Breast cancer diagnosis is one of the many areas that has taken advantage of artificial intelligence to achieve better performance, despite the fact that the availability of a large medical image dataset remains a challenge. Transfer learning (TL) is a phenomenon that enables deep learning algorithms to overcome the issue of shortage of training data in constructing an efficient model by transferring knowledge from a given source task to a target task. However, in most cases, ImageNet (natural images) pre-trained models that do not include medical images, are utilized for transfer learning to medical images. Considering the utilization of microscopic cancer cell line images that can be acquired in large amount, we argue that learning from both natural and medical datasets improves performance in ultrasound breast cancer image classification. The proposed multistage transfer learning (MSTL) algorithm was implemented using three pre-trained models: EfficientNetB2, InceptionV3, and ResNet50 with three optimizers: Adam, Adagrad, and stochastic gradient de-scent (SGD). Dataset sizes of 20,400 cancer cell images, 200 ultrasound images from Mendeley and 400 ultrasound images from the MT-Small-Dataset were used. ResNet50-Adagrad-based MSTL achieved a test accuracy of 99 ± 0.612% on the Mendeley dataset and 98.7 ± 1.1% on the MT-Small-Dataset, averaging over 5-fold cross validation. A p-value of 0.01191 was achieved when comparing MSTL against ImageNet based TL for the Mendeley dataset. The result is a significant improvement in the performance of artificial intelligence methods for ultrasound breast cancer classification compared to state-of-the-art methods and could remarkably improve the early diagnosis of breast cancer in young women.

Various convolutional neural network (CNN)-based approaches have been recently proposed to improve the performance of motor imagery based-brain-computer interfaces (BCIs). However, the classification accuracy of CNNs is compromised when target data are distorted. Specifically for motor imagery electroencephalogram (EEG), the measured signals, even from the same person, are not consistent and can be significantly distorted. To overcome these limitations, we propose to apply a capsule network (CapsNet) for learning various properties of EEG signals, thereby achieving better and more robust performance than previous CNN methods. The proposed CapsNet-based framework classifies the two-class motor imagery, namely right-hand and left-hand movements. The motor imagery EEG signals are first transformed into 2D images using the short-time Fourier transform (STFT) algorithm and then used for training and testing the capsule network. The performance of the proposed framework was evaluated on the BCI competition IV 2b dataset. The proposed framework outperformed state-of-the-art CNN-based methods and various conventional machine learning approaches. The experimental results demonstrate the feasibility of the proposed approach for classification of motor imagery EEG signals.

The Internet of Things (IoT) attempts to help people access internet-connected devices, applications, and services anytime and anywhere. However, how providing an efficient and intuitive method of interaction between people and IoT devices is still an open challenge. In this work, we propose a novel interaction system called Watch & Do, where users can control an IoT device by gazing at it and doing simple gestures. The proposed system mainly consists of 1) object detection module, 2) gaze estimation module, 3) hand gesture recognition module, and 4) IoT controller module. The target device is identified by various deep learning-based gaze estimation and object detection techniques. Afterwards, hand gesture recognition is applied to generate an IoT device control command which is transmitted to the IoT platform. The experimental results and case studies demonstrate the feasibility of the proposed system and imply the future research directions.

Facial emotion recognition (FER) systems are imperative in recent advanced artificial intelligence (AI) applications to realize better human–computer interactions. Most deep learning-based FER systems have issues with low accuracy and high resource requirements, especially when deployed on edge devices with limited computing resources and memory. To tackle these problems, a lightweight FER system, called Light-FER, is proposed in this paper, which is obtained from the Xception model through model compression. First, pruning is performed during the network training to remove the less important connections within the architecture of Xception. Second, the model is quantized to half-precision format, which could significantly reduce its memory consumption. Third, different deep learning compilers performing several advanced optimization techniques are benchmarked to further accelerate the inference speed of the FER system. Lastly, to experimentally demonstrate the objectives of the proposed system on edge devices, Light-FER is deployed on NVIDIA Jetson Nano.

The demand for wheelchairs has increased recently as the population of the elderly and patients with disorders increases. However, society still pays less attention to infrastructure that can threaten the wheelchair user, such as sidewalks with cracks/potholes. Although various studies have been proposed to recognize such challenges, they mainly depend on RGB images or IMU sensors, which are sensitive to outdoor conditions such as low illumination, bad weather, and unavoidable vibrations, resulting in unsatisfactory and unstable performance. In this paper, we introduce a novel system based on various convolutional neural networks (CNNs) to automatically classify the condition of sidewalks using images captured with depth and infrared modalities. Moreover, we compare the performance of training CNNs from scratch and the transfer learning approach, where the weights learned from the natural image domain (e.g., ImageNet) are fine-tuned to the depth and infrared image domain. In particular, we propose applying the ResNet-152 model pre-trained with self-supervised learning during transfer learning to leverage better image representations. Performance evaluation on the classification of the sidewalk condition was conducted with 100% and 10% of training data. The experimental results validate the effectiveness and feasibility of the proposed approach and bring future research directions.

In smart interactive environments, such as digital museums or digital exhibition halls, it is important to accurately understand the user’s intent to ensure successful and natural interaction with the exhibition. In the context of predicting user intent, gaze estimation technology has been considered one of the most effective indicators among recently developed interaction techniques (e.g., face orientation estimation, body tracking, and gesture recognition). Previous gaze estimation techniques, however, are known to be effective only in a controlled lab environment under normal lighting conditions. In this study, we propose a novel deep learning-based approach to achieve a successful gaze estimation under various low-light conditions, which is anticipated to be more practical for smart interaction scenarios. The proposed approach utilizes a generative adversarial network (GAN) to enhance users’ eye images captured under low-light conditions, thereby restoring missing information for gaze estimation. Afterward, the GAN-recovered images are fed into the convolutional neural network architecture as input data to estimate the direction of the user gaze. Our experimental results on the modified MPIIGaze dataset demonstrate that the proposed approach achieves an average performance improvement of 4.53%–8.9% under low and dark light conditions, which is a promising step toward further research.