Sahand Hajifar

Bio: Sahand Hajifar is an academic researcher from University at Buffalo. The author has contributed to research in topics: Medicine & Wearable computer. The author has an hindex of 1, co-authored 6 publications receiving 12 citations. Previous affiliations of Sahand Hajifar include Sharif University of Technology.

Online Domain Adaptation for Continuous Cross-Subject Liver Viability Evaluation Based on Irregular Thermal Data

Abstract: Accurate evaluation of liver viability during its procurement is a challenging issue and has traditionally been addressed by taking invasive biopsy on liver. Recently, people have started to investigate on the non-invasive evaluation of liver viability during its procurement using the liver surface thermal images. However, existing works include the background noise in the thermal images and do not consider the cross-subject heterogeneity of livers, thus the viability evaluation accuracy can be affected. In this paper, we propose to use the irregular thermal data of the pure liver region, and the cross-subject liver evaluation information (i.e., the available viability label information in cross-subject livers), for the real-time evaluation of a new liver's viability. To achieve this objective, we extract features of irregular thermal data based on tools from graph signal processing (GSP), and propose an online domain adaptation (DA) and classification framework using the GSP features of cross-subject livers. A multiconvex block coordinate descent based algorithm is designed to jointly learn the domain-invariant features during online DA and learn the classifier. Our proposed framework is applied to the liver procurement data, and classifies the liver viability accurately.

Investigation of Heterogeneity Sources for Occupational Task Recognition via Transfer Learning.

Abstract: Human activity recognition has been extensively used for the classification of occupational tasks. Existing activity recognition approaches perform well when training and testing data follow an identical distribution. However, in the real world, this condition may be violated due to existing heterogeneities among training and testing data, which results in degradation of classification performance. This study aims to investigate the impact of four heterogeneity sources, cross-sensor, cross-subject, joint cross-sensor and cross-subject, and cross-scenario heterogeneities, on classification performance. To that end, two experiments called separate task scenario and mixed task scenario were conducted to simulate tasks of electrical line workers under various heterogeneity sources. Furthermore, a support vector machine classifier equipped with domain adaptation was used to classify the tasks and benchmarked against a standard support vector machine baseline. Our results demonstrated that the support vector machine equipped with domain adaptation outperformed the baseline for cross-sensor, joint cross-subject and cross-sensor, and cross-subject cases, while the performance of support vector machine equipped with domain adaptation was not better than that of the baseline for cross-scenario case. Therefore, it is of great importance to investigate the impact of heterogeneity sources on classification performance and if needed, leverage domain adaptation methods to improve the performance.

Online Domain Adaptation for Continuous Cross-Subject Liver Viability Evaluation Based on Irregular Thermal Data

Abstract: Accurate evaluation of liver viability during its procurement is a challenging issue and has traditionally been addressed by taking an invasive biopsy of the liver. Recently, people have started to...

Graph Adaptive Knowledge Transfer for Unsupervised Domain Adaptation

Abstract: Unsupervised domain adaptation has caught appealing attentions as it facilitates the unlabeled target learning by borrowing existing well-established source domain knowledge. Recent practice on domain adaptation manages to extract effective features by incorporating the pseudo labels for the target domain to better solve cross-domain distribution divergences. However, existing approaches separate target label optimization and domain-invariant feature learning as different steps. To address that issue, we develop a novel Graph Adaptive Knowledge Transfer (GAKT) model to jointly optimize target labels and domain-free features in a unified framework. Specifically, semi-supervised knowledge adaptation and label propagation on target data are coupled to benefit each other, and hence the marginal and conditional disparities across different domains will be better alleviated. Experimental evaluation on two cross-domain visual datasets demonstrates the effectiveness of our designed approach on facilitating the unlabeled target task learning, compared to the state-of-the-art domain adaptation approaches.

Multimodal Data Guided Spatial Feature Fusion and Grouping Strategy for E-Commerce Commodity Demand Forecasting

Abstract: E-commerce offers various merchandise for selling and purchasing with frequent transactions and commodity flows. An accurate prediction of customer needs and optimized allocation of goods is required for cost reduction. The existing solutions have significant errors and are unsuitable for addressing warehouse needs and allocation. That is why businesses cannot respond to customer demands promptly, as they need accurate and reliable demand forecasting. Therefore, this paper proposes spatial feature fusion and grouping strategies based on multimodal data and builds a neural network prediction model for e-commodity demand. The designed model extracts order sequence features, consumer emotional features, and facial value features from multimodal data from e-commerce products. Then, a bidirectional long short-term memory network- (BiLSTM-) based grouping strategy is proposed. The proposed strategy fully learns the contextual semantics of time series data while reducing the influence of other features on the group’s local features. The output features of multimodal data are highly spatially correlated, and this paper employs the spatial dimension fusion strategy for feature fusion. This strategy effectively obtains the deep spatial relations among multimodal data by integrating the features of each column in each group across spatial dimensions. Finally, the proposed model’s prediction effect is tested using e-commerce dataset. The experimental results demonstrate the proposed algorithm’s effectiveness and superiority.

Digital Twin for Monitoring Ergonomics during Manufacturing Production

Abstract: Within the era of smart factories, concerning the ergonomics related to production processes, the Digital Twin (DT) is the key to set up novel models for monitoring the performance of manual work activities, which are able to provide results in near real time and to support the decision-making process for improving the working conditions. This paper aims to propose a methodological framework that, by implementing a human DT, and supports the monitoring and the decision making regarding the ergonomics performances of manual production lines. A case study, carried out in a laboratory, is presented for demonstrating the applicability and the effectiveness of the proposed framework. The results show how it is possible to identify the operational issues of a manual workstation and how it is possible to propose and test improving solutions.

Short and long-run causal effects of agriculture, forestry, and other land use on greenhouse gas emissions: evidence from China using VECM approach.

Abstract: Climate change caused by different anthropogenic activities is a subject of attention globally. There is a concern on how to maintain a clean environment and at the same time achieve optimal use of land. To this end, this study examines the causal effects of land use including agricultural, forestry, and other land categories on greenhouse gas (GHG) emissions. The data for China is collected over the period 1990 to 2012 for the empirical examination. By employing vector error correction model (VECM), it is found that there is significant long-run causality among variables. However, in the short run expectedly, only land under agriculture has strong causality with the GHG emissions. The results in case of variance decomposition analysis highlight that land under agriculture and other use significantly causes the GHG emissions in the long run. Further, impulse responses of variables are also measured with the Cholesky one standard deviation. The results are robust and support the argument that different land uses cause GHG emissions in China. The study provides insights for policy makers to improve the activities occurring on agricultural and other land uses. Assessment of overall potential, including bio energy, needs to include analysis of trade-offs and feedbacks with land-use competition. Many positive linkages with sustainable development and with adaptation exist but are case and site specific as they depend on scale, scope, and pace of implementation.

Smart Wearables for the Detection of Occupational Physical Fatigue: A Literature Review

Abstract: Today’s world is changing dramatically due to the influence of various factors. Whether due to the rapid development of technological tools, advances in telecommunication methods, global economic and social events, or other reasons, almost everything is changing. As a result, the concepts of a “job” or work have changed as well, with new work shifts being introduced and the office no longer being the only place where work is done. In addition, our non-stop active society has increased the stress and pressure at work, causing fatigue to spread worldwide and becoming a global problem. Moreover, it is medically proven that persistent fatigue is a cause of serious diseases and health problems. Therefore, monitoring and detecting fatigue in the workplace is essential to improve worker safety in the long term. In this paper, we provide an overview of the use of smart wearable devices to monitor and detect occupational physical fatigue. In addition, we present and discuss the challenges that hinder this field and highlight what can be done to advance the use of smart wearables in workplace fatigue detection.