Comparison of background parenchymal enhancement and fibroglandular density at breast magnetic resonance imaging between BRCA gene mutation carriers and non-carriers.

OBJECTIVE: High background parenchymal enhancement and amount of fibroglandular tissue on breast magnetic resonance imaging are related to increased breast cancer risk. This study sought to compare these parameters between BRCA mutation carriers and non-carriers and to evaluate the potential implications of the findings for short term follow-up. MATERIALS AND METHODS: Magnetic resonance imaging studies of known BRCA mutation carriers, were compared to age-matched non-carrier studies performed in the same center during the same period. The groups were compared for qualitative background parenchymal enhancement and amount of fibroglandular tissue using the Breast Imaging Reporting and Data System (BI-RADS). RESULTS: Breast parenchymal enhancement was high in up to one-third of the cohort: 22% of carriers and 33% of controls (p = 0.013). These results were sustained on separate analysis of menstrual-cycle-timed examinations. Amount of fibroglandular tissue was high in most cases: 62% of carriers and 75% of controls (p = 0.004). A BI-RADS final assessment score of 3 was more common in patients with high parenchymal enhancement, especially controls. CONCLUSION: BRCA mutation carriers demonstrated lower levels of breast parenchymal enhancement and amount of fibroglandular tissue than age-matched non-carriers. These differences are probably influenced by hormonal status, as well as highlight different risks in distinctive subgroups of breast cancer (hormone-enriched, mutation-associated defective DNA damage repair), affecting considerations of preventive medical treatment. Differences in the indications for imaging between the carrier and non-carrier groups (screening for mutations and breast cancer evaluation, respectively) probably accounted for the higher rate of BI-RADS 3 in the control group.

Malignancy Detection on Mammography Using Dual Deep Convolutional Neural Networks and Genetically Discovered False Color Input Enhancement.

Breast cancer is the most prevalent malignancy in the US and the third highest cause of cancer-related mortality worldwide. Regular mammography screening has been attributed with doubling the rate of early cancer detection over the past three decades, yet estimates of mammographic accuracy in the hands of experienced radiologists remain suboptimal with sensitivity ranging from 62 to 87% and specificity from 75 to 91%. Advances in machine learning (ML) in recent years have demonstrated capabilities of image analysis which often surpass those of human observers. Here we present two novel techniques to address inherent challenges in the application of ML to the domain of mammography. We describe the use of genetic search of image enhancement methods, leading us to the use of a novel form of false color enhancement through contrast limited adaptive histogram equalization (CLAHE), as a method to optimize mammographic feature representation. We also utilize dual deep convolutional neural networks at different scales, for classification of full mammogram images and derivative patches combined with a random forest gating network as a novel architectural solution capable of discerning malignancy with a specificity of 0.91 and a specificity of 0.80. To our knowledge, this represents the first automatic stand-alone mammography malignancy detection algorithm with sensitivity and specificity performance similar to that of expert radiologists.

Do Prenatal Corticosteroids Affect Brain Maturation of the Premature Infant? An Electroencephalography Study.

OBJECTIVE: To assess whether prenatal treatment with betamethasone has a significant influence on cerebral maturation indices as measured by electroencephalographic (EEG) indices. STUDY DESIGN: Infants born less than 35 weeks postmenstrual age (PMA) were prospectively enrolled if their mother received a full course of bethametasone prior to delivery (study group) or not (control group); infants with major intracranial abnormalities were excluded as well as those who were sedated or needed assisted ventilation. EEG was recorded during the first 10 days of life. Interburst intervals and maximal amplitudes of theta and delta bandwidths were calculated by a signal processing software. A multivariate general linear model was used to analyze the relationship between the 2 groups and the different electrophysiologic parameters, adjusting for PMA and mode of delivery. RESULTS: Thirty-eight infants were included in the study group and 36 in the control group. Univariate analysis demonstrated a negative correlation between PMA at test and EEG indices (interburst interval and delta and theta frequencies). Multivariate analysis demonstrated a less robust correlation of PMA and EEG indices and a positive correlation of prenatal betamethasone treatment with Theta frequencies. Repeating the data analysis separately for each study group, the above results remained significant mainly in the study group. CONCLUSIONS: Our findings suggest a possible stabilization effect of corticosteroids on the central nervous system and a possible delay of the maturation of cerebral activity related to prenatal corticosteroids use. These findings may relate to a better neurodevelopmental outcome of infants treated prenatally with corticosteroids.

Influence of antiepileptic drugs on amplitude-integrated electroencephalography.

Amplitude-integrated electroencephalography monitors different aspects of cerebral function in neonatal intensive care units. To examine the influence of various antiepileptic drugs on the background patterns and voltage of amplitude-integrated electroencephalography recordings, we screened 191 tracing segments originating from 77 newborns treated with antiepileptic drugs. The influences of lorazepam, diazepam, and phenobarbital given as bolus doses, and midazolam and lidocaine given in continuous infusion, were examined. Voltages and patterns before and after drug administration were assessed. Time taken to return to previous voltage was assessed in clinically significant cases. Chi-square and Wilcoxon tests were used for statistical analyses. Significant changes were evident after lorazepam, diazepam, phenobarbital, and midazolam administration. Depending on the voltage-assessment method, a clinically significant depression of the lower voltage border occurred in 25-35% of tracings, and of the upper border in 16-32%. In 12% of tracings, change to a worse pattern was noted. The average time for recovery to predrug administration voltage was 2.5 hours (range, 15 minutes to 15 hours). Changes in amplitude-integrated electroencephalography tracings occur after antiepileptic drugs are infused. These changes include deterioration of pattern and depression of voltage that may persist for a considerable period. The potential depressing effects of these drugs should be taken into consideration when assessing amplitude-integrated electroencephalogram tracings.