The goal of this study is to propose a computer-aided diagnosis system to differentiate between four breast imaging reporting and data system (Bi-RADS) classes in digitised mammograms. This system is inspired by the approach of the doctor during the radiologic examination as it was agreed in BI-RADS, where masses are described by their form, their boundary and their density. The segmentation of masses in the authors' approach is manual because it is supposed that the detection is already made. When the segmented region is available, the features extraction process can be carried out. 22 visual characteristics are automatically computed from shape, edge and textural properties; only one human feature is used in this study, which is the patient's age. Classification is finally done using a multi-layer perceptron according to two separate schemes; the first one consists of classify masses to distinguish between the four BI-RADS classes (2, 3, 4 and 5). In the second one the authors classify abnormalities on two classes (benign and malign). The proposed approach has been evaluated on 480 mammographic masses extracted from the digital database for screening mammography, and the obtained results are encouraging. This paper is organised as the following: Section 2 presents the practical background related to the BI-RADS assessment; in Section 3, we review related researches; in Section 4, we explain the features extraction methods investigated and classification schemes used in the present work; in Section 5, we show the

The main aim of the work is to assess physical parameters of forest woodchips and their impact on the prices achieved by the supplier in transactions with a power plant. During fragmentation of logging residue, high content of green matter and contaminants negatively impacts the quality parameters that serve as basis for settlements. The analysis concerns data on the main parameters -water content, fuel value, sulphur and ash content -from 252 days of deliveries of forest chips to a power plant. The deliveries were realised from forested areas on an average about 340 km from the plant. Average water content and the resultant fuel value of forest chips was within 27-47% and 8.7-12.9 GJ×Mg −1 (appropriately), respectively. They depend on the month in which they are delivered to the power plant. The threshold values for the above-mentioned parameters are set by the plant at a real level and the suppliers have no problems with meeting them. The parameter that is most frequently exceeded is ash content (11.5% of cases). The settlement system does not differentiate on the basis of the transport distance but gives possibility to lower the settlement price when the quality parameters are not met but provides no reward for deliveries with parameters better than the average ones. On the basis of results obtained, it was calculated that average annual settlement price is lower than the contract price by about 0.20 PLN×GJ −1 , which in case of the analysed company may translate into an average daily loss of about 700 PLN.

No abstract

Several artificial intelligence approaches, particularly case-based reasoning (CBR), which is analogous to the context of human reasoning for problem resolution, have demonstrated their efficiency and reliability in the medical field. In recent years, deep learning represents the latest iteration of an advance in artificial intelligence technologies in medicine to aid in data classification, diagnosis of new diseases, and complex decision-making. Although these two independent approaches have good results in the medical field, the latter is still a complex field. This chapter reviews the available literature on CBR systems, deep learning systems, and CBR deep learning systems in medicine. The methods used and results obtained are discussed, and key findings are highlighted. Further, in the light of this review, some directions for future research are given. This chapter presents the proposed approach, which helps to make the retrieval phase of the CBR cycle more reliable and robust.