Search

Bovine twin birth is associated with detriments, including increased embryo/fetal losses, malpresentation, and dystocia. Incidence of these is lessened in bilateral compared with unilateral twin pregnancy. This study was undertaken to assess the use of follicular ablation by aspiration to create bilateral twin pregnancies in females with genetic potential for ~3.5 ovulations per cycle (Trio allele carriers). In experiment 1, carriers (n = 30) and noncarriers (n = 10) were synchronized for ovulation and timed artificial insemination (TAI). Follicles (>5 mm) in excess of one per ovary were aspirated ~16 h preceding TAI. Follicle count for females with follicles on only one ovary was reduced to two. Blood was sampled 2 wk post-TAI to assess progesterone (P4) concentrations; embryo count was determined by ultrasound 6 wk post-TAI. Circulating P4 concentration post-TAI was significantly (P < 0.001) associated with both genotype and subsequent pregnancy status (pregnant noncarriers: 7.06 ± 0.68 ng/mL; pregnant carriers: 5.54 ± 0.55 ng/mL; nonpregnant noncarriers: 5.22 ± 1.05 ng/mL; nonpregnant carriers: 3.13 ± 0.42 ng/mL). Experiment 2 was undertaken to offset the negative effects of follicular aspiration on subsequent P4 concentration observed in experiment 1. Carriers (n = 38) and noncarriers (n = 32) were submitted to TAI and follicle ablation as described for experiment 1. Additionally, accessory corpora lutea (CL) were induced in carriers by the administration of human chorionic gonadotropin (carriers) at day 6 post-TAI. Consequently, P4 concentration post-TAI was significantly (P < 0.05) associated with subsequent pregnancy status (pregnant: 8.48 ± 0.61 ng/mL; nonpregnant: 6.70 ± 0.63 ng/mL) but not with genotype (carrier: 8.01 ± 0.59 ng/mL; noncarrier: 7.17 ± 0.64 ng/mL). Embryo number was greater in carriers (exp. 1: 1.64 ± 0.81; exp 2: 1.45 ± 0.09) vs. noncarriers (1.00 ± 0.00, both experiments). Single, twin, and triplet pregnancies occurred in carriers in experiment 1, whereas multiples in experiment 2 were limited to twin pregnancies. Genotype effects on pregnancy rate were not significant (P > 0.10) in either experiment. Results suggest that follicular ablation to create bilateral twin pregnancies in Trio carriers is feasible but requires the induction of accessory CL to offset the negative effects of follicular aspiration on subsequent P4 concentration and associated fertility outcomes.

This study aimed to validate a 7 d progesterone (P4)-based fixed-time AI (FTAI) protocol for Bos indicus cattle by comparing to 8 and 9 d-type protocols. The first study compared 7 vs. 8 d protocols in Nelore heifers (Exp. 1.1; n = 742) and cows (Exp. 1.2; n = 2488), and the second study compared 7 vs. 9 d protocols in cows (Exp. 2; n = 1343). On experimental Day −10 and Day −11 the 8 and 9 d groups received an intravaginal P4 implant, 2.0 mg estradiol benzoate (EB) and 0.5 mg cloprostenol sodium (PGF). On Day −9 the 7 d group received the same treatments (P4, EB, and PGF). Then, on Day −2 all groups had the P4 implants removed, and PGF, 0.6 mg estradiol cypionate, and 300 IU equine chorionic gonadotropin (eCG) was administered. Fixed-time AI was performed 48 h later (Day 0) and 8.4 mg buserelin acetate (GnRH) was administered to 7d-G, 8d-G and 9d-G groups, whereas 7d-0, 8d-0 and 9d-0 groups did not receive GnRH at AI. Estrus was detected using tail-chalk between Day −2 and Day 0. Pregnancy per AI (P/AI) was evaluated by ultrasound 30 d after AI. Effects were considered significant when P ≤ 0.05, whereas a tendency was designated when P ≤ 0.10 and P > 0.05. In heifers (Exp. 1.1), incidence of estrus was similar regardless of protocol length (7 or 8 d). There was no independent treatment effect on P/AI or interaction between protocol length and GnRH at AI for P/AI (7d-0: 46.9, 7d-G: 51.4, 8d-0: 47.7, and 8d-G: 43.6%). Heifers in estrus had greater P/AI, and GnRH had no additional effect. More cows (Exp. 1.2) from the 8 d protocol were in estrus than cows submitted to the 7 d protocol. Additionally, despite no interaction between protocol length and GnRH on P/AI (7d-0: 55.9, 7d-G: 60.9, 8d-0: 56.2, and 8d-G: 60.8%), GnRH at AI increased P/AI. There was no interaction between estrus and GnRH, but cows displaying estrus had greater P/AI. Cows not expressing estrus tended (P = 0.06) to have greater P/AI when receiving GnRH. In Exp. 2, more 9 d cows were in estrus than 7 d cows. Protocol length did not affect P/AI but tended (P = 0.08) to interact with GnRH (7d-G had greater P/AI [57.9%] than 7d-0 [47.6%], but 9d-0 [54.6%] and 9d-G [55.4%] were not different from other groups). Moreover, GnRH increased P/AI only for the 7 d protocol. No interaction between estrus and GnRH was detected but estrus improved P/AI, and GnRH tended (P = 0.09) to improve P/AI of cows in estrus. In conclusion, despite longer protocols being more conducive to expression of estrus, there were no detectable effects of protocol length on P/AI. In addition, GnRH at FTAI may improve fertility in cows, particularly when cows are treated with shorter protocols.

This paper extends a Support Vector Machine (SVM) approach for the detection of seismic events, at the level of a seismic station. In previous works, it was shown that this approach produced excellent results, in terms of the Recall and Specificity measures, whether applied off-line or in a continuous scheme. The drawback was the time taken for achieving the detection, too large to be applied in a Early-Warning System (EWS). This paper shows that, by using alternative input features, a similar performance can be obtained, with a significant reduction in detection time. Additionally, it is experimentally proved that, whether off-line or in continuous operation, the best results are obtained when the SVM detector is trained with data originated from the respective seismic station.

This study describes research to design a seismic detection system to act at the level of a seismic station, providing a similar role to that of STA/LTA ratio-based detection algorithms. In a first step, Multi-Layer Perceptrons (MLPs) and Support Vector Machines (SVMs), trained in supervised mode, were tested. The sample data consisted of 2903 patterns extracted from records of the PVAQ station, one of the seismographic network’s stations of the Institute of Meteorology of Portugal (IM). Records’ spectral variations in time and characteristics were reflected in the input ANN patterns, as a set of values of power spectral density at selected frequencies. To ensure that all patterns of the sample data were within the range of variation of the training set, we used an algorithm to separate the universe of data by hyper-convex polyhedrons, determining in this manner a set of patterns that have a mandatory part of the training set. Additionally, an active learning strategy was conducted, by iteratively incorporating poorly classified cases in the training set. The proposed system best results, in terms of sensitivity and selectivity in the whole data ranged between 98% and 100%. These results compare very favourably with the ones obtained by the existing detection system, 50%, and with other approaches found in the literature. Subsequently, the system was tested in continuous operation for unseen (out of sample) data, and the SVM detector obtained 97.7% and 98.7% of sensitivity and selectivity, respectively. The classifier presented 88.4% and 99.4% of sensitivity and selectivity when applied to data of a different seismic station of IM. Due to the input features used, the average time taken for detection with this approach is in the order of 100 s. This is too long to be used in an early-warning system. In order to decrease this time, an alternative set of input features was tested. A similar performance was obtained, with a significant reduction in the average detection time (around 1.3 s). Additionally, it was experimentally proved that, whether off-line or in continuous operation, the best results are obtained when the SVM detector is trained with data originated from the respective seismic station.

This study describes the on-line operation of a seismic detection system to act at the level of a seismic station providing similar role to that of a STA /LTA ratio-based detection algorithms. The intelligent detector is a Support Vector Machine (SVM), trained with data consisting of 2903 patterns extracted from records of the PVAQ station, one of the seismographic network's stations of the Institute of Meteorology of Portugal (IM). Records' spectral variations in time and characteristics were reflected in the SVM input patterns, as a set of values of power spectral density at selected frequencies. To ensure that all patterns of the sample data were within the range of variation of the training set, we used an algorithm to separate the universe of data by hyper-convex polyhedrons, determining in this manner a set of patterns that have a mandatory part of the training set. Additionally, an active learning strategy was conducted, by iteratively incorporating poorly classified cases in the training set. After having been trained, the proposed system was experimented in continuous operation for unseen (out of sample) data, and the SVM detector obtained 97.7% and 98.7% of sensitivity and selectivity, respectively. The same type of ANN presented 88.4 % and 99.4% of sensitivity and selectivity when applied to data of a different seismic station of IM. © 2013 Springer-Verlag Berlin Heidelberg.

Abstract This experimental study focuses on a detection system at the seismic station level that should have a similar role to the detection algorithms based on the ratio STA/LTA. We tested two types of neural network: Multi-Layer Perceptrons and Support Vector Machines, trained in supervised mode. The universe of data consisted of 2903 patterns extracted from records of the PVAQ station, of the seismography network of the Institute of Meteorology of Portugal. The spectral characteristics of the records and its variation in time were reflected in the input patterns, consisting in a set of values of power spectral density in selected frequencies, extracted from a spectrogram calculated over a segment of record of pre-determined duration. The universe of data was divided, with about 60% for the training and the remainder reserved for testing and validation. To ensure that all patterns in the universe of data were within the range of variation of the training set, we used an algorithm to separate the universe of data by hyper-convex polyhedrons, determining in this manner a set of patterns that have a mandatory part of the training set. Additionally, an active learning strategy was conducted, by iteratively incorporating poorly classified cases in the training set. The best results, in terms of sensitivity and selectivity in the whole data ranged between 98% and 100%. These results compare very favorably with the ones obtained by the existing detection system, 50%.