Dynamic forecasting of individual cow milk yield in automatic milking systems

Accurate forecasting of dairy cow milk yield is useful to dairy farmers, both in relation to financial planning and for detection of deviating yield patterns, which can be an indicator of mastitis and other diseases. In this study we developed a dynamic linear model (DLM) designed to forecast milk yields of individual cows per milking, as they are milked in milking robots. The DLM implements a Wood's function to account for the expected total daily milk yield. It further implements a second-degree polynomial function to account for the effect of the time intervals between milkings on the proportion of the expected total daily milk yield. By combining these 2 functions in a dynamic framework, the DLM was able to continuously forecast the amount of milk to be produced in a given milking. Data from 169,774 milkings on 5 different farms in 2 different countries were used in this study. A separate farm-specific implementation of the DLM was made for each of the 5 farms. To determine which factors would influence the forecast accuracy, the standardized forecast errors of the DLM were described with a linear mixed effects model (lme). This lme included lactation stage (early, middle, or late), somatic cell count (SCC) level (nonelevated or elevated), and whether or not the proper farm-specific version of the DLM was used. The standardized forecast errors of the DLM were only affected by SCC level and interactions between SCC level and lactation stage. Therefore, we concluded that the implementation of Wood's function combined with a second-degree polynomial is useful for dynamic modeling of milk yield in milking robots, and that this model has potential to be used as part of a mastitis detection system.     

Lack of combinatorial productivity in language processing with simple recurrent networks

An astronomical set of sentences can be produced in natural language by combining relatively simple sentence structures with a human-size lexicon. These sentences are within the range of human language performance. Here, we investigate the ability of simple recurrent networks (SRNs) to handle such combinatorial productivity. We successfully trained SRNs to process sentences formed by combining sentence structures with different groups of words. Then, we tested the networks with test sentences in which words from different training sentences were combined. The networks failed to process these sentences, even though the sentence structures remained the same and all words appeared on the same syntactic positions as in the training sentences. In these combination cases, the networks produced work–word associations, similar to the condition in which words are presented in the context of a random word sequence. The results show that SRNs have serious difficulties in handling the combinatorial productivity that underlies human language performance. We discuss implications of this result for a potential neural architecture of human language processing.     

Implementation of a process analytical technology system in a freeze-drying process using Raman spectroscopy for in-line process monitoring

The aim of the present study was to propose a strategy for the implementation of a Process Analytical Technology system in freeze-drying processes. Mannitol solutions, some of them supplied with NaCl, were used as models to freeze-dry. Noninvasive and in-line Raman measurements were continuously performed during lyophilization of the solutions to monitor real time the mannitol solid state, the end points of the different process steps (freezing, primary drying, secondary drying), and physical phenomena occurring during the process. At-line near-infrared (NIR) and X-ray powder diffractometry (XRPD) measurements were done to confirm the Raman conclusions and to find out additional information. The collected spectra during the processes were analyzed using principal component analysis and multivariate curve resolution. A two-level full factorial design was used to study the significant influence of process (freezing rate) and formulation variables (concentration of mannitol, concentration of NaCl, volume of freeze-dried sample) upon freeze-drying. Raman spectroscopy was able to monitor (i) the mannitol solid state (amorphous, alpha, beta, delta, and hemihydrate), (ii) several process step end points (end of mannitol crystallization during freezing, primary drying), and (iii) physical phenomena occurring during freeze-drying (onset of ice nucleation, onset of mannitol crystallization during the freezing step, onset of ice sublimation). NIR proved to be a more sensitive tool to monitor sublimation than Raman spectroscopy, while XRPD helped to unravel the mannitol hemihydrate in the samples. The experimental design results showed that several process and formulation variables significantly influence different aspects of lyophilization and that both are interrelated. Raman spectroscopy (in-line) and NIR spectroscopy and XRPD (at-line) not only allowed the real-time monitoring of mannitol freeze-drying processes but also helped (in combination with experimental design) us to understand the process.     

A comparative analysis of parallel programming language complexity and performance

Several parallel programming languages, libraries and environments have been developed to ease the task of writing programs for multiprocessors. Proponents of each approach often point out various language features that are designed to provide the programmer with a simple programming interface. However, virtually no data exist that quantitatively evaluate the relative ease of use of different parallel programming languages. The paper borrows techniques from the software engineering field to quantify the complexity of three predominant programming models: shared-memory, message-passing and high-performance Fortran. It is concluded that traditional software complexity metrics are effective indicators of the relative complexity of parallel programming languages. The impact of complexity on run-time performance is also discussed in the context of message-passing vs. HPF on an IBM SP2. © 1998 John Wiley & Sons, Ltd.     

Comparative Investigation of Hoveyda–Grubbs Catalysts bearing Modified N‐Heterocyclic Carbene Ligands

This paper reports the structural modification of Hoveyda–Grubbs complexes through the introduction of either an N‐alkyl‐N′‐mesityl heterocyclic carbene, an N‐alkyl‐N′‐(2,6‐diisopropylphenyl) heterocyclic carbene, or an N‐alkyl‐N′‐alkyl heterocyclic carbene. The effect of the modified N‐heterocyclic carbene (NHC) ligand was investigated in representative ring‐opening metathesis polymerization (ROMP), ring‐closing metathesis (RCM) and cross metathesis (CM) reactions. A pronounced influence on both catalyst activity and selectivity was found to be exerted by the NHC amino substituents, which emphasizes that a rigorously selected steric environment is critical in olefin metathesis catalyst design.     

Effect of catecholamine depletion on myocardial infarct size in dogs: role of catecholamines in ischemic preconditioning

OBJECTIVES: Cardioprotective adaptation to brief periods of ischemia and reperfusion is termed ischemic preconditioning (PC). Limitation of infarct size by preconditioning is associated with marked slowing of ischemic metabolism. The cause of metabolic slowing has not been determined but may involve either pro- or anti-adrenergic mechanisms. Hypothetically, adrenergic stimulation could signal the adaptive response. Alternatively, metabolic slowing during the sustained ischemic challenge could occur through a reduction in beta-adrenergic stimulation. This study was designed to test the role of cardiac norepinephrine (NE) in PC. METHODS: The effect of PC on myocardial infarct size was studied in control dogs and dogs depleted of catecholamines by pretreatment with reserpine (RES; 0.25 mg/kg i.v.). PC was induced by four cycles of 5 min of ischemia and 5 min of reperfusion. Infarcts were produced by 60 min of ischemia and 3 h of reperfusion. Cardiac NE depletion was verified by radioimmunoassay of tissue samples and by absence of hemodynamic response to a tyramine bolus (1.4 mg/kg) administered at the end of each experiment. Infarct size, expressed as percent of area at risk, was controlled for variation in collateral blood flow using analysis of covariance (ANCOVA). RESULTS: Adjusted mean infarct size was 25.5 +/- 3.2% in untreated controls vs. 19.1 +/- 3.3% in RES-treated controls (P = NS). PC limited infarct size in untreated dogs (7.4 +/- 1.8 vs. 25.5 +/- 3.2%; PC vs. control; P < 0.01) but not in RES-treated dogs (15.7 +/- 3.0% vs. 19.1 +/- 3.3%; RES + PC vs. RES; P = NS). Infarct size was larger in dogs with RES + PC than with PC alone, even though there was a trend toward a slight beneficial effect with RES alone. CONCLUSION: The cardioprotective effect of ischemic preconditioning cannot be explained entirely as an anti-adrenergic effect. On the contrary, adrenergic receptor stimulation may be required for the full expression of ischemic preconditioning in canine myocardium.     

A phase II trial of interleukin-2 and interferon-alpha in the treatment of metastatic colorectal carcinoma

A total of 29 patients with stage IV colorectal cancer were entered into a phase II trial of bolus interleukin-2 (IL-2) and interferon-alpha (IFN alpha) (3 x 10(6) U/m2 of each cytokine given i.v. q8h x 15 doses and repeated in 2 weeks). Immunologic parameters measured on isolated peripheral blood lymphocytes revealed increased activated T cells with upregulated natural killer and lymphokine-activated killer activity. Among 24 evaluable patients, there were 4 partial responses (17%) of short duration ( < or = 6 months). Three of the responding patients had been refractory to prior chemotherapy. Overall median survival in the 24 evaluable patients was 18.5 months. Therapy necessitated an inpatient setting, with the most common toxicities being hypotension, hepatic insufficiency, fever, hypocalcemia, nausea/vomiting, and renal insufficiency. There were two treatment-related deaths. Because neither IL-2 nor IFN alpha alone has significant activity against colorectal cancer, the responses observed in this study suggest a potential synergistic effect between the two cytokines. However, the toxicity and short duration of response without survival benefit do not support the routine use of this regimen as a therapeutic modality for this tumor histology.