Today, medical image analysis papers require solid experiments to prove the usefulness of proposed methods. However, experiments are often performed on data selected by the researchers, which may come from different institutions, scanners, and populations. Different evaluation measures may be used, making it difficult to compare the methods. In this paper, we introduce a dataset of 7909 breast cancer histopathology images acquired on 82 patients, which is now publicly available from http://web.inf.ufpr.br/vri/breast-cancer-database. The dataset includes both benign and malignant images. The task associated with this dataset is the automated classification of these images in two classes, which would be a valuable computer-aided diagnosis tool for the clinician. In order to assess the difficulty of this task, we show some preliminary results obtained with state-of-the-art image classification systems. The accuracy ranges from 80% to 85%, showing room for improvement is left. By providing this dataset and a standardized evaluation protocol to the scientific community, we hope to gather researchers in both the medical and the machine learning field to advance toward this clinical application.

The performance of most conventional classification systems relies on appropriate data representation and much of the efforts are dedicated to feature engineering, a difficult and time-consuming process that uses prior expert domain knowledge of the data to create useful features. On the other hand, deep learning can extract and organize the discriminative information from the data, not requiring the design of feature extractors by a domain expert. Convolutional Neural Networks (CNNs) are a particular type of deep, feedforward network that have gained attention from research community and industry, achieving empirical successes in tasks such as speech recognition, signal processing, object recognition, natural language processing and transfer learning. In this paper, we conduct some preliminary experiments using the deep learning approach to classify breast cancer histopathological images from BreaKHis, a publicly dataset available at http://web.inf.ufpr.br/vri/breast-cancer-database. We propose a method based on the extraction of image patches for training the CNN and the combination of these patches for final classification. This method aims to allow using the high-resolution histopathological images from BreaKHis as input to existing CNN, avoiding adaptations of the model that can lead to a more complex and computationally costly architecture. The CNN performance is better when compared to previously reported results obtained by other machine learning models trained with hand-crafted textural descriptors. Finally, we also investigate the combination of different CNNs using simple fusion rules, achieving some improvement in recognition rates.

Breast cancer (BC) is a deadly disease, killing millions of people every year. Developing automated malignant BC detection system applied on patient's imagery can help dealing with this problem more efficiently, making diagnosis more scalable and less prone to errors. Not less importantly, such kind of research can be extended to other types of cancer, making even more impact to help saving lives. Recent results on BC recognition show that Convolution Neural Networks (CNN) can achieve higher recognition rates than hand-crafted feature descriptors, but the price to pay is an increase in complexity to develop the system, requiring longer training time and specific expertise to fine-tune the architecture of the CNN. DeCAF (or deep) features consist of an in-between solution it is based on reusing a previously trained CNN only as feature vectors, which is then used as input for a classifier trained only for the new classification task. In the light of this, we present an evaluation of DeCaf features for BC recognition, in order to better understand how they compare to the other approaches. The experimental evaluation shows that these features can be a viable alternative to fast development of high-accuracy BC recognition systems, generally achieving better results than traditional hand-crafted textural descriptors and outperforming task-specific CNNs in some cases.

In many countries, water quality monitoring is limited due to the high cost of logistics and professional equipment such as multiparametric probes. However, low-cost sensors integrated with the Internet of Things can enable real-time environmental monitoring networks, providing valuable water quality information to the public. To facilitate the widespread adoption of these sensors, it is crucial to identify which sensors can accurately measure key water quality parameters, their manufacturers, and their reliability in different environments. Although there is an increasing body of work utilizing low-cost water quality sensors, many questions remain unanswered. To address this issue, a systematic literature review was conducted to determine which low-cost sensors are being used for remote water quality monitoring. The results show that there are three primary vendors for the sensors used in the selected papers. Most sensors range in price from US$6.9 to US$169.00 but can cost up to US$500.00. While many papers suggest that low-cost sensors are suitable for water quality monitoring, few compare low-cost sensors to reference devices. Therefore, further research is necessary to determine the reliability and accuracy of low-cost sensors compared to professional devices.

Recent public cooperation between the Federal University of Technology – Parana (UTFPR) and the Toledo Municipality plans to implement the concept of smart cities in this city. In this context, one of the applications under development intends to track the recyclable garbage collector trucks in real time over the Internet. Actually, fleet vehicle tracking is one of the main applications for smart cities. LoRaWAN stands out among network technologies for smart cities due to operating in an open frequency range, covering long distances with low power consumption and low equipment cost. However, the coverage and performance of LoRaWAN is directly affected by both the environment and configuration parameters. In addition, tracking devices must be able to send its coordinates to the Internet even when the vehicle goes through zones where there are obstacles for electromagnetic waves such as elevated buildings or valleys. In this paper we perform experimental investigations to evaluate four LoRaWAN tracking devices, two available out of the box and two assembled and programmed. The behavior of each tracking device is analyzed when moving at a constant speed through three representative urban areas totaling 10.71 km². The two most efficient tracking devices are analyzed in a stretch of 3.5 km with speeds ranging from 0 to 30 km/h, 0 to 50 km/h and 0 to 100 km/h. Results include a quantitative and qualitative aspects, including the received signal strength indication (RSSI), signal-to-noise ratio (SNR), packet delivery ratio (PDR), and spreading factor (SF) for the received geographic coordinates. As the devices depend on the quality of the signal offered by the network, we also present the results of the development and evaluation of the LoRaWAN network, by planning its coverage throughout the city.