Model-data-driven adversarial active learning for brain tumor segmentation.

Active learning (AL) attempts to select informative samples in a dataset to minimize the number of required labels while maximizing the performance of the model. Current AL in segmentation tasks is limited to the expansion of popular classification-based methods including entropy, MC-dropout, etc. Meanwhile, most applications in the medical field are simply migrations that fail to consider the nature of medical images, such as high class imbalance, high domain difference, and data scarcity. In this study, we address these challenges and propose a novel AL framework for medical image segmentation task. Our approach introduces a pseudo-label-based filter addressing excessive blank patches in medical abnormalities segmentation tasks, e.g., lesions, and tumors, used before the AL selection. This filter helps reduce resource usage and allows the model to focus on selecting more informative samples. For the sample selection, we propose a novel query strategy that combines both model impact and data stability by employing adversarial attack. Furthermore, we harness the adversarial samples generated during the query process to enhance the robustness of the model. The experimental results verify our framework's effectiveness over various state-of-the-art methods. Our proposed method only needs less than 14% annotated patches in 3D brain MRI multiple sclerosis (MS) segmentation tasks and 20% for Low-Grade Glioma (LGG) tumor segmentation to achieve competitive results with full supervision. These promising outcomes not only improve performance but alleviate the time burden associated with expert annotation, thereby facilitating further advancements in the field of medical image segmentation. Our code is available at https://github.com/HelenMa9998/adversarial_active_learning.

An Information Gain-Based Model and an Attention-Based RNN for Wearable Human Activity Recognition.

Wearable sensor-based HAR (human activity recognition) is a popular human activity perception method. However, due to the lack of a unified human activity model, the number and positions of sensors in the existing wearable HAR systems are not the same, which affects the promotion and application. In this paper, an information gain-based human activity model is established, and an attention-based recurrent neural network (namely Attention-RNN) for human activity recognition is designed. Besides, the attention-RNN, which combines bidirectional long short-term memory (BiLSTM) with attention mechanism, was tested on the UCI opportunity challenge dataset. Experiments prove that the proposed human activity model provides guidance for the deployment location of sensors and provides a basis for the selection of the number of sensors, which can reduce the number of sensors used to achieve the same classification effect. In addition, experiments show that the proposed Attention-RNN achieves F1 scores of 0.898 and 0.911 in the ML (Modes of Locomotion) task and GR (Gesture Recognition) task, respectively.

A gigabyte interpreted seismic dataset for automatic fault recognition.

The lack of large-scale open-source expert-labelled seismic datasets is one of the barriers to applying today's AI techniques to automatic fault recognition tasks. The dataset present in this article consists of a large number of processed seismic images and their corresponding fault annotations. The processed seismic images, which are originally from a seismic survey called Thebe Gas Field in the Exmouth Plateau of the Carnarvan Basin on the NW shelf of Australia, are represented in Python Numpy format, which can be easily adopted by various AI models and will facilitate cooperation with researchers in the field of computer science. The corresponding fault annotations were firstly manually labelled by expert interpreters of faults from seismic data in order to investigate the structural style and associated evolution of the basin. Then the fault interpretation and seismic survey are processed and collected using Petrel software and Python programs separately. This dataset can help to train, validate, and evaluate the performance of different automatic fault recognition workflow.

An Energy-Efficient Fall Detection Method Based on FD-DNN for Elderly People.

A fall detection module is an important component of community-based care for the elderly to reduce their health risk. It requires the accuracy of detections as well as maintains energy saving. In order to meet the above requirements, a sensing module-integrated energy-efficient sensor was developed which can sense and cache the data of human activity in sleep mode, and an interrupt-driven algorithm is proposed to transmit the data to a server integrated with ZigBee. Secondly, a deep neural network for fall detection (FD-DNN) running on the server is carefully designed to detect falls accurately. FD-DNN, which combines the convolutional neural networks (CNN) with long short-term memory (LSTM) algorithms, was tested on both with online and offline datasets. The experimental result shows that it takes advantage of CNN and LSTM, and achieved 99.17% fall detection accuracy, while its specificity and sensitivity are 99.94% and 94.09%, respectively. Meanwhile, it has the characteristics of low power consumption.