Artificial intelligence-assisted medical diagnosis is enhancing accuracy with the contribution of several state of the art technologies such as Deep Learning (DL), Machine Learning (ML) and Image Processing (IP). From the detection of diseases to the selection of proper treatment plans, AI-powered assistance is effectively employed by healthcare professionals. Despite these advancements, the application of AI in animal healthcare is lagging behind, presenting a significant scope for AI adoption in veterinary medical diagnostics. This study addresses this gap by focusing on the automated diagnosis of canine Babesia infection, a parasitic disease that affects red blood cells (RBC). Our research contributed by developing a labeled dataset of microscopic images of red blood cells of infected and uninfected cases. During this work, four AI models are developed for automated classification: a custom Convolutional Neural Network (CNN), two pre-trained models (VGG16 ,DenseNet121) and a hybrid model (DenseNet121 + Support Vector Machine (SVM)). The performance of these models was 96.88%, 94%, 96.37% and 95.50% respectively. To further enhance the accuracy, a weighted average ensemble technique was employed. The ensemble model achieved an improved accuracy of 97.75%, demonstrating its potential. The enhanced performance of the ensemble model highlights the effectiveness of our method, significantly outperforming traditional methods and providing veterinarians with an efficient early diagnosis tool. This study is one of the few to address disease detection from microscopic images in animals using the potential of Artificial Intelligence.

The progression of Artificial Intelligence (AI) is poised to revolutionize healthcare, augmenting human expertise with computational capabilities to transform clinical decision-making. This study delves into the intricate dynamics among societal values, computational algorithms, and medical practice, shedding light on the juncture where AI functions as a collaborative tool, elevating diagnostic precision and treatment selection while upholding the human dimension in patient care. The evolution of AI technology highlights its capacity to process extensive medical data, unveiling nuances imperceptible to human observation. Nevertheless, the assimilation of AI within clinical contexts prompts ethical considerations concerning the continuity of patient-centric care. The essence of patient-provider relationships—characterized by empathy and emotional resonance—intersects with AI's computational efficacy. This inquiry navigates this intersection by exploring models of collaboration that harmonize AI with human judgment, enriching clinical accuracy while honoring the complexities of human experience. In response to the dynamic healthcare landscape, various dimensions come into focus, spanning the augmentation of diagnostic precision to the intricate intricacies of treatment selection. These facets underscore the symbiotic rapport between AI algorithms and human expertise. Nonetheless, challenges emerge in the integration process—ensuring AI's transparency, preserving patient autonomy, and addressing potential erosion of trust. This work underscores the essential equilibrium between AI's capabilities and the essence of compassionate human care. By traversing diagnostic accuracy, treatment selection, and the preservation of human attributes, this study embarks on a holistic exploration. It navigates the metamorphosis of medical practice, steered by AI-augmented healthcare. This investigation illuminates the trajectory toward an ethically grounded, patient-centric integration of AI, heralding a transformative era in healthcare while honoring the timeless principles of empathetic care.

This research introduces a novel method that integrates both unsupervised and supervised learning, leveraging SimCLR (Simple Framework for Contrastive Learning of Visual Representations) for self-supervised learning along with different pre-trained models to improve microscopic image classification of Babesia parasite in canines. We focused on three popular CNN architectures, namely ResNet, EfficientNet, and DenseNet, and evaluated the impact of SimCLR pre-training on their performance. A detailed comparison of the different variants of ResNet, EfficientNet, and Densenet in terms of classification accuracy and training efficiency is presented. Base models such as different variants of the ResNet, EfficientNet, and DenseNet models were utilized within the SimCLR framework. Firstly, the models were pre-trained on unlabeled images, followed by training classifiers on labeled datasets. This approach significantly improved the robustness and accuracy, demonstrating the potential benefits of combining contrastive learning with conventional supervised techniques. The highest accuracy of 97.07% was achieved by Efficientnet_b2. Thus, detection of Babesia or other hemoparasites in microscopic blood smear images could be automated with high accuracy without using a labelled dataset.

Veterinary diagnostics is a field that demands AI-based solutions suitable for resource constraint environment. Traditional animal disease detection methods have significant scope for AI adoption to automate certain aspects of the diagnosis process. In this study, twelve CNN-ML hybrid models are developed, leveraging the power of Convolutional Neural Networks (CNN) combined with traditional machine learning approaches. Pre-trained models are employed for feature extraction paired with traditional ML classifiers such as Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boost (XGBoost) and AdaBoost. These twelve hybrid models were categorized based on their CNNbackbone: category I (DenseNet121+ML models), category II (VGG16+ML models) and category III(InceptionV3+ML models). The top-performing hybrid models from each category i.e., DenseNet121+RF (96.00% accuracy) from category I, VGG16+SVM (93.5% accuracy) from category II and InceptionV3+SVM (92.60% accuracy) from category III are selected for ensemble modeling. The weighted ensemble achieved an accuracy of 95.5%, while the voting ensemble is able to achieve the accuracy to 96.62%. The ensemble model is shown to be more generalized and robust compared to the individual hybrid models. Additionally, as training time was a crucial consideration in resource-constrained environment, we concluded that weighted ensemble model is efficient as its training time is only 0.97 seconds.

Disorders such as malaria, leukemia, thrombocytopenia, sickle cell anemia, and many others cause morphological and physiological changes in the blood cells which can be used as an indicator in the diagnosis process of such diseases. Blood smear analysis is a routine work carried out in laboratories to diagnose a disease. Microscopic evaluation is one of the widely used techniques for blood smear analysis. However, the diagnostic result may vary depending on the skill and experience of the technician, instruments and methods used to analyze the blood sample. Manual microscopic examination is time consuming and is less repeatability. The advent of digital imaging taken using Light microscopy has enhanced diagnostic role. In recent years several technological innovations in the field of image processing and Convolutional Neural Networks (CNN) are employed in automated medical diagnosis. This paper focuses on the recent works reported in the field of automated disease diagnosis from blood-smeared images.