No abstract

MATERIALS AND METHODSplasia is also found as an isolated anomaly (3,4) or in association with minor deformities compatible with survival, i.e. oligohydramnios deformation sequence following prolonged rupture of the fetal membranes (5).Over 20 years ago, both Blanc et al. (6) and Bain et al. (7) demonstrated, using perinatal autopsy data, that prolonged oligohydramnios, irrespective of its cause, was associated with fetal lung hypoplasia. More recently, experimental evidence derived from work in various mammalian species confirmed this association (8-11). Two important questions remain unanswered: I) what is the mechanism by which lack of amniotic fluid interferes with fetal lung growth? 2) What are the influences of both gestational age at the onset of oligohydramnios, and its duration, on the degree of lung hypoplasia? This report focuses on the latter question using the guinea pig model of oligohydramnios. A preliminary account of these experiments has been published previously (12).Fetal lung growth proceeds in three consecutive stages, i.e. the pseudoglandular, canalicular, and terminal sac stages. Although this sequence is invariable, its timing in gestation varies among species. We induced oligohydramnios between days 40 and 55 of gestation in the guinea pig, a period roughly corresponding to 18 to 29 wk of gestation in the human. The five time-frames studied are illustrated in Figure I, along with the stages of fetal lung development involved. More specifically, the experiments were as follows: experiment A: drainage between days 40 and 45 (5 days); experiment B: drainage between 45 and 50 (5 days); experiment C: drainage between days 50 and 55 (5 days); experiment D: drainage between days 40 and 50 (10 days); experiment E: drainage between days 45 and 55 (10 days).All fetuses were sacrificed at the end of the 5-or 10-days study period to assess fetal and lung growth. In order to overcome the effects of variation between litters, we used untouched littermate fetuses as controls. Whenever possible, untouched littermate fetuses located at the same level in the opposite uterine horn were chosen as paired controls in order to control for the naturally occurring intralitter variation in fetal size.Pregnant guinea pigs (Cavia porcellus, Camm-Hartley strain) were obtained 30 days after mating from the Camm Research Laboratory, Wayne, NJ. Pregnant animals were caged individually and provided with guinea pig chow no. 5025 (Ralston Purina Co.) and tap water ad libitum. Following an overnight fast, the animals were anesthetized with intramuscular injections of ketamine hydrochloride (40 mg/kg) (Ketaset, Bristol Lab.) and acepromazine maleate (3 mg/kg) (Tech America Group, Inc.). After local anesthesia with lidocaine, a midline laparotomy was performed. Both uterine horns were exteriorized and the position of each fetus was recorded. Half the litter was randomly assigned to the experimental group and the other half served as 951 ABSTRACf. We drained amniotic fluid for periods of 5 and 10 days at various times in g...

Background: Eoinophilic Esophagitis (EoE) is a chronic inflammatory condition diagnosed by >=15 eosinophils (Eos) per high-power field (HPF). There is no gold standard for clinical remission and Eo-associated metrics are poorly correlated with symptoms. Deep learning can be used to explore the relationships of tissue features with clinical response. Objectives: To determine if deep learning can elucidate tissue patterns in EoE that predict treatments or symptoms at remission. Methods: We created two deep learning models using esophageal biopsies from histologically normal and EoE patients: one to identify Eos in esophageal biopsies and a second to broadly classify esophageal tissue as EoE vs. normal. We used these models to analyze biopsies at diagnosis and first remission timepoint, as defined by <15 Eos/HPF, in a subset of 19 treatment-naive patients. Differences in deep learning metrics between patient groups were assessed using Wilcoxon Rank-Sum tests. Results: All initial patients were symptomatic at diagnosis and a majority were still suffering from dysphagia at remission. The Eo identification model had a low mean (SD) error of -0.3 (11.5) Eos/HPF. Higher peak and average Eo counts at diagnosis were associated with higher likelihood of being on a food-elimination diet at remission than steroids or proton-pump inhibitor (p<0.05). The EoE classification model had an F1-score of 0.97 for distinguishing normal tissue from EoE. There was a significant decrease from diagnosis in the percentage of EoE-classified tissue among asymptomatic remission patients (p<0.05). Conclusions: Deep learning may have utility in diagnosing EoE and predicting future treatment response at diagnosis and resolution of symptoms at follow-up.

No abstract