The paper describes the research made toward im-proving medical case retrieval for Alzheimer’s Disease (AD). Our approach considers using Magnetic Resonance Images (MRI) as an input for the search. To improve the retrieval process we used longitudinal information extracted from different sets of scans ac-quired at different time points and automatically extracted de-scriptors to represent input images. All experiments were per-formed with and without quality control (QC) to determine the in-fluence of the errors caused by the automated processing to the results relevance. For the experiments, a total of 267 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database with available scans at baseline, the 6-month, 12-month, and 24-month follow-ups were selected. The obtained results showed that the selection of the time points for extraction of the longitudinal information influences the retrieval performance. Results also showed that not all automatically generated descriptors lead to im-provement of the results. Longitudinal volume changes provide the most relevant representation. Adding QC phase in the experi-ments leads to improvements in all examined scenarios. The re-sults showed that the most frequently automatically selected fea-tures are common semantic markers for AD.

A bite from a bug may expose the affected person to serious, life-threatening conditions, which may require immediate medical attention. The identification of the bug bite may be challenging even for experienced medical personnel due to the different manifestations of the bites and similarity to other skin conditions. This motivated our work on a computer-aided system that offers information on the bug bite based on the classification of bug bite images. Recently, there have been significant advances of methods for image classification for the detection of various skin conditions. However, there are very few sources that discuss the classification of bug bites. The goal of our research is to fill in this gap in the literature and offer a comprehensive approach for the analysis of this topic. This includes (1) the creation of a dataset that is larger than those considered in the related sources; (2) the exploration and analysis of the application of pre-trained state-of-the-art deep learning architectures with transfer learning, used in this study to overcome the challenges of low-size datasets and computational burden; (3) the further improvement of the classification performance of the individual CNNs by proposing an ensemble of models, and finally, (4) the implementation and description of an end-to-end system for bug bite classification from images taken with mobile phones, which should be beneficial to the medical personnel in the diagnostic process. In this paper, we give a detailed discussion of the models’ architecture, back-end architecture, and performance. According to the general evaluation metrics, DenseNet169 with an accuracy of 78% outperformed the other individual CNN models. However, the overall best performance (accuracy of 86%) was achieved by the proposed stacking ensemble model. These results are better than the results in the limited related work. Additionally, they show that deep CNNs and transfer learning can be successfully applied to the problem of the classification of bug bites.

This paper presents medical image retrieval for Alzheimer’s disease based on information extracted from multiple time points. The aim is to analyze the retrieval performance directed by the information combination of different time points and from each time point separately. For each subject, Magnetic Resonance Images (MRI) of four consecutive time points are obtained from ADNI database. Measurements of cortical and subcortical brain structures, including volumes and cortical thickness of the brain regions are used as features. The feature selection is performed aiming to select the most relevant features and reduce redundant and possibly noisy data. The evaluation is based on ten scenarios defined for separate time points and for a combination of multiple time points. According to the obtained results, it can be concluded that the retrieval performance gets better while using the time point that is more temporally distant from the baseline, regarding the scenarios in which separate information at any time point is used. When using a combination of multiple time points, the retrieval performance is improved only in the case when more than two time points are available and the selected features from the last two points are used. Feature selection algorithm leads to better results in all cases, while significantly reduces the feature vector dimensionality. The selected features are known as significant markers for Alzheimer’s Disease.