation problem for a satellite network is considered, where variations of fading conditions are added to those of traffic load. Since the capacity of the system is finite and divided in finite discrete portions, the resource allocation problem reveals to be a discrete stochastic programming one, which is typically NP-Hard. In practice, a good approximation of the optimal solution could be obtained through the adoption of a closed-form expression of the performance measure in steady-state conditions. Once we have summarized the drawbacks of such optimization strategy, we address two novel optimization approaches. The first one derives from Gokbayrak and Cassandras and is based on the minimization over the discrete constraint set using an estimate of the gradient, obtained through a "relaxed continuous extension" of the performance measure. The computation of the gradient estimation is based on infinitesimal perturbation analysis (IPA). Neither closed forms of the performance measures, nor additional feedbacks concerning the state of the system and very mild assumptions about the stochastic environment are requested. The second one is the main contribution of the present work, and is based on an open-loop feedback control (OLFC) strategy, aimed at providing optimal reallocation strategies as functions of the state of the network. The optimization approach leads us to a functional optimization problem, and we investigate the adoption of a neural network-based technique, in order to approximate its solution. As is shown in the simulation results, we obtain near-optimal reallocation strategies with a small real time computational effort and avoid the suboptimal transient periods introduced by the IPA gradient descent algorithm.

Ultrasound diagnosis of fetal macrosomia: a comparison of weight prediction models using computer simulation

OBJECTIVE

To assess the frequency of the diagnosis of macrosomia in relation to differing weight estimation formulae in unselected pregnancies.

METHODS

Computer modeling techniques were employed. Computer modeling software generated correlated fetal biometry measurements according to published British standards, from 37 to 41 weeks' gestation. For each set of measurements, estimated fetal weights were obtained by a panel of 18 ultrasound weight formulae. The diagnosis of macrosomia was made if the fetal weight estimate was greater than 4500 g. Cohorts of 5000 pregnancies for each week of gestation were studied.

RESULTS

The frequency of diagnosis of macrosomia increased progressively with advancing gestational age, with large increases between 40 and 41 weeks. The type of weight estimation formula had a profound influence on the frequency of diagnosis of macrosomia. Five of the formulae tested almost never returned a weight estimate greater than 4500 g. Three formulae yielded false positive rates in excess of 15%. The Hadlock group of formulae yielded frequencies of 0.3% to 14.6%.

CONCLUSIONS

Most formulae tend to over-diagnose macrosomia at term. Intervention rates for suspected fetal macrosomia may be influenced by gestational age at the time of scan and the type of fetal weight estimation formula in use.

Third-trimester ultrasound dating algorithms derived from pregnancies conceived with artificial reproductive techniques

OBJECTIVE

To derive an accurate formula for ultrasound estimation of gestational age in late pregnancy.

METHODS

A database of 123 singleton pregnancies conceived by artificial reproductive techniques with third-trimester ultrasound measurements was studied. Biometry variables included the fetal head circumference (HC) and femur length (FL). The dataset was equally divided into a derivation sample and a target sample. To derive the equations of best fit, regression analysis was used, with true menstrual age as the dependent variable and fetal biometry measurements as independent variables. The formulae were tested on the target set and the menstrual age estimates were compared with the actual menstrual age. Clinical performance was estimated in terms of systematic and random errors, absolute errors and their 95% CI values.

RESULTS

The menstrual ages at time of scanning ranged from 26 to 41 weeks, with a mean of 33 weeks. The best performing derived formula was a combination of HC and FL. This had a random error of 7.5 days and prediction errors within a 95% confidence limit of -13 to +17 days. With FL only, the random error was 8.2 days, whereas using the HC only yielded a random error of 9.4 days.

CONCLUSIONS

Ultrasound estimation of gestational age in late pregnancy is better than indicated by older publications. It is more accurately estimated by combining the HC with the FL than by using the FL measurements alone.

Accuracy of ultrasound dating formulae in the late second-trimester in pregnancies conceived with in-vitro fertilization

PURPOSE

To investigate the accuracy of ultrasound dating formulae in the late second trimester of pregnancy.

MATERIAL AND METHODS

A dataset of 136 singleton pregnancies conceived by artificial reproductive techniques was studied to assess the accuracy of ultrasound dating formulae in the late second trimester, and compared with early second trimester. A total of 21 published ultrasound-dating formulae were tested.

RESULTS

For most formulae, late second trimester scans yielded results that were marginally less accurate than the early second trimester. The best performance was obtained with dating formulae based on femur length, either alone or combined with the biparietal diameter. These formulae had mean absolute errors of 3-3.5 days. Combining two or more parameters did not result in any substantial gain in accuracy.

CONCLUSIONS

Pregnancy dating by ultrasound in the late second trimester is sufficiently accurate for routine clinical use. Formulae based on femur length appear to be at least as accurate as head measurements.

Factors influencing fetal growth

Birth weight remains one of the most important measures we have of the health status of a population, and is a strong predictor of both neonatal mortality and morbidity. It reflects maternal nutritional status and fetal growth rates in the antenatal period. Although a large number of physiological and pathological variables can affect the growth rate, for most cases of growth-restricted fetuses a cause cannot be determined. Awareness of the physiological aspects of fetal growth is important in assessing whether the fetus is growing normally. This article is a review of recent findings and key historical material on factors influencing birth weight.

Attitudes of Singapore women toward cesarean and vaginal deliveries

OBJECTIVES

To assess the attitude, knowledge, and expectations of Asian pregnant women toward cesarean and vaginal deliveries.

METHODS

Written questionnaires were given to pregnant women attending the National University Hospital antenatal clinics, and 160 responses were tabulated and analyzed using SPSS software.

RESULTS

The participation rate was 65% and 50% of the respondents were Chinese, 20% Indian, 21% Malay, 2% White, and 9.2% Other. The median age was 31 years, and approximately 43% were primiparas. Only 3.7% of them would prefer an elective cesarean delivery, and although 50% had friends or relatives who requested one, only 3% felt that this influenced their preference. The most common reasons for choosing a cesarean delivery were avoiding labor pains and lowering the risk of fetal distress. When asked which modality they would recommend to their friends, only 2% expressed that they would recommend cesarean delivery; however, 71% stated that women should have the right to request a cesarean delivery. Awareness of complications arising from vaginal and cesarean deliveries was generally low and related to the frequency of these complications. There was no significant correlation between demographic characteristics and maternal preference for mode of delivery.

CONCLUSIONS

Asian women largely prefer a vaginal delivery, and their attitude toward cesarean delivery on demand is comparable with that of Western women. Cultural or ethnic differences are unlikely to affect maternal preference for cesarean delivery in Singapore women.

Induction policy and missed post-term pregnancies: a mathematical model

The aim of this study was to compare the clinical performance of ultrasound dates and ultrasound dates combined with menstrual dates for the detection of post-maturity. A computer model was designed which uses the statistical distributions of the duration of normal pregnancy, day of ovulation in relation to the menstrual cycle and ultrasound error for estimating gestational age. The clinical performance of the different dating methods was then analysed from these variables, on simulations of 30,000 cases. The efficacy of different dating methods for detecting post-maturity was determined by generating receiver-operator characteristics (ROC) curves. The proportion of post-term pregnancies (294 days and over) predicted by the model (3.5%) agrees with published values. There is a steep rise in missed cases if induction is delayed beyond 10 days from the expected date of delivery, reaching 20% on day 294. Elective delivery on day 290 will detect 98.9% of cases destined to deliver post-term, with an induction rate of 10%; the respective figures for induction on day 294 are 79% and 3.8%. The ROC curves for the detection of post-maturity suggest that use of the mid-trimester biparietal diameter (BPD) is better than a 7-day or 10-day rule. Timing of elective delivery is the most important variable affecting the detection rate for post-maturity There is no advantage in using menstrual dates when ultrasound biometry is available.

A fetal growth standard derived from multiple modalities

OBJECTIVE

To create a fetal weight growth standard from published data on ultrasound-dated pregnancies.

METHODS

A fetal growth standard was calculated from published birth weight data, sonographic weight standards, and one MRI study. The birth weights from the East Midlands Obstetric Database were modified by an incremental function to compensate for the lower weights of preterm infants. Published sonographic and MRI standards, and the modified birth weight curve were transformed to yield fractional growth curves that express fetal size as a proportion of the expected term weight. These three curves were then averaged, and the standard deviation was taken as 12% of the median. Curves for Europeans and Chinese were then generated.

RESULTS

The average curve for fetal weight is a virtually linear function of gestational age, with a weekly weight gain of 5.2% of the expected term birth weight.

CONCLUSIONS

Ultrasound-derived fetal growth curves are subject to bias inherent in weight estimation formulae. A standard derived from multiple modalities is likely to be a more robust estimate of normal fetal growth.

Sociobiological variables and pregnancy outcome

Recent advances are beginning to shed light on the mechanisms whereby adverse psychosocial factors can influence pregnancy outcome. High levels of maternal stress have been linked to endocrine disturbances, which in turn increases the risk of preterm labor considerably. These observations have been supported by experimental animal models. Birth weight is subject to considerable ethnic variation, and on its own is a nonspecific indicator of pregnancy outcome. The benefits of social and psychological intervention have been best documented in the intrapartum situation, whereas antenatal intervention is most likely to be of benefit when focused on improving socioeconomic conditions and access to healthcare providers.

Fetal growth

Recent epidemiological and experimental studies show that abnormal fetal growth can lead to serious complications, including stillbirth, perinatal morbidity and disorders extending well beyond the neonatal period. It is now clear that the intrauterine milieu is as important as genetic endowment in shaping the future health of the conceptus. Maternal characteristics such as weight, height, parity and ethnic group need to be adjusted for, and pathological factors such as smoking excluded, to establish appropriate standards and improve the distinction between what is normal and abnormal. Currently, the aetiology of growth restriction is not well understood and preventative measures are ineffective. Elective delivery remains the principal management option, which emphasizes the need for better screening techniques for the timely detection of intrauterine growth failure.

Effect of labour and delivery on plasma hepatic enzymes in the newborn

OBJECTIVE

To study the relationship between cord blood hepatic enzymes and obstetric and neonatal outcome in a Chinese population.

MATERIALS AND METHODS

The study group consisted of 288 low-risk Chinese women with singleton term pregnancies. The following enzymes were assayed in cord blood: lactate dehydrogenase (LDH), glutamyl transferase (GGT), aspartate aminotransferase (AST) and alanine transferase (ALT). These were correlated to maternal and neonatal characteristics.

RESULTS

A strong correlation was noted between cord blood AST and LDH (R = 0.582, p < 0.01), which was absent amongst those infants delivered by elective cesarean section. LDH, AST and ALT were negatively correlated with cord arterial pH and base excess (BE). GGT was inversely related only to gestational age (R = -0.18, p < 0.01). Both LDH and AST were weakly correlated with the duration of the first and second stages of labour. LDH was most closely linked to arterial pH, whereas AST was related to both arterial BE and duration of the second stage.

CONCLUSIONS

The reference values are comparable to those published for Caucasian populations. There are moderate elevations in LDH and AST associated with the onset of labour and changes in acid-base status.

Symphysis-fundus height and pregnancy characteristics in ultrasound-dated pregnancies

OBJECTIVE

To investigate the relation between maternal and pregnancy characteristics and symphysis-fundus height values at term in an obstetric population dated by sonography.

METHODS

Three hundred twenty-five women were recruited from the antenatal clinics of the Queen's Medical Centre, Nottingham, United Kingdom for measurement of fundal height and for ultrasound scans. Symphysis-fundus height measurements were analyzed by multivariate regression analysis in relation to gestational age, maternal height and weight, ethnic group, and smoking.

RESULTS

Gestational age and maternal characteristics explained nearly half of the variability in symphysis-fundus height. Gestational age was the most important determinant, followed by maternal weight, parity, and sex of the infant. The other variables were not significantly correlated.

CONCLUSION

Maternal characteristics had statistically significant effects on the expected symphysis-fundus height, which suggests that individually adjusted fundal height charts may improve the precision of clinical screening for fetal growth restriction.

[Pericardial involvement as initial manifestation of multiple myeloma]

Pericardial involvement is a rare complication of multiple myeloma, caused by amyloidosis, infections, or plasmacell infiltration, usually at late or terminal disease stage. We report a patient with pericarditis coming from a department of Cardiology where a preceding (15 years before) diagnosis of breast cancer and present bloody pericardial effusion with probably malignant cells permitted at first to orientate towards metastatic pericardial involvement in breast cancer. Laboratory findings (pancytopenia, hypogammaglobulinemia, proteinuria) suggested to perform bone marrow aspirate, serum and concentrated urine immunoelectrophoresis, measurement of 24-h urine protein excretion, and further cytologic and immunocytochemical assay of pericardial fluid. Acquired data allowed to diagnose light chain multiple myeloma with pericardial involvement caused by plasmacell infiltration. We diagnosed this complication, representing first and main clinical feature of multiple myeloma, owing to a complete clinical and laboratory evaluation and repetition of cytologic and immunocytochemical assay of pericardial fluid.