4 October 2022 in IEEE Transactions on Computers
Rowhammer Attack, a new DRAM-based attack, was developed exploiting weak cells to alter their content. Such attacks can be launched at the user level without requiring access permission to the victim memory cells. Leveraging such attacks, a new bit-flip-based adversarial weights attack (BFA) was developed targeting deep neural network models. When BFA attackers acquire a DNN model, they manipulate the existing DNN adversarial attack into locating vulnerable bits in the target DNN model. By flipping a subset of them using Rowhammer, they can crash that model within 30 trails. In this paper, we propose a lightweight and easy-to-deploy defense mechanism in the bit-level, Randomized Rotated and Nonlinear Encoding (RREC), which generates both robustness and fault-tolerant against BFA. Since flipping the most significant bit (MSB) in quantized data is too dangerous, we introduce randomized Rotation to obfuscate the bit order of model data and efficiently hide truly vulnerable bits with less vulnerable ones. Further, RREC reduces the average bit-flipped distance by more than 3x from the nonlinear encoding. It decreases the bit-flip distance among the majority of bits (including those vulnerable bits). Theoretically, RREC minimized the impact of a single bit BFA to 1/24 compared with baseline. Experimentally, RREC tolerates more than 17x flipped bits versus baseline model and 4.8x and 5.7x more bits compared with the existing BFA defenses (4B QAT and WR) with 0.01x to 0.08x of runtime latency. Moreover, we evaluate RREC against a newly emerged attack, Targeted-BFA, and it improves the defense rate from $5\%$ to $95\%$ .
3 September 2021 in IEEE Transactions on Instrumentation and Measurement
Remote photoplethysmography (rPPG) enables noncontact heart rate (HR) estimation using facial videos and exhibits significant convenience over traditional contact-based HR measurement approaches. However, its effectiveness depends mainly on the number of pixels of the facial region; thus, the estimation accuracy is easily degraded due to missing rPPG information when the resolution of facial images is low. Given that, in most real application scenarios, such as group-oriented and long-distance vital signs measurement, only low-resolution facial videos are available, finding a solution to recover rPPG information is a key to improve the performance of noncontact HR measurement. In this article, we propose a two-stage deep learning scheme to achieve accurate HR measurement from low-resolution facial videos. In the first stage, an rPPG information recovery network is proposed to recover the rPPG information during video super-resolution (SR) processing. The up-scaled images are then fed into a temporally aware HR measurement network in the second stage. An attention mechanism that reassigns the weights of the temporal information in each feature channel is designed to improve the measurement accuracy. The mean absolute percentage error achieves 3.320 and 4.231, while Pearson's correlation coefficient reaches 0.930 and 0.892 on two datasets, respectively. Experimental results show that rPPG information could be effectively recovered through SR processing. Compared with the state-of-the-art approaches that treat high-resolution images as the input, the proposed method achieved closely comparable measurement performance using merely low-resolution images.
13 June 2019 in IEEE Journal of Biomedical and Health Informatics
Cervical cancer ranks as the second most common cancer in women worldwide. In clinical practice, colposcopy is an indispensable part of screening for cervical intraepithelial neoplasia (CIN) grades and cervical cancer but exhibits high misdiagnosis rate. Existing computer-assisted algorithms for analyzing cervigram images have neglected that colposcopy is a sequential and multistate process, which is unsuitable for clinical applications. In this work, we construct a cervigram-based recurrent convolutional neural network (C-RCNN) to classify different CIN grades and cervical cancer. Convolutional neural networks (CNN) are leveraged to extract spatial features. We develop a sequence-encoding module to encode discriminative temporal features and a multistate-aware convolutional layer to integrate features from different states of cervigram images. To train and evaluate the performance of C-RCNN, we leveraged a dataset of 4,753 real cervigrams and obtained 96.13% test accuracy with a specificity and sensitivity of 98.22% and 95.09%, respectively. Areas under each receiver operating characteristic curves (AUC) are above 0.94, proving that visual representations and sequential dynamics can be jointly and effectively optimized in the training phase. Comparative analysis demonstrated the effectiveness of the proposed C-RCNN against competing methods, showing significant improvement over only focusing on a single frame. This architecture can be extended to other applications in medical image analysis.
8 March 2021 in IEEE Journal of Biomedical and Health Informatics
Automatic acetowhite lesion segmentation in colposcopy images (cervigrams) is essential in assisting gynecologists for the diagnosis of cervical intraepithelial neoplasia grades and cervical cancer. It can also help gynecologists determine the correct lesion areas for further pathological examination. Existing computer-aided diagnosis algorithms show poor segmentation performance because of specular reflections, insufficient training data and the inability to focus on semantically meaningful lesion parts. In this paper, a novel computer-aided diagnosis algorithm is proposed to segment acetowhite lesions in cervigrams automatically. To reduce the interference of specularities on segmentation performance, a specular reflection removal mechanism is presented to detect and inpaint these areas with precision. Moreover, we design a cervigram image classification network to classify pathology results and generate lesion attention maps, which are subsequently leveraged to guide a more accurate lesion segmentation task by the proposed lesion-aware convolutional neural network. We conducted comprehensive experiments to evaluate the proposed approaches on 3045 clinical cervigrams. Our results show that our method outperforms state-of-the-art approaches and achieves better Dice similarity coefficient and Hausdorff Distance values in acetowhite legion segmentation.
13 December 2017 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Three-dimensional (3D) NAND flash is one of the most prospective advances in flash memory industry. While 3D flash improves cell density and reduces lithography cost through die stacking, it suffers from severe program disturbance, which leads to significant performance and lifetime degradation for 3D flash based SSDs. To address the above challenge, we propose ApproxFTL, an approximate-write aware flash translation layer design, that uses approximate-write operations to store error-resilient data of modern applications. By reducing the maximal threshold voltage and tightening the guard bands between multi-level cell states, approximate write operations not only finish early but also exhibit large disturbance reduction, which can be exploited to alleviate disturbance in physical blocks that save both precise and approximate data. ApproxFTL maximizes the disturbance mitigation through approximate-write aware data placement, wear leveling, and garbage collection enhancements. Our experimental results show that ApproxFTL, while preserving high data quality, improves the read and write response time of flash accesses by 41.38% and 45.64% on average, respectively, and extends the lifetime of 3D flash based SSDs by 5.75% when comparing to the state-of-the-art.
30 April 2019 in IEEE Transactions on Knowledge and Data Engineering
With the fast advances in Internet technologies, social networks have become a major platform for social interaction, lifestyle demonstration, and message dissemination. Effective community detection in social networks helps to assess public sentiment, identify community leaders, and produce personalized recommendations. While different community detection approaches have been proposed in the literature, the trust model based detection schemes model user interactions as trust transfer, which helps to capture the implicit relation in the network. Unfortunately, trust model based detection schemes face cold start problem, i.e., they cannot accurately model newly joined users as these users have few interactions for a duration after joining the network. In this paper, we propose TLCDA, a novel trust model based community detection algorithm. By enhancing the traditional trust computation with inter-node relation strength and similarity in social networks, TLCDA detects communities through coarse-grained K-Mediods clustering. Our evaluation on real social networks shows that the communities detected by TLCDA exhibit superior preference cohesion while satisfying the topology cohesion.