Sensory event-related potential morphology predicts age in premature infants

OBJECTIVE

We investigated whether sensory-evoked cortical potentials could be used to estimate the age of an infant. Such a model could be used to identify infants who deviate from normal neurodevelopment.

METHODS

Infants aged between 28- and 40-weeks post-menstrual age (PMA) (166 recording sessions in 96 infants) received trains of visual and tactile stimuli. Neurodynamic response functions for each stimulus were derived using principal component analysis and a machine learning model trained and validated to predict infant age.

RESULTS

PMA could be predicted accurately from the magnitude of the evoked responses (training set mean absolute error and 95% confidence intervals: 1.41 [1.14; 1.74] weeks,p = 0.0001; test set mean absolute error: 1.55 [1.21; 1.95] weeks,p = 0.0002). Moreover, we show that their predicted age (their brain age) is correlated with a measure known to relate to maturity of the nervous system and is linked to long-term neurodevelopment.

CONCLUSIONS

Sensory-evoked potentials are predictive of age in premature infants and brain age deviations are related to biologically and clinically meaningful individual differences in nervous system maturation.

SIGNIFICANCE

This model could be used to detect abnormal development of infants' response to sensory stimuli in their environment and may be predictive of neurodevelopmental outcome.

The roles, challenges, and merits of the p value

Since the 18th century, the p value has been an important part of hypothesis-based scientific investigation. As statistical and data science engines accelerate, questions emerge: to what extent are scientific discoveries based on p values reliable and reproducible? Should one adjust the significance level or find alternatives for the p value? Inspired by these questions and everlasting attempts to address them, here, we provide a systematic examination of the p value from its roles and merits to its misuses and misinterpretations. For the latter, we summarize modest recommendations to handle them. In parallel, we present the Bayesian alternatives for seeking evidence and discuss the pooling of p values from multiple studies and datasets. Overall, we argue that the p value and hypothesis testing form a useful probabilistic decision-making mechanism, facilitating causal inference, feature selection, and predictive modeling, but that the interpretation of the p value must be contextual, considering the scientific question, experimental design, and statistical principles.

Personalized Longitudinal Assessment of Multiple Sclerosis Using Smartphones

Personalized longitudinal disease assessment is central to quickly diagnosing, appropriately managing, and optimally adapting the therapeutic strategy of multiple sclerosis (MS). It is also important for identifying idiosyncratic subject-specific disease profiles. Here, we design a novel longitudinal model to map individual disease trajectories in an automated way using smartphone sensor data that may contain missing values. First, we collect digital measurements related to gait and balance, and upper extremity functions using sensor-based assessments administered on a smartphone. Next, we treat missing data via imputation. We then discover potential markers of MS by employing a generalized estimation equation. Subsequently, parameters learned from multiple training datasets are ensembled to form a simple, unified longitudinal predictive model to forecast MS over time in previously unseen people with MS. To mitigate potential underestimation for individuals with severe disease scores, the final model incorporates additional subject-specific fine-tuning using data from the first day. The results show that the proposed model is promising to achieve personalized longitudinal MS assessment; they also suggest that features related to gait and balance as well as upper extremity function, remotely collected from sensor-based assessments, may be useful digital markers for predicting MS over time.

Probing potential priming: Defining, quantifying, and testing the causal priming effect using the potential outcomes framework

Having previously seen an item helps uncover the item another time, given a perceptual or cognitive cue. Oftentimes, however, it may be difficult to quantify or test the existence and size of a perceptual or cognitive effect, in general, and a priming effect, in particular. This is because to examine the existence of and quantify the effect, one needs to compare two outcomes: the outcome had one previously seen the item vs. the outcome had one not seen the item. But only one of the two outcomes is observable. Here, we argue that the potential outcomes framework is useful to define, quantify, and test the causal priming effect. To demonstrate its efficacy, we apply the framework to study the priming effect using data from a between-subjects study involving English word identification. In addition, we show that what has been used intuitively by experimentalists to assess the priming effect in the past has a sound mathematical foundation. Finally, we examine the links between the proposed method in studying priming and the multinomial processing tree (MPT) model, and how to extend the method to study experimental paradigms involving exclusion and inclusion instructional conditions.

The temporal dynamics of sleep disturbance and psychopathology in psychosis: a digital sampling study

BACKGROUND

Sleep disruption is a common precursor to deterioration and relapse in people living with psychotic disorders. Understanding the temporal relationship between sleep and psychopathology is important for identifying and developing interventions which target key variables that contribute to relapse.

METHODS

We used a purpose-built digital platform to sample self-reported sleep and psychopathology variables over 1 year, in 36 individuals with schizophrenia. Once-daily measures of sleep duration and sleep quality, and fluctuations in psychopathology (positive and negative affect, cognition and psychotic symptoms) were captured. We examined the temporal relationship between these variables using the Differential Time-Varying Effect (DTVEM) hybrid exploratory-confirmatory model.

RESULTS

Poorer sleep quality and shorter sleep duration maximally predicted deterioration in psychosis symptoms over the subsequent 1-8 and 1-12 days, respectively. These relationships were also mediated by negative affect and cognitive symptoms. Psychopathology variables also predicted sleep quality, but not sleep duration, and the effect sizes were smaller and of shorter lag duration.

CONCLUSIONS

Reduced sleep duration and poorer sleep quality anticipate the exacerbation of psychotic symptoms by approximately 1-2 weeks, and negative affect and cognitive symptoms mediate this relationship. We also observed a reciprocal relationship that was of shorter duration and smaller magnitude. Sleep disturbance may play a causal role in symptom exacerbation and relapse, and represents an important and tractable target for intervention. It warrants greater attention as an early warning sign of deterioration, and low-burden, user-friendly digital tools may play a role in its early detection.

Sleep monitoring using ear-centered setups: Investigating the influence from electrode configurations

Modern sleep monitoring development is shifting towards the use of unobtrusive sensors combined with algorithms for automatic sleep scoring. Many different combinations of wet and dry electrodes, ear-centered, forehead-mounted or headband-inspired designs have been proposed, alongside an ever growing variety of machine learning algorithms for automatic sleep scoring.

OBJECTIVE

Among candidate positions, those in the facial area and around the ears have the benefit of being relatively hairless, and in our view deserve extra attention. In this paper, we seek to determine the limits to sleep monitoring quality within this spatial constraint.

METHODS

We compare 13 different, realistic sensor setups derived from the same data set and analysed with the same pipeline.

RESULTS

All setups which include both a lateral and an EOG derivation show similar, state-of-the-art performance, with average Cohen's kappa values of at least 0.80.

CONCLUSION

If large electrode distances are used, positioning is not critical for achieving large sleep-related signal-to-noise-ratio, and hence accurate sleep scoring.

SIGNIFICANCE

We argue that with the current competitive performance of automated staging approaches, there is a need for establishing an improved benchmark beyond current single human rater scoring.

P–684 Impact of GnRH antagonist pretreatment on oocyte yield after ovarian stimulation: a retrospective analysis

Study question

Does a 3-day pretreatment course with a GnRH antagonist in the early follicular phase increase the number of oocytes in a GnRH antagonist stimulation protocol?

Summary answer

The administration of 3 days of GnRH antagonist before starting ovarian stimulation in a GnRH antagonist protocol increases the number of COCs (Cumulus-Oocyte-Complexes).

What is known already

The GnRH antagonist protocol is characterized by higher gonadotropin and E2 serum levels at the start of ovarian stimulation (OS), compared with a long pituitary down regulation protocol. The unsuppressed FSH level at the start of a GnRH antagonist cycle allows the initial growth of follicles before addition of exogenous FSH, which may result in asynchrony of the follicular cohort. Menstrual administration of a GnRH antagonist can inhibit follicle growth and improve homogeneity of recruitable follicles. Previous studies showed a trend toward higher numbers of COCs and improved maturation and fertilization rates of retrieved oocytes.

Study design, size, duration

Retrospective single center crossover study, including consecutive women enrolled in an IVF program in a university hospital from January 2011 to December 2020. All women underwent one standard GnRH antagonist stimulation cycle (“standard cycle”) and one GnRH antagonist stimulation cycle preceded by early administration of GnRH antagonist for 3 days (“pretreatment cycle”). Women with basal progesterone levels >1.5ng/ml, and women undergoing oocyte freezing, oocyte donation or PGT were excluded. In total, 427 patients were included.

Participants/materials, setting, methods

Women were included when the pretreatment cycle occurred within a time interval of < 12 months following the start of stimulation in the standard cycle. The primary outcome was the total number of COCs.

Main results and the role of chance

The average female age was 35.1 ± 4.7 years. Indications for fertility treatment included unexplained infertility (34.3%), male-factor infertility (33.3%), age (16.9%), PCOS (8.2%) and endometriosis (2.6%). All cycles were divided into two groups: group 1 (standard, 427 cycles) and group 2 (pretreatment, 427 cycles). The mean duration of stimulation was similar in both groups (10.3 vs 10.3 days, p = 0.2). The starting dose of gonadotropin (196.8 vs 234.9IU, p < 0.001) and total amount of gonadotropin used (2000.7 vs 2415.2IU, p < 0.001) were higher in group 2. The total number of obtained COCs (6.2 vs 8.8 p < 0.001) and the number of mature oocytes (4.2 vs 6.4 p < 0.001) were significantly higher in group 2. The Generalized estimating equation (GEE) multivariate regression analysis showed that the pretreatment strategy had a significant positive effect on the number of COCs (coefficient 2.8, p value <0.001 after adjusting for the confounders age, indication of infertility, stimulation dose, type and total amount of gonatropins used).

Limitations, reasons for caution

Despite the large dataset, the presence of biases related to the retrospective study design cannot be excluded. Besides, the impact of GnRH pretreatment on pregnancy rate cannot be assessed because of the crossover design.

Wider implications of the findings: A 3-day course of GnRH antagonist pretreatment increases the number of COCs obtained after OS. Furthermore, since the initiation of OS in a GnRH antagonist protocol relies on the occurrence of spontaneous menses, addition of three days of GnRH antagonist pretreatment may enhance scheduling flexibility without reducing efficacy.

Trial registration number

Not applicable

P–318 Short (seven days) versus conventional (fourteen days) estrogen priming in an artificial frozen embryo transfer cycle: a randomised controlled trial

Study question

What is the impact of seven days versus fourteen days’ estrogen (E2) priming on the clinical outcome of frozen-embryo-transfer in artificially prepared endometrium (FET-HRT) cycles?

Summary answer

No significant difference in clinical/ongoing pregnancy rate was observed when comparing 7 versus 14 days of estrogen priming before starting progesterone (P) supplementation.

What is known already

One (effective) method for endometrial preparation prior to frozen embryo transfer is hormone replacement therapy (HRT), a sequential regimen with E2 and P, which aims to mimic the endocrine exposure of the endometrium in a physiological cycle. The average duration of E2 supplementation is generally 12–14 days, however, this protocol has been arbitrarily chosen whereas, the optimal duration of E2 implementation remains unknown.

Study design, size, duration

This is a single-center, randomized, controlled, open-label pilot study. All FET-HRT cycles were performed in a tertiary centre between October 2018 and December 2020. Overall, 150 patients were randomized of whom 132 were included in the analysis after screening failure and drop-out.

Participants/materials, setting, methods

The included patients were randomized into one of 2 groups; group A (7 days of E2 prior to P supplementation) and group B (14 days of E2 prior to P supplementation). Both groups received blastocyst stage embryos for transfer on the 6th day of vaginal P administration. Pregnancy was assessed by an hCG blood test 12 days after FET and clinical pregnancy was confirmed by transvaginal ultrasound at 7 weeks of gestation.

Main results and the role of chance

Following the exclusion of drop-outs and screening failures, 132 patients were finally included both in group A (69 patients) or group B (63 patients). Demographic characteristics for both groups were comparable. The positive pregnancy rate was 46.4% and 53.9%, (p 0.462) for group A and group B, respectively. With regard to the clinical pregnancy rate at 7 weeks, no statistically significant difference was observed (36.2% vs 36.5% for group A and group B, respectively, p = 0.499). The secondary outcomes of the study (biochemical pregnancy, miscarriage and live birth rate) were also comparable between the two arms for both PP and ITT analysis. Multivariable logistic regression showed that the HRT scheme is not associated with pregnancy rate, however, the P value on the day of ET is significantly associated with the pregnancy outcome.

Limitations, reasons for caution

This study was designed as a proof of principle trial with a limited study population and therefore underpowered to determine the superiority of one intervention over another. Instead, the purpose of the present study was to explore trends in outcome differences and to allow us to safely design larger RCTs.

Wider implications of the findings: The results of this study give the confidence to perform larger-scale RCTs to confirm whether a FET-HRT can be performed safely in a shorter time frame, thus, reducing the TTP, while maintaining comparable pregnancy and live birth rates.

Trial registration number

NCT03930706

P–462 First live birth after fertility preservation using vitrified oocytes in a woman with mosaic Turner syndrome

Study question

Is oocyte vitrification an option for preserving the fertility of women diagnosed with Turner syndrome (TS)?

Summary answer

We report the first live birth achieved using cryopreserved oocytes in a woman diagnosed with mosaic Turner syndrome.

What is known already

Women with TS are at extremely high risk for premature ovarian insufficiency (POI) and infertility. Although the desire of becoming parents may be fulfilled through egg donation or adoption, fertility preservation using ovarian tissue cryopreservation or oocyte vitrification has been offered to adolescents with TS before complete exhaustion of their follicular stockpile. However, women with TS exhibit higher rates of pregnancy loss and obstetric complications, and the feasibility of fertility preservation in TS is hampered by the reduced follicular pool and by concerns about the X chromosomal content of oocytes and follicular cells.

Study design, size, duration

Case report in a university hospital.

Participants/materials, setting, methods

A 25-year-old woman with Turner syndrome mosaicism (45,X0[14]/46,XX[86]) was referred for fertility preservation (FP) counseling. Serum antimüllerian hormone (AMH) level was normal (6.4 µg/L). In view of parenthood postponement and because of the unpredictable rate of follicle loss, the woman underwent two cycles of ovarian stimulation using recombinant follicle stimulating hormone (rFSH), 200–250 IU/day for 8 resp. 12 days, in a GnRH antagonist protocol.

Main results and the role of chance

In total, 29 metaphase II oocytes (MII) were vitrified. Five years later, the patient returned to the clinic with a desire for pregnancy. Because of evidence of considerable AMH decline (–56% in an interval of four years), the patient was advised to utilize her cryopreserved oocytes for in-vitro fertilization with preimplantation genetic testing for aneuploidy screening (PGT-A). All 29 MII oocytes were thawed; 26 oocytes survived (89.7%) and were inseminated using intracytoplasmic sperm injection (ICSI). Thirteen oocytes were fertilized normally. Three good quality blastocysts ensued and were vitrified after trophectoderm biopsy for PGT-A using array-CGH. Two blastocysts were found euploid. One was thawed and transferred into the uterus using a HRT priming protocol. An uneventful pregnancy occurred. The patient delivered a healthy baby girl weighing 3490 g at 40 weeks of gestation.

Limitations, reasons for caution

Cryopreservation of oocytes and/or ovarian tissue in selected postmenarchal girls or young women with Turner syndrome is an investigational FP approach that may result in genetic parenthood. The feasibility of FP in TS individuals is limited to those with evidence of ovarian function, before POI occurs.

Wider implications of the findings: Cryopreservation of mature oocytes after ovarian stimulation is a realistic option for FP in selected postmenarchal individuals with mosaic TS. Whether PGT-A may reduce the risk of pregnancy loss in TS has to be confirmed by further studies.

Trial registration number

Not applicable

P–464 What is the optimal ovarian stimulation (OS) protocol for women who undergo planned oocyte cryopreservation (POC)?

Study question

When repeated cycles of OS for planned oocyte cryopreservation using a standard GnRH antagonist protocol are required, can OS protocol modifications improve oocyte yield?

Summary answer

Compared to repeating a standard GnRH antagonist protocol, switching to a long GnRH agonist protocol for POC results in a higher number of cryopreserved oocytes.

What is known already

The total number of cryopreserved oocytes is a key parameter of POC programs because of its association with livebirth. A substantial proportion of women embarking on POC will undergo repeated cycles of OS to reach their desired target number of vitrified oocytes. According to recent guidelines, the GnRH antagonist protocol with GnRH agonist triggering is considered the first choice protocol for POC, because of its safety profile and convenience. However, in women with normal ovarian reserve, the long GnRH agonist protocol results in a higher number of oocytes retrieved. Evidence regarding the optimal protocol for POC is limited.

Study design, size, duration

This is a single-centre, retrospective cohort study including 283 women who had a first cycle for POC using a standard GnRH antagonist protocol and who requested a second OS cycle to increase their total number of vitrified oocytes for later use. The choice of protocol for the second cycle was left at the discretion of the reproductive medicine specialist. All OS cycles took place between January 2009 and December 2019 in a tertiary referral hospital.

Participants/materials, setting, methods

After ovarian reserve testing, the first cycle OS was performed using rFSH or HPhMG in a GnRH antagonist protocol. For the second cycle, a GnRH antagonist protocol with or without antagonist pretreatment, or a long GnRH agonist protocol was prescribed. The primary outcome was the number of mature oocytes (MII) vitrified per cycle. Cycle characteristics were compared. Data were assessed by generalized estimating equation (GEE) regression analysis adjusting for covariates.

Main results and the role of chance

In total, 226 (79.9%) women had a GnRH antagonist protocol and 57 (20.1%) had a long GnRH agonist protocol in their second OS cycle for POC. Overall, mean age was 36.6±2.4 years. The median (CI) number of mature oocytes vitrified after the second OS cycle was significantly higher than that after the first cycle [8 (5–11) vs. 7 (4–10), p < 0.001]. According to GEE multivariate regression, adjusting for relevant confounders, switching from a GnRH antagonist protocol in the first cycle to a long GnRH agonist protocol in the second cycle was the only significant predictor of the number of vitrified oocytes after the subsequent cycle (coefficient 1.59, CI 0.29–2.89, p-value = 0.017). Age, AFC, initial dose and type of gonadotropins did not predict the number of vitrified oocytes. None of the women developed moderate or severe OHSS.

Similarly, of 174 women who underwent their first OS cycle with a standard GnRH antagonist protocol, 133 women (76.4%) had the same protocol for their second cycle and 41 women (23.6%) an additional three-day course of GnRH antagonist pretreatment. According to GEE multivariate regression, this protocol modification did not result in more mature oocytes available for vitrification (coefficient –0.25, CI –1.86–1.36, p-value = 0.76).

Limitations, reasons for caution

These data should be interpreted with caution because of the retrospective design and limited sample. Although more oocytes were obtained with a long GnRH agonist protocol we have no data on livebirth in women returning to use their oocytes to support the choice for a specific OS protocol for POC.

Wider implications of the findings: Although oocyte yield in the context of POC is an important parameter that may be modulated by the choice of OS protocol, the ultimate outcome measure of a successful POC program is livebirth after oocyte vitrification. Future research of oocyte parameters reflecting oocyte quality is paramount.

Trial registration number

Not applicable

Building a Machine-Learning Framework to Remotely Assess Parkinson's Disease Using Smartphones

OBJECTIVE

Parkinson's disease (PD) is a neurodegenerative disorder that affects multiple neurological systems. Traditional PD assessment is conducted by a physician during infrequent clinic visits. Using smartphones, remote patient monitoring has the potential to obtain objective behavioral data semi-continuously, track disease fluctuations, and avoid rater dependency.

METHODS

Smartphones collect sensor data during various active tests and passive monitoring, including balance (postural instability), dexterity (skill in performing tasks using hands), gait (the pattern of walking), tremor (involuntary muscle contraction and relaxation), and voice. Some of the features extracted from smartphone data are potentially associated with specific PD symptoms identified by physicians. To leverage large-scale cross-modality smartphone features, we propose a machine-learning framework for performing automated disease assessment. The framework consists of a two-step feature selection procedure and a generic model based on the elastic-net regularization.

RESULTS

Using this framework, we map the PD-specific architecture of behaviors using data obtained from both PD participants and healthy controls (HCs). Utilizing these atlases of features, the framework shows promises to (a) discriminate PD participants from HCs, and (b) estimate the disease severity of individuals with PD.

SIGNIFICANCE

Data analysis results from 437 behavioral features obtained from 72 subjects (37 PD and 35 HC) sampled from 17 separate days during a period of up to six months suggest that this framework is potentially useful for the analysis of remotely collected smartphone sensor data in individuals with PD.

A Deep Learning Framework for the Remote Detection of Parkinson'S Disease Using Smart-Phone Sensor Data

The assessment of Parkinson's disease (PD) using wearable sensors in non-clinical environments presents an opportunity for objective disease classification and severity prediction on a high-frequency and longitudinal basis. However, many challenges exist in analysing remotely collected data due to many sources of data corruption. Using a cohort of 1,815 participants (866 controls and 949 with PD) we implement a range of classification algorithms on Alternate Finger Tapping test data collected on smart-phones in remote environments. We compare the disease classification ability of two traditional machine learning methods against two state-of-the-art deep learning approaches, determining if the latter is successful without the definition of an explicit feature set. We find the deep learning approaches capable of disease classification, often outperforming traditional methods. We show similarities between the participants successfully classified through use of a manually extracted feature set, and the features learnt by a convolutional neural network. Finally, we discuss the broader challenges of analysing remotely collected datasets whilst highlighting the suitability of deep learning for assessing PD when large participant numbers are available.

Predicting motor, cognitive & functional impairment in Parkinson's

OBJECTIVE

We recently demonstrated that 998 features derived from a simple 7-minute smartphone test could distinguish between controls, people with Parkinson's and people with idiopathic Rapid Eye Movement sleep behavior disorder, with mean sensitivity/specificity values of 84.6-91.9%. Here, we investigate whether the same smartphone features can be used to predict future clinically relevant outcomes in early Parkinson's.

METHODS

A total of 237 participants with Parkinson's (mean (SD) disease duration 3.5 (2.2) years) in the Oxford Discovery cohort performed smartphone tests in clinic and at home. Each test assessed voice, balance, gait, reaction time, dexterity, rest, and postural tremor. In addition, standard motor, cognitive and functional assessments and questionnaires were administered in clinic. Machine learning algorithms were trained to predict the onset of clinical outcomes provided at the next 18-month follow-up visit using baseline smartphone recordings alone. The accuracy of model predictions was assessed using 10-fold and subject-wise cross validation schemes.

RESULTS

Baseline smartphone tests predicted the new onset of falls, freezing, postural instability, cognitive impairment, and functional impairment at 18 months. For all outcome predictions AUC values were greater than 0.90 for 10-fold cross validation using all smartphone features. Using only the 30 most salient features, AUC values greater than 0.75 were obtained.

INTERPRETATION

We demonstrate the ability to predict key future clinical outcomes using a simple smartphone test. This work has the potential to introduce individualized predictions to routine care, helping to target interventions to those most likely to benefit, with the aim of improving their outcome.

Cortical region-specific sleep homeostasis in mice: effects of time of day and waking experience

Sleep–wake history, wake behaviors, lighting conditions, and circadian time influence sleep, but neither their relative contribution nor the underlying mechanisms are fully understood. The dynamics of electroencephalogram (EEG) slow-wave activity (SWA) during sleep can be described using the two-process model, whereby the parameters of homeostatic Process S are estimated using empirical EEG SWA (0.5–4 Hz) in nonrapid eye movement sleep (NREMS), and the 24 hr distribution of vigilance states. We hypothesized that the influence of extrinsic factors on sleep homeostasis, such as the time of day or wake behavior, would manifest in systematic deviations between empirical SWA and model predictions. To test this hypothesis, we performed parameter estimation and tested model predictions using NREMS SWA derived from continuous EEG recordings from the frontal and occipital cortex in mice. The animals showed prolonged wake periods, followed by consolidated sleep, both during the dark and light phases, and wakefulness primarily consisted of voluntary wheel running, learning a new motor skill or novel object exploration. Simulated SWA matched empirical levels well across conditions, and neither waking experience nor time of day had a significant influence on the fit between data and simulation. However, we consistently observed that Process S declined during sleep significantly faster in the frontal than in the occipital area of the neocortex. The striking resilience of the model to specific wake behaviors, lighting conditions, and time of day suggests that intrinsic factors underpinning the dynamics of Process S are robust to extrinsic influences, despite their major role in shaping the overall amount and distribution of vigilance states across 24 hr.

Machine-learning-derived sleep-wake staging from around-the-ear electroencephalogram outperforms manual scoring and actigraphy

Quantification of sleep is important for the diagnosis of sleep disorders and sleep research. However, the only widely accepted method to obtain sleep staging is by visual analysis of polysomnography (PSG), which is expensive and time consuming. Here, we investigate automated sleep scoring based on a low-cost, mobile electroencephalogram (EEG) platform consisting of a lightweight EEG amplifier combined with flex-printed cEEGrid electrodes placed around the ear, which can be implemented as a fully self-applicable sleep system. However, cEEGrid signals have different amplitude characteristics to normal scalp PSG signals, which might be challenging for visual scoring. Therefore, this study evaluates the potential of automatic scoring of cEEGrid signals using a machine learning classifier ("random forests") and compares its performance with manual scoring of standard PSG. In addition, the automatic scoring of cEEGrid signals is compared with manual annotation of the cEEGrid recording and with simultaneous actigraphy. Acceptable recordings were obtained in 15 healthy volunteers (aged 35 ± 14.3 years) during an extended nocturnal sleep opportunity, which induced disrupted sleep with a large inter-individual variation in sleep parameters. The results demonstrate that machine-learning-based scoring of around-the-ear EEG outperforms actigraphy with respect to sleep onset and total sleep time assessments. The automated scoring outperforms human scoring of cEEGrid by standard criteria. The accuracy of machine-learning-based automated scoring of cEEGrid sleep recordings compared with manual scoring of standard PSG was satisfactory. The findings show that cEEGrid recordings combined with machine-learning-based scoring holds promise for large-scale sleep studies.

Big data in Parkinson's disease: using smartphones to remotely detect longitudinal disease phenotypes

OBJECTIVE

To better understand the longitudinal characteristics of Parkinson's disease (PD) through the analysis of finger tapping and memory tests collected remotely using smartphones.

APPROACH

Using a large cohort (312 PD subjects and 236 controls) of participants in the mPower study, we extract clinically validated features from a finger tapping and memory test to monitor the longitudinal behaviour of study participants. We investigate any discrepancy in learning rates associated with motor and non-motor tasks between PD subjects and healthy controls. The ability of these features to predict self-assigned severity measures is assessed whilst simultaneously inspecting the severity scoring system for floor-ceiling effects. Finally, we study the relationship between motor and non-motor longitudinal behaviour to determine if separate aspects of the disease are dependent on one another.

MAIN RESULTS

We find that the test performances of the most severe subjects show significant correlations with self-assigned severity measures. Interestingly, less severe subjects do not show significant correlations, which is shown to be a consequence of floor-ceiling effects within the mPower self-reporting severity system. We find that motor performance after practise is a better predictor of severity than baseline performance suggesting that starting performance at a new motor task is less representative of disease severity than the performance after the test has been learnt. We find PD subjects show significant impairments in motor ability as assessed through the alternating finger tapping (AFT) test in both the short- and long-term analyses. In the AFT and memory tests we demonstrate that PD subjects show a larger degree of longitudinal performance variability in addition to requiring more instances of a test to reach a steady state performance than healthy subjects.

SIGNIFICANCE

Our findings pave the way forward for objective assessment and quantification of longitudinal learning rates in PD. This can be particularly useful for symptom monitoring and assessing medication response. This study tries to tackle some of the major challenges associated with self-assessed severity labels by designing and validating features extracted from big datasets in PD, which could help identify digital biomarkers capable of providing measures of disease severity outside of a clinical environment.

Heart rate variability in bipolar disorder and borderline personality disorder: a clinical review

Heart rate variability (HRV) in psychiatric disorders has become an increasing area of interest in recent years following technological advances that enable non-invasive monitoring of autonomic nervous system regulation. However, the clinical interpretation of HRV features remain widely debated or unknown. Standardisation within studies of HRV in psychiatric disorders is poor, making it difficult to reproduce or build on previous work. Recently, a Guidelines for Reporting Articles on Psychiatry and Heart rate variability checklist has been proposed to address this issue. Here we assess studies of HRV in bipolar disorder and borderline personality disorder against this checklist and discuss the implication for ongoing research in this area.

Autogenous femoral head as grafting material for mandibular augmentation

Bone grafting is commonly used for augmentation of the atrophic edentulous maxilla and mandible. Although bone substitutes and allogeneic frozen bone grafts have been applied successfully, fresh autogenous bone grafts remain the 'gold standard' in maxillofacial reconstructive surgery. A disadvantage of harvesting autogenous bone is the resulting donor-site morbidity. The authors present a case in which an autogenous femoral head, which was removed because of a prosthetic hip replacement, was used for augmentation of the extreme atrophic mandible. Using this procedure avoids donor-site morbidity.

[A sudden redness and swelling of the face]

Two patients were referred to a department of oral and maxillofacial surgery with a redness and swelling of the face which had suddenly developed together with a mild illness. The diagnosis of erysipelas was made clinically. This skin infection is generally caused by betahaemolytic streptococci group-A. Treatment is generally in the first instance medicinal. The drugs of choice for treating erysipelas in the vast majority of cases are narrow-spectrum penicillins.

RASSF6 is a novel member of the RASSF family of tumor suppressors

RASSF family proteins are tumor suppressors that are frequently downregulated during the development of human cancer. The best-characterized member of the family is RASSF1A, which is downregulated by promoter methylation in 40-90% of primary human tumors. We now identify and characterize a novel member of the RASSF family, RASSF6. Like the other family members, RASSF6 possesses a Ras Association domain and binds activated Ras. Exogenous expression of RASSF6 promoted apoptosis, synergized with activated K-Ras to induce cell death and inhibited the survival of specific tumor cell lines. Suppression of RASSF6 enhanced the tumorigenic phenotype of a human lung tumor cell line. Furthermore, RASSF6 is often downregulated in primary human tumors. RASSF6 shares some similar overall properties as other RASSF proteins. However, there are significant differences in biological activity between RASSF6 and other family members including a discrete tissue expression profile, cell killing specificity and impact on signaling pathways. Moreover, RASSF6 may play a role in dictating the degree of inflammatory response to the respiratory syncytial virus. Thus, RASSF6 is a novel RASSF family member that demonstrates the properties of a Ras effector and tumor suppressor but exhibits biological properties that are unique and distinct from those of other family members.

Radiological sacroiliitis, a hallmark of spondylitis, is linked with CARD15 gene polymorphisms in patients with Crohn's disease

BACKGROUND

Sacroiliitis is a common extraintestinal manifestation of Crohn's disease but its association with the HLA-B27 phenotype is less evident. Polymorphisms in the CARD15 gene have been linked to higher susceptibility for Crohn's disease. In particular, associations have been found with ileal and fibrostenosing disease, young age at onset of disease, and familial cases.

OBJECTIVES

To investigate whether the presence of sacroiliitis in patients with Crohn's disease is linked to the carriage of CARD15 polymorphisms.

METHODS

102 consecutive patients with Crohn's disease were clinically evaluated by a rheumatologist. Radiographs of the sacroiliac joints were taken and assessed blindly by two investigators. The RFLP-PCR technique was used to genotype all patients for three single nucleotide polymorphisms (SNP) in the CARD15 gene. Every SNP was verified by direct sequencing. The HLA-B27 phenotype was determined.

RESULTS

Radiological evidence of sacroiliitis with or without ankylosing spondylitis was found in 23 patients (23%), of whom only three were HLA-B27 positive. In contrast, 78% of patients with sacroiliitis carried a CARD15 variant v 48% of those without sacroiliitis (p = 0.01; odds ratio 3.8 (95% confidence interval, 1.3 to 11.5)). Multivariate analysis (logistic regression) showed that the association between sacroiliitis and CARD15 polymorphisms was independent of other CARD15 related phenotypes (ileal and fibrostenosing disease, young age at onset of disease, familial Crohn's disease) (p = 0.039).

CONCLUSIONS

CARD15 variants were identified as genetic predictors of Crohn's disease related sacroiliitis. An association was demonstrated between these polymorphisms and an extraintestinal manifestation of Crohn's disease.

[A cystic lesion in the mandibular angle]

A 27-year-old male patient was referred by his dentist to a department of Oral and Maxillofacial Surgery, because of a radiolucent lesion in the mandibular angle. There were no clinical signs or symptoms. The orthopantomogram showed a sharp, demarcated, oval (1.5 x 2.5 cm), unilocular radiolucency caudal of the mandibular canal. Additional radiographic evaluation (CT scan, sialogram) revealed an oval depression in the lingual cortex of the mandible filled with salivary gland tissue. The diagnosis Stafne defect was made. At radiographic follow-up after 1 year, no progression of the lesion was seen. Treatment is not needed.

Differential expression of the early lung cancer detection marker, heterogeneous nuclear ribonucleoprotein-A2/B1 (hnRNP-A2/B1) in normal breast and neoplastic breast cancer

Heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNP-A2/B1) is highly expressed during critical stages of lung development and carcinogenesis. To determine if the expression of hnRNP-A2/B1 is an informative biomarker in breast carcinogenesis, we analyzed hnRNP-A2/B1 overexpression by immunohistochemistry in archived specimens. Expression was detected in 48/85 (56.5%) primary invasive breast cancers and 7/72 (9.7%) specimens of normal breast tissue. Northern analysis of breast cancer cells also demonstrated higher level of hnRNP-A2/B1 expression compared to normal or transformed breast cells. Expression of hnRNP-A2/B1 in breast cancer cells was decreased by exposure to retinoids coordinately with decreased cell growth. These results warrant further evaluation of hnRNP-A2/B1 as a marker of breast carcinogenesis.

Ras uses the novel tumor suppressor RASSF1 as an effector to mediate apoptosis

Although activated Ras proteins are usually associated with driving growth and transformation, they may also induce senescence, apoptosis, and terminal differentiation. The subversion of these anti-neoplastic effects during Ras-dependent tumor development may be as important as the acquisition of the pro-neoplastic effects. None of the currently identified potential Ras effector proteins can satisfactorily explain the apoptotic action of Ras. Consequently, we have sought to identify novel Ras effectors that may be responsible for apoptosis induction. By examining the EST data base, we identified a potential Ras association domain in the tumor suppressor RASSF1. We now show that RASSF1 binds Ras in a GTP-dependent manner, both in vivo and directly in vitro. Moreover, activated Ras enhances and dominant negative Ras inhibits the cell death induced by transient transfection of RASSF1 into 293-T cells. This cell death appears to be apoptotic in nature, as RASSF1-transfected 293-T cells exhibit membrane blebbing and can be rescued by the addition of a caspase inhibitor. Thus, the RASSF1 tumor suppressor may serve as a novel Ras effector that mediates the apoptotic effects of oncogenic Ras.

Relationship of arachidonic acid metabolizing enzyme expression in epithelial cancer cell lines to the growth effect of selective biochemical inhibitors

Arachidonic acid (AA) metabolizing enzymes are emerging as significant mediators of growth stimulation for epithelial cells. The relative contribution of the various family members of AA metabolizing enzymes to epithelial cancer cell growth is not known. To study this question, we first analyzed a series of epithelial cancer cells to establish the relative frequency of expression for the various enzymes. We analyzed the expression of five AA metabolizing enzymes as well as 5-lipoxygenase activating protein (FLAP) in a panel of human epithelial cancer cell lines (n = 20) using reverse transcription-PCR. From this analysis, we found that cyclooxygenase-1 (COX-1), 5-lipoxygenase (5-LOX), and FLAP were universally expressed in all cancer cell lines tested. For the remaining enzymes, the expression of COX-2, 12-LOX, and 15-LOX varied among cell lines, 60, 35, and 90%, respectively. Although the pattern of expression varied among the different cell types, all of the enzymes were expressed in all major cancer histologies. Using a panel of selective biochemical AA metabolizing enzyme inhibitors, we then evaluated the effect of these agents on cell lines with known expression status for the AA metabolizing enzymes. For the enzymes that were not universally expressed, growth inhibition by selective biochemical inhibitors did not closely correlate with the expression status of specific enzymes (P > 0.05). For the universally expressed enzymes, the LOX inhibitors were more potent growth inhibitors than the COX inhibitors. The frequent expression of the AA metabolizing enzymes suggests that AA metabolism pathway may be modulated in response to xenobiotic exposure during carcinogenesis. Although establishing a priori AA metabolizing enzyme status was not consistently informative about what AA metabolizing enzyme inhibition would be most growth inhibitory, the frequent inhibition of many epithelial cancers by these biochemical inhibitors opens a new avenue for cancer therapy and intervention in carcinogenesis.

Phenotypic and functional characterization of a panel of cytotoxic murine NK cell clones that are heterogeneous in their enhancement of Ig secretion in vitro

NK cells not only function as cytotoxic effector cells, but also have immunoregulatory roles including the enhancement of Ig secretion. To have a stable and uniform population of NK cells to study their role in Ig secretion, we generated murine NK clones. Thus, culture of splenocytes from mice that were homozygous for a mutation in the p53 tumor suppressor gene (p53-KO) with IL-2 and poly(IC) resulted in a long-term NK line, from which four stable clones were derived. This approach also yielded a long-term NK line from splenocytes of normal C57BL/6 mice. Identification of the clones as members of the NK lineage was based on large granular morphology, expression of NK-TR and absence of TCR gene rearrangement. Flow cytometry revealed that all clones expressed IL-2R alpha and beta, chains and B220, but no CD3, NK1.1, DX5 or Ly-49. RT-PCR analysis showed heterogeneity in NK1.1 gene expression, and demonstrated expression of perforin and several granzymes in all clones. Three out of four clones lysed YAC-1, but not P815 target cells, corresponding to a pattern of NK specificity. All NK clones enhanced Ig secretion in an in vitro model for T cell-independent type 2 antigens, albeit to varying degrees. We found no correlation between the degree of helper activity of the NK clones and the level of their cytotoxic activity on YAC-1 targets. Thus, we established murine NK clones, and show that they mediate both cytotoxicity and enhancement of Ig secretion.

Peptide-amidating enzymes are expressed in the stellate epithelial cells of the thymic medulla

C-terminal amidation is a post-translational processing step necessary to convey biological activity to a large number of regulatory peptides. In this study we have demonstrated that the peptidyl-glycine alpha-amidating monooxygenase enzyme complex (PAM) responsible for this activity is located in the medullary stellate epithelial cells of the thymus and in cultured epithelial cells bearing a medullary phenotype, using Northern blot, immunocytochemistry, in situ hybridization, and enzyme assays. Immunocytochemical localization revealed a granular pattern in the cytoplasm of the stellate cells, which were also positive for cytokeratins and a B-lymphocyte-associated antigen. The presence of PAM activity in medium conditioned by thymic epithelial cell lines suggests that PAM is a secreted product of these cells. Among the four epithelial cell lines examined, there was a direct correlation between PAM activity and content of oxytocin, an amidated peptide. Taken together, these data provide convincing evidence that thymic epithelial cells have the capacity to generate amidated peptides that may influence T-cell differentiation and suggest that the amidating enzymes could play an important role in the regulation of thymic physiology.

Purification and characterization of a protein that permits early detection of lung cancer. Identification of heterogeneous nuclear ribonucleoprotein-A2/B1 as the antigen for monoclonal antibody 703D4

We have reported that a mouse monoclonal antibody 703D4, detects lung cancer 2 years earlier than routine chest x-ray or cytomorphology. We purified the 703D4 antigen to elucidate its role in early lung cancer biology, using Western blot detection after SDS-polyacrylamide gel electrophoresis. Purification steps included anion exchange chromatography, preparative isoelectric focusing, polymer-based C18-like, and analytical C4 reverse phase high performance liquid chromatography. After 25-50,000-fold purification, the principal immunostaining protein was > 95% pure by Coomassie staining. The NH2 terminus was blocked, so CNBr digestion was used to generate internal peptides. Three sequences, including one across a site of alternate exon splicing, all identified a single protein, heterogeneous nuclear ribonucleoprotein-A2 (hnRNP-A2). A minor co-purifying immunoreactive protein resolved at the final C4 high performance liquid chromatography step is the splice variant hnRNP-B1. Northern analysis of RNA from primary normal bronchial epithelial cells demonstrated a low level of hnRNP-A2/B1 expression, consistent with immunohistochemical staining of clinical samples, and increased hnRNP-A2/B1 expression was found in lung cancer cells. hnRNP-A2/B1 expression is under proliferation-dependent control in normal bronchial epithelial cell primary cultures, but not in SV40-transformed bronchial epithelial cells or tumor cell lines. With our clinical data, this information suggests that hnRNP-A2/B1 is an early marker of lung epithelial transformation and carcinogenesis.

Growth control of lung cancer by interruption of 5-lipoxygenase-mediated growth factor signaling

Signal transduction pathways shared by different autocrine growth factors may provide an efficient approach to accomplish clinically significant control of lung cancer growth. In this study, we demonstrate that two autocrine growth factors activate 5-lipoxygenase action of the arachidonic acid metabolic pathway in lung cancer cell lines. Both growth factors increased the production of 5(S)-hydrooxyeicosa-6E,8Z,11Z,14Z-tetraeno ic acid (5-HETE), a major early 5-lipoxygenase metabolic product. Exogenously added 5-HETE stimulated lung cancer cell growth in vitro. Inhibition of 5-lipoxygenase metabolism by selective antagonists resulted in significant growth reduction for a number of lung cancer cell lines. Primary clinical specimens and lung cancer cell lines express the message for the 5-lipoxygenase enzymes responsible for the generation of active metabolites. In vivo evaluation demonstrated that interruption of 5-lipoxygenase signaling resulted in enhanced levels of programmed cell death. These findings demonstrate that 5-lipoxygenase activation is involved with growth factor-mediated growth stimulation for lung cancer cell lines. Pharmacological intervention with lipoxygenase inhibitors may be an important new clinical strategy to regulate growth factor-dependent stages of lung carcinogenesis.

Expression in human lung cancer cell lines of genes of prohormone processing and the neuroendocrine phenotype

Lung tumor cells and cell lines, principally the histologically classified small cell lung cancer, are characterized by the expression of neuroendocrine (NE) features including AADC (aromatic amino acid decarboxylase, previously called DOPA decarboxylase) and the production of many peptide hormones. The general mechanisms by which most aspects of the NE phenotype affect the clinical behavior of lung tumor cells are unknown, but it is well recognized that peptide hormones can have systemic effects (paraneoplastic syndromes) and several have been shown to be autocrine growth factors for cancer cells. In order to determine the relationship between expression of different aspects of the NE phenotype in lung cancer cell lines, we have compared expression of a gene required for biosynthesis of some active peptide hormones (PAM, peptidylglycine alpha-amidating monooxygenase) to the gene for AADC in 32 lung cancer cell lines. Expression of these genes was quantified by both steady state Northern blot analysis and radiochemical enzymatic activity measurements. To ensure a range of expression of NE markers, non-small cell lung cancer (NSCLC) cell lines were chosen to include several which had previously been shown to express NE markers, and several small cell lung cancer (SCLC) cell lines with previous low levels of AADC were included. PAM enzyme activity and Northern blot analysis showed a two to three log variation in levels of expression in both the small cell and non-small cell lines. A smaller range was found for AADC expression. Using the highly sensitive PAM enzyme assays, all cell lines were found to express detectable PAM. PAM activities were secreted into the growth medium of all cell lines. There was no simple correlation apparent between AADC and PAM gene expression in the lung cancer cell lines. However, classic small cell lines demonstrated high levels of expression of both PAM and AADC genes, as did the carcinoid subset of the NSCLC lines. NSCLC lines expressed levels of PAM mRNA and enzyme activities equivalent to those of SCLC but had infrequent expression of AADC (principally only carcinoid NSCLC expressed AADC). These data demonstrate that separate aspects of the NE phenotype can be differentially expressed in lung cancer histological sub-types. Expression of PAM enzymes in all sub-types of lung cancer suggests that peptide prohormone activation may be a common mechanism for autocrine growth stimulation even in non-Ne NSCLC cell lines, or may reflect maintenance in cell lines of a common pathway of lung tumor promotion.

Human peptidylglycine alpha-amidating monooxygenase transcripts derived by alternative mRNA splicing of an unreported exon

We are characterizing the alternatively spliced human peptidylglycine alpha-amidating monooxygenase (hPAM)-encoding mRNA transcripts expressed by human cells. Reverse transcription coupled to the polymerase chain reaction (RT-PCR) has been used to identify four alternatively spliced variants that differ in the region joining the two catalytic domains. Two of the transcripts represent previously reported splice variants differentiated by the presence (hPAM-A) or absence (hPAM-B) of a 321-nucleotide (nt) linker (optional exon A) which in the rat produce functionally distinct enzymes. Different mRNAs represent two splice variants, hPAM-C and hPAM-D, that show the presence of an exon unreported for PAM in any other species. This new exon, designated exon C, is 54 nt in length, encodes an 18-amino-acid (aa) peptide containing a conserved dibasic aa endoproteolytic processing motif, and is located 3' of exon A in human genomic DNA. We propose that cell-specific regulation of mRNA splicing would provide a mechanism for control of prohormone activation by these variants of the PAM enzyme.