Attention module incorporated transfer learning empowered deep learning-based models for classification of phenotypically similar tropical cattle breeds (Bos indicus).

Accurate breed identification in dairy cattle is essential for optimizing herd management and improving genetic standards. A smart method for correctly identifying phenotypically similar breeds can empower farmers to enhance herd productivity. A convolutional neural network (CNN) based model was developed for the identification of Sahiwal and Red Sindhi cows. To increase the classification accuracy, first, cows's pixels were segmented from the background using CNN model. Using this segmented image, a masked image was produced by retaining cows' pixels from the original image while eliminating the background. To improve the classification accuracy, models were trained on four different images of each cow: front view, side view, grayscale front view, and grayscale side view. The masked images of these views were fed to the multi-input CNN model which predicts the class of input images. The segmentation model achieved intersection-over-union (IoU) and F1-score values of 81.75% and 85.26%, respectively with an inference time of 296 ms. For the classification task, multiple variants of MobileNet and EfficientNet models were used as the backbone along with pre-trained weights. The MobileNet model achieved 80.0% accuracy for both breeds, while MobileNetV2 and MobileNetV3 reached 82.0% accuracy. CNN models with EfficientNet as backbones outperformed MobileNet models, with accuracy ranging from 84.0% to 86.0%. The F1-scores for these models were found to be above 83.0%, indicating effective breed classification with fewer false positives and negatives. Thus, the present study demonstrates that deep learning models can be used effectively to identify phenotypically similar-looking cattle breeds. To accurately identify zebu breeds, this study will reduce the dependence of farmers on experts.

Evaluation of salivary electrolytes during estrous cycle in Murrah buffaloes with reference to estrus detection.

AIM: Timely estrus detection is one of the critical factors for increasing reproductive efficiency in buffaloes. In recent decades, saliva has become a more popular as a noninvasive source for determining physiological status of animals by various biochemical electrolytes. This study was designed to assess and correlate changes in different salivary minerals concentration (calcium, inorganic phosphorus, magnesium, sodium, potassium, and chloride) during different stages of the estrous cycle in Murrah buffaloes. MATERIALS AND METHODS: The saliva samples were collected during the different phases of the estrous cycle from 20 Murrah buffaloes in early morning hours and were assayed using respective minerals assay kits. RESULTS: The concentrations of calcium (8.76±0.08-12.11±0.11 mg/dl), inorganic phosphorus (6.56±0.13-14.72±4.50 mg/dl), magnesium (2.27±0.14-5.79±0.15 mg/dl), sodium (139.47±0.31-159.62±1.22 mmol/L), potassium (12.40±0.22-26.85±1.22 mmol/L), and chloride (109.28±0.41-137.07±0.68 mmol/L) varied during the different phases of estrous cycle. The concentration of calcium, inorganic phosphorus, magnesium, sodium, potassium, and chloride in saliva were significantly (p<0.01) higher during estrus phase compared to other phases of the estrous cycle. All these minerals were positively and significantly (p<0.0001) related to estrogen concentration while salivary concentrations of calcium, magnesium, sodium, and chloride showed a significant (p<0.0001) negative correlation with progesterone level in blood plasma. CONCLUSION: These preliminary findings indicate that there are definite variations in salivary mineral and electrolyte concentrations during different phases of the estrous cycle. These results may be used as an aid for estrus detection/confirmation in buffaloes although validation of the results using a large number of animals is required.