In the era of digitized images, the goal is to extract information from them and create new knowledge thanks to Computer Vision techniques, Machine Learning and Deep Learning. This enables the use of images for early diagnosis and subsequent treatment of a wide range of diseases. In the dermatological field, deep neural networks are used to distinguish between melanoma and non-melanoma images. In this paper, we have underlined two essential points of melanoma detection research. The first aspect considered is how even a simple modification of the parameters in the dataset determines a change of the accuracy of classifiers. In this case, we investigated the Transfer Learning issues. Following the results of this first analysis, we suggest that continuous training-test iterations are needed to provide robust prediction models. The second point is the need to have a more flexible system architecture that can handle changes in the training datasets. In this context, we proposed the development and implementation of a hybrid architecture based on Cloud, Fog and Edge Computing to provide a Melanoma Detection service based on clinical and dermoscopic images. At the same time, this architecture must deal with the amount of data to be analyzed by reducing the running time of the continuous retrain. This fact has been highlighted with experiments carried out on a single machine and different distribution systems, showing how a distributed approach guarantees output achievement in a much more sufficient time.

Melanoma is considered the deadliest skin cancer and when it is in an advanced state it is difficult to treat. Diagnoses are visually performed by dermatologists, by naked-eye observation. This paper proposes an augmented reality smartphone application for supporting the dermatologist in the real-time analysis of a skin lesion. The app augments the camera view with information related to the lesion features generally measured by the dermatologist for formulating the diagnosis. The lesion is also classified by a deep learning approach for identifying melanoma. The real-time process adopted for generating the augmented content is described. The real-time performances are also evaluated and a user study is also conducted. Results revealed that the real-time process may be entirely executed on the Smartphone and that the support provided is well judged by the target users.

In this paper we present an approach to automate the architecture recovery process of software systems. The approach is built on information retrieval and clustering techniques, and, in particular, uses Latent Semantic Indexing (LSI) to get similarities among software entities (e.g., programs or classes) and the k-means clustering algorithm to form groups of software entities that implement similar functionality. In order to improve computational time in the context of the software evolution and then reduce energy waste, the architecture recovery process can be also applied by using fold-in and fold-out mechanisms that, respectively, add and remove software entities to the LSI representation of the understudy software system. The approach has been implemented in a prototype of a supporting software system as an Eclipse plug-in. Finally, to assess the approach and the plug-in, we have conducted an empirical investigation on five open source software systems implemented using the programming languages Java and C/C++. In the investigation special emphasis has been also given to the effect of using the fold-in and fold-out mechanisms.

Detecting emotions is very useful in many fields, from health-care to human-computer interaction. In this paper, we propose an iterative user-centered methodology for supporting the development of an emotion detection system based on low-cost sensors. Artificial Intelligence techniques have been adopted for emotion classification. Different kind of Machine Learning classifiers have been experimentally trained on the users’ biometrics data, such as hearth rate, movement and audio. The system has been developed in two iterations and, at the end of each of them, the performance of classifiers (MLP, CNN, LSTM, Bidirectional-LSTM and Decision Tree) has been compared. After the experiment, the SAM questionnaire is proposed to evaluate the user’s affective state when using the system. In the first experiment we gathered data from 47 participants, in the second one an improved version of the system has been trained and validated by 107 people. The emotional analysis conducted at the end of each iteration suggests that reducing the device invasiveness may affect the user perceptions and also improve the classification performance.

In this paper, we present case studies of recovering structural design patterns from object-oriented source code. The proposed recovery technique is based on the use of visual language parsing techniques, and is supported by a visual environment automatically produced by a grammar based visual environment generator. We have applied the recovery technique to public-domain programs and libraries obtaining encouraging results. In particular, for the considered software our recovery approach is characterized by higher recall and precision values with respect to other recovery techniques

In this paper we present a two phase approach to the recovery of structural design pattern. In the first phase, the design pattern instances are identified at a course-grained level by considering the design structure only and using a visual language parsing technique. Then, the identified candidate patterns are validated by a fine-grained source code analysis phase. The latter phase is an enhancement of a previous approach developed by the authors aiming at improving the results of precision and time performances. The retrieval effectiveness of the approach is assessed by applying the recovery technique on four software systems