Biomedical subjects

N Sundararajan

Publications and source records attributed to N Sundararajan.

3 recordsLinked to original sources

PubMed1999

Impact of the hydrodyne process on tenderness, microbial load, and sensory characteristics of pork longissimus muscle.

Paired, boneless pork loin muscles were obtained from 76 market hogs to evaluate tenderness, meat quality characteristics, sensory attributes, and microbial characterization of pork muscle exposed to the Hydrodyne Process (H) compared with untreated control (C) loin. A subset of 16 paired loins was randomly selected for use in sensory evaluation and microbial characterization. Loins were vacuum packaged and immersed in a heat shrink tank prior to the H treatment. The Hydrodyne treatment exposed the loin to the pressure equivalent of a 150-g explosive, generating a pressure distribution of approximately 703 kg/cm2 at the surface of the samples. Meat quality assessments taken following treatment included subjective color, firmness/wetness, marbling scores (1 to 5 scale), Minolta reflectance and color readings, drip loss, and lipid content. The P-value for statistical significance for main effects and interactions was set at <.05 in all analyses. Administration of H resulted in a 17% improvement in Warner-Bratzler shear force (2.69 vs. 3.24 kg), with the shear force similar at two end-point cooking times (11 and 16 min) corresponding to approximately 75 and 83 degrees C, respectively. No differences between H and C were observed for color score, firmness score, Minolta L, Minolta Y, or drip loss on uncooked samples. The H loins had lower marbling scores (P<.05) and intramuscular lipid (P<.05) content than the paired C loin. Sensory evaluation on the randomly selected (n = 16) paired loins samples showed no improvement in Warner-Bratzler shear force. Sensory panelists were also unable to detect a difference between H and C loins for both initial and sustained tenderness scores. No differences between H and C loins were found for pork flavor, off-flavor, cohesiveness, or number of chews before swallowing, but H loins had a significantly lower juiciness score and more cooking loss than C loins. Microbial analysis results showed no differences in coliform bacteria counts, aerobic plate counts, and no detectable levels of Escherichia coli bacteria in any loins. The findings support the ability of the Hydrodyne procedure to improve tenderness without impacting other muscle quality attributes of pork.

Animals↗

PubMed1997

A sequential learning scheme for function approximation using minimal radial basis function neural networks.

This article presents a sequential learning algorithm for function approximation and time-series prediction using a minimal radial basis function neural network (RBFNN). The algorithm combines the growth criterion of the resource-allocating network (RAN) of Platt (1991) with a pruning strategy based on the relative contribution of each hidden unit to the overall network output. The resulting network leads toward a minimal topology for the RBFNN. The performance of the algorithm is compared with RAN and the enhanced RAN algorithm of Kadirkamanathan and Niranjan (1993) for the following benchmark problems: (1) hearta from the benchmark problems database PROBEN1, (2) Hermite polynomial, and (3) Mackey-Glass chaotic time series. For these problems, the proposed algorithm is shown to realize RBFNNs with far fewer hidden neurons with better or same accuracy.

Algorithms↗

PubMed1995

Analysis of training set parallelism for backpropagation neural networks.

Training set parallelism and network based parallelism are two popular paradigms for parallelizing a feedforward (artificial) neural network. Training set parallelism is particularly suited to feedforward neural networks with backpropagation learning where the size of the training set is large in relation to the size of the network. This paper analyzes training set parallelism for feedforward neural networks when implemented on a transputer array configured in a pipelined ring topology. Theoretical expressions for the time per epoch (iteration) and optimal size of a processor network are derived when the training set is equally distributed among the processing nodes. These show that the speed up is a function of the number of patterns per processor, communication overhead per epoch and the total number of processors in the topology. Further analysis of how to optimally distribute the training set on a given processor network when the number of patterns in the training set is not an integer multiple of the number of processors, is also carried out. It is shown that optimal allocation of patterns in such cases is a mixed integer programming problem. Using this analysis it is found that equal distribution of training patterns among the processors is not the optimal way to allocate the patterns even when the training set is an integer multiple of the number of processors. Extension of the analysis to processor networks comprising processors of different speeds is also carried out. Experimental results from a T805 transputer array are presented to verify all the theoretical results.

Algorithms↗