The field of medical imaging diagnostic makes use of a modality of imaging tests, e.g., X-rays, ultrasounds, computed tomographies, and magnetic resonance imaging, to assist physicians with the diagnostic of patients’ illnesses. Due to their state-of-the-art results in many challenging image classification tasks, deep neural networks (DNNs) are suitable tools for use by physicians to provide diagnostic support when dealing with medical images. To further advance the field, the present work proposes a two-phase algorithm capable of automatically generating compact DNN architectures given a database, called here DNNDeepeningPruning. In the first phase, also called the deepening phase, the algorithm grows a DNN by adding blocks of residual layers one after another until the model overfits the given data. In the second phase, called the pruning phase, the algorithm prunes the created DNN model from the first phase to produce a DNN with a small amount of floating-point operations guided by some preference given by the user. The proposed algorithm unifies the two separate fields of DNN architecture searching and pruning under a single framework, and it is tested in two medical imaging data sets with satisfactory results.

The ubiquitous use of cameras in a home environment raises privacy concerns, which is one of the major barriers to the deployment of smart home systems for elderly and disabled care. Social robots are equipped with cameras that can witness embarrassing situations of their owners, such as nakedness during a morning bath. In this paper, our proposed solution is to let the robot detect such privacy-sensitive moments and then turn away to avoid observing the person. We proposed a method based on the Convolutional Neural Networks (CNN) to detect nakedness in a daily living scenario. In this method, two CNNs are used: one detects whether a person is present or not in the current scene, and the other detects privacy-sensitive situations. We implemented the method on the ASCC home service robot developed in our lab. The CNNs were trained with a database composed of more than 2900 pictures collected from the Internet and from our ASCC Smart Home. In our experiments, we considered three human poses: sitting, standing, and lying. If the robot detects a privacy-sensitive situation, it will turn around and verbally announce its intention, which ensures the person that he or she is not being watched. Our experiments validate the proposed privacy preserving method. We also conducted a user study and find that people prefer that the robot takes some actions to avoid directly watching them in privacy-sensitive situations.

Recent research has shown that the ubiquitous use of cameras and voice monitoring equipment in a home environment can raise privacy concerns and affect human mental health. This can be a major obstacle to the deployment of smart home systems for elderly or disabled care. This study uses a social robot to detect embarrassing situations. Firstly, we designed an improved neural network structure based on the You Only Look Once (YOLO) model to obtain feature information. By focusing on reducing area redundancy and computation time, we proposed a bounding-box merging algorithm based on region proposal networks (B-RPN), to merge the areas that have similar features and determine the borders of the bounding box. Thereafter, we designed a feature extraction algorithm based on our improved YOLO and B-RPN, called F-YOLO, for our training datasets, and then proposed a real-time object detection algorithm based on F-YOLO (RODA-FY). We implemented RODA-FY and compared models on our MAT social robot. Secondly, we considered six types of situations in smart homes, and developed training and validation datasets, containing 2580 and 360 images, respectively. Meanwhile, we designed three types of experiments with four types of test datasets composed of 960 sample images. Thirdly, we analyzed how a different number of training iterations affects our prediction estimation, and then we explored the relationship between recognition accuracy and learning rates. Our results show that our proposed privacy detection system can recognize designed situations in the smart home with an acceptable recognition accuracy of 94.48%. Finally, we compared the results among RODA-FY, Inception V3, and YOLO, which indicate that our proposed RODA-FY outperforms the other comparison models in recognition accuracy.