Zero-shot transfer learned generic AI models for prediction of optimally ripe climacteric fruits.

Ideally, ripe fruits offer appropriate nutritional content and best quality in terms of taste and flavour. Prediction of ripe climacteric fruits acts as the main marketing indicator for quality from the consumer perspective and thus renders it a genuine industrial concern for all the stakeholders of the fruit supply chain. However, the building of fruit-specific individual model for the prediction of ripeness level remains an existing challenge due to the scarcity of sufficient labeled experimental data for each fruit. This paper describes the development of generic AI models based on the similarity in physico-chemical degradation phenomena of climacteric fruits for prediction of 'unripe' and 'ripe' levels using 'zero-shot' transfer learning techniques. Experiments were performed on a variety of climacteric and non-climacteric fruits, and it was observed that transfer learning works better for fruits within a cluster (climacteric fruits) as compared to across clusters (climacteric to non-climacteric fruits). The main contributions of this work are two-fold (i) Using domain knowledge of food chemistry to label the data in terms of age of the fruit, (ii) We hypothesize and prove that the zero-shot transfer learning works better within a set of fruits, sharing similar degradation chemistry depicted by their visual properties like black spot formations, wrinkles, discoloration, etc. The best models trained on banana, papaya and mango dataset resulted in s zero-shot transfer learned accuracies in the range of 70 to 82 for unknown climacteric fruits. To the best of our knowledge, this is the first study to demonstrate the same.

Data based predictive models for odor perception.

Machine learning and data analytics are being increasingly used for quantitative structure property relation (QSPR) applications in the chemical domain where the traditional Edisonian approach towards knowledge-discovery have not been fruitful. The perception of odorant stimuli is one such application as olfaction is the least understood among all the other senses. In this study, we employ machine learning based algorithms and data analytics to address the efficacy of using a data-driven approach to predict the perceptual attributes of an odorant namely the odorant characters (OC) of "sweet" and "musky". We first analyze a psychophysical dataset containing perceptual ratings of 55 subjects to reveal patterns in the ratings given by subjects. We then use the data to train several machine learning algorithms such as random forest, gradient boosting and support vector machine for prediction of the odor characters and report the structural features correlating well with the odor characters based on the optimal model. Furthermore, we analyze the impact of the data quality on the performance of the models by comparing the semantic descriptors generally associated with a given odorant to its perception by majority of the subjects. The study presents a methodology for developing models for odor perception and provides insights on the perception of odorants by untrained human subjects and the effect of the inherent bias in the perception data on the model performance. The models and methodology developed here could be used for predicting odor characters of new odorants.

Human ß-defensin-1: A natural antimicrobial peptide present in amniotic fluid that is increased in spontaneous preterm labor with intra-amniotic infection.

PROBLEM: Human ß-defensins (HBDs) are antimicrobial peptides that participate in the soluble innate immune mechanisms against infection. Herein, we determined whether HBD-1 was present in amniotic fluid during normal pregnancy and whether its concentrations change with intra-amniotic inflammation and/or infection. METHOD OF STUDY: Amniotic fluid was collected from 219 women in the following groups: (a) midtrimester who delivered at term (n = 35); (b) term with (n = 33) or without (n = 17) labor; (c) preterm labor with intact membranes who delivered at term (n = 29) or who delivered preterm with (n = 19) and without (n = 29) intra-amniotic inflammation and infection or with intra-amniotic inflammation but without infection (n = 21); and (d) preterm prelabor rupture of membranes (pPROM) with (n = 19) and without (n = 17) intra-amniotic inflammation/infection. Amniotic fluid HBD-1 concentrations were determined using a sensitive and specific ELISA kit. RESULTS: (a) HBD-1 was detectable in all amniotic fluid samples; (b) amniotic fluid concentrations of HBD-1 were changed with gestational age (midtrimester vs term no labor), being higher in midtrimester; (c) amniotic fluid concentrations of HBD-1 were similar between women with and without spontaneous labor at term; (d) among patients with spontaneous preterm labor, amniotic fluid concentrations of HBD-1 in women with intra-amniotic inflammation/infection and in those with intra-amniotic inflammation without infection were greater than in women without intra-amniotic inflammation or infection who delivered preterm or at term; and (e) the presence of intra-amniotic inflammation and infection in patients with pPROM did not change amniotic fluid concentrations of HBD-1. CONCLUSION: HBD-1 is a physiological constituent of amniotic fluid that is increased in midtrimester during normal pregnancy and in the presence of culturable microorganisms in the amniotic cavity. These findings provide insight into the soluble host defense mechanisms against intra-amniotic infection.

Early Evaluation of the Fetal Heart.

Evaluation of the fetal heart at 11-13 + 6 weeks of gestation is indicated for women with a family history of congenital heart defects (CHD), a previous child with CDH, or an ultrasound finding associated with cardiac anomalies. The accuracy for early detection of CHD is highly related to the experience of the operator. The 4-chamber view and outflow tracts are the most important planes for identification of an abnormal heart, and can be obtained in the majority of fetuses from 11 weeks of gestation onward. Transvaginal ultrasound is the preferred route for fetal cardiac examination prior to 12 weeks of gestation, whereas, after 12 weeks, the fetal heart can be reliably evaluated by transabdominal ultrasound. Cardiac defects, such as ventricular septal defects, tetralogy of Fallot, Ebstein's anomaly, or cardiac tumors, are unlikely to be identified at ≤14 weeks of gestation. Additional ultrasound techniques such as spatiotemporal image correlation and the evaluation of volumes by a fetal-heart expert can improve the detection of congenital heart disease. The evaluation of the fetal cardiac function at 11-13 + 6 weeks of gestation can be useful for early identification of fetuses at risk of anemia due to hemoglobinopathies, such as hemoglobin Bart's disease.

Intrahepatic Cholestasis of Pregnancy Leading to Severe Vitamin K Deficiency and Coagulopathy.

Intrahepatic cholestasis of pregnancy is seldom associated with significant vitamin K deficiency. We report a case of a 16-year-old primigravid patient at 24 weeks and 3 days of gestation who presented with pruritus, hematuria, and preterm labor. Laboratory work-up showed severe coagulopathy with Prothrombin Time (PT) of 117.8 seconds, International Normalized Ratio (INR) of 10.34, and elevated transaminases suggestive of intrahepatic cholestasis of pregnancy. Her serum vitamin K level was undetectable (<0.1 nMol/L). Initial therapy consisted of intramuscular replacement of vitamin K and administration of fresh frozen plasma. Her hematuria and preterm labor resolved and she was discharged. She presented in active labor and delivered at 27 weeks and 1 day. Her bile acids (93 µ/L) and INR (2.32) had worsened. She delivered a male infant, 1150 grams with Apgar scores 7 and 9. The newborn received 0.5 mg of intramuscular vitamin K shortly after delivery but went on to develop bilateral grade III intraventricular hemorrhages by day 5. Intrahepatic cholestasis in pregnancy and nutrition issues were identified as the main risk factors for the severe coagulopathy of this patient. This case underlines the importance of evaluation of possible severe coagulopathy in patients with intrahepatic cholestasis of pregnancy in order to avoid serious maternal or fetal adverse outcomes.

Management of Liddle Syndrome in Pregnancy: A Case Report and Literature Review.

Liddle syndrome is an autosomal dominant genetic condition that causes hypertension and hypokalemia due to a gain-of-function mutation in the SCNN1B or SCNN1G genes which code for the epithelial sodium channel in the kidney. This leads to increased sodium and water reabsorption causing hypertension. We report a case of a 27-year-old pregnant woman who was admitted for hypertension and hypokalemia and later diagnosed and treated for Liddle syndrome using amiloride. Maintaining a high suspicion of Liddle syndrome in pregnancy is essential in such cases to be able to adequately and effectively treat the hypertension. Due to physiological effects of pregnancy, the dose of amiloride may need to be increased as gestational age progresses up to a maximum dose of 30 mg orally per day.

Successful Continuation of Pregnancy After Treatment of Group A Streptococci Sepsis.

BACKGROUND: Invasive group A streptococci infections in pregnancy have historically led to severe maternal and neonatal morbidity and mortality. We are reporting a rare and novel case of successful treatment of third-trimester group A streptococci infection with early, aggressive intervention and maintenance of the pregnancy to term. CASE: A 35 year old woman initially presented with fever, flu-like symptoms, and preterm contractions at 34 weeks of gestation. She demonstrated signs of early stages of septic shock, ultimately attributed to group A streptococci bacteremia. Early, aggressive intervention allowed the pregnancy to continue until 38 weeks of gestation with normal maternal and neonatal outcomes. CONCLUSION: Early and aggressive treatment of invasive group A streptococci infection during pregnancy can potentially avoid severe maternal and perinatal morbidity and mortality with a successful continuation of pregnancy.

Dynamic Changes in the Myometrium during the Third Stage of Labor, Evaluated Using Two-Dimensional Ultrasound, in Women with Normal and Abnormal Third Stage of Labor and in Women with Obstetric Complications.

OBJECTIVE: To investigate dynamic changes in myometrial thickness during the third stage of labor. METHODS: Myometrial thickness was measured using ultrasound at one-minute time intervals during the third stage of labor in the mid-region of the upper and lower uterine segments in 151 patients including: women with a long third stage of labor (n = 30), postpartum hemorrhage (n = 4), preterm delivery (n = 7) and clinical chorioamnionitis (n = 4). Differences between myometrial thickness of the uterine segments and as a function of time were evaluated. RESULTS: There was a significant linear increase in the mean myometrial thickness of the upper uterine segments, as well as a significant linear decrease in the mean myometrial thickness of the lower uterine segments until the expulsion of the placenta (p < 0.001). The ratio of the measurements of the upper to the lower uterine segments increased significantly as a function of time (p < 0.0001). In women with postpartum hemorrhage, preterm delivery, and clinical chorioamnionitis, an uncoordinated pattern among the uterine segments was observed. CONCLUSION: A well-coordinated activity between the upper and lower uterine segments is demonstrated in normal placental delivery. In some clinical conditions this pattern is not observed, increasing the time for placental delivery and the risk of postpartum hemorrhage.