This study explores the use of deep learning neural networks and transfer learning to estimate the age of sheep from their dental images. This is an important aspect of agriculture for meat quality, animal welfare, breeding, and health management. Using cutting-edge techniques, MobileNet, ResNet50, and ResNet102, we compare two deep learning approaches: fine-tuning and feature extraction using the pre-trained version of these models as part of our investigation. We collected 540 images of sheep from nearby farms, concentrating on three age groups: young, middle-aged, and elderly, for the purpose of our study. With an interesting recognition accuracy of 96.9%, the experimental results demonstrate that ResNet102 is the best performer both when fine-tuned and when employing its deep features that are retrieved from its pre-trained version. These findings highlight how cutting-edge machine learning techniques have the potential to completely transform long-standing methods in the sheep sector and pave the way for developing a novel mobile application that improves economic outcomes and cultural conformity concerning sheep age recognition.

Using SEER data, this work seeks to create a machine learning (ML) model to predict lung metastases (LM) and three-month prognostic variables in hepatocellular carcinoma (HCC) patients. The study comprised 34,861 HCC patients, 1,783 (5.11 %) of whom had lung metastases, and 859 were suitable for the 3-month prognostic model. The ML models were cross-validated twice, and with an AUC of 1 and an F1 score of 0.997, the random forest (RF) classifier was found to be the best choice for predicting LM. The Easy Ensemble (EE) classifier was utilized to address the dataset's class imbalance problem. The study also indicated that employing resampling approaches such as SMOTE can result in synthetic data, which can reduce model reliability, making EE the recommended choice for addressing the class imbalance. Overall, this study aims to improve clinical decision-making by providing a comprehensive predictive model for HCC patients with LM and a 3-month prognosis.

The usefulness of the oversampling approach to class-imbalanced structured medical datasets is discussed in this paper. In this regard, we basically look into the oversampling approach's prevailing assumption that synthesized instances do belong to the minority class. We used an off-the-shelf oversampling validation system to test this assumption. According to the experimental results from the validation system, at least one of the three medical datasets used had newly generated samples that were not belonging to the minority class as a result of the oversampling methods validated. Additionally, the error rate varied based on the dataset and oversampling method tested. Therefore, we claim that synthesizing new instances without first confirming that they are aligned with the minority class is a risky approach, especially in medical fields where misdiagnosis can have serious repercussions. As alternatives to oversampling, ensemble, data partitioning, and method-level approaches are advised since they do not make false assumptions.

Classifying and modeling texture images, especially those with significant rotation, illumination, scale, and view-point variations, is a hot topic in the computer vision field. Inspired by local graph structure (LGS), local ternary patterns (LTP), and their variants, this paper proposes a novel image feature descriptor for texture and material classification, which we call Petersen Graph Multi-Orientation based Multi-Scale Ternary Pattern (PGMO-MSTP). PGMO-MSTP is a histogram representation that efficiently encodes the joint information within an image across feature and scale spaces, exploiting the concepts of both LTP-like and LGS-like descriptors, in order to overcome the shortcomings of these approaches. We first designed two single-scale horizontal and vertical Petersen Graph-based Ternary Pattern descriptors ($PGTP_{h}$and$PGTP_{v}$). The essence of$PGTP_{h}$and$PGTP_{v}$is to encode each$5\times 5$image patch, extending the ideas of the LTP and LGS concepts, according to relationships between pixels sampled in a variety of spatial arrangements (i.e., up, down, left, and right) of Petersen graph-shaped oriented sampling structures. The histograms obtained from the single-scale descriptors$PGTP_{h}$and$PGTP_{v}$are then combined, in order to build the effective multi-scale PGMO-MSTP model. Extensive experiments are conducted on sixteen challenging texture data sets, demonstrating that PGMO-MSTP can outperform state-of-the-art handcrafted texture descriptors and deep learning-based feature extraction approaches. Moreover, a statistical comparison based on the Wilcoxon signed rank test demonstrates that PGMO-MSTP performed the best over all tested data sets.

Research on content-based image retrieval (CBIR) has been under development for decades, and numerous methods have been competing to extract the most discriminative features for improved representation of the image content. Recently, deep learning methods have gained attention in computer vision, i ncluding CBIR. In this paper, we present a comparative investigation of different features, including low-level and high-level features, for CBIR. We compare the performance of CBIR systems using different deep features with state-of-the-art low-level features such as SIFT, SURF, HOG, LBP, and LTP, using different dictionaries and coefficient learning techniques. Furthermore, we conduct comparisons with a set of primitive and popular features that have been used in this field, including colour histograms and Gabor features. We also investigate the discriminative power of deep features using certain similarity measures under different validation approaches. Furthermore, we investigate the effects of the dimensionality reduction of deep features on the performance of CBIR systems using principal component analysis, discrete wavelet transform, and discrete cosine transform. Unprecedentedly, the experimental results demonstrate high (95% and 93%) mean average precisions when using the VGG-16 FC7 deep features of Corel-1000 and Coil-20 datasets with 10-D and 20-D K-SVD, respectively.