Chaoyang Zhang

Mahid Ahmed, Ali Dogru, Chaoyang Zhang et al.

Strategically locating sawmills is critical for the efficiency, profitability, and sustainability of timber supply chains, yet it involves a series of complex decision-making affected by various factors, such as proximity to resources and markets, proximity to roads and rail lines, distance from the urban area, slope, labor market, and existing sawmill data. Although conventional Multi-Criteria Decision-Making (MCDM) approaches utilize these factors while locating facilities, they are susceptible to bias since they rely heavily on expert opinions to determine the relative factor weights. Machine learning (ML) models provide an objective, data-driven alternative for site selection that derives these weights directly from the patterns in large datasets without requiring subjective weighting. Additionally, ML models autonomously identify critical features, eliminating the need for subjective feature selection. In this study, we propose integrated ML and MCDM methods and showcase the utility of this integrated model to improve sawmill location decisions via a case study in Mississippi. This integrated model is flexible and applicable to site selection problems across various industries.

Chaoyang Zhang, Shaojie Chang, Ti Bai et al.

Photon counting spectral CT (PCCT) can produce reconstructed attenuation maps in different energy channels, reflecting energy properties of the scanned object. Due to the limited photon numbers and the non-ideal detector response of each energy channel, the reconstructed images usually contain much noise. With the development of Deep Learning (DL) technique, different kinds of DL-based models have been proposed for noise reduction. However, most of the models require clean data set as the training labels, which are not always available in medical imaging field. Inspiring by the similarities of each channel's reconstructed image, we proposed a self-supervised learning based PCCT image enhancement framework via multi-spectral channels (S2MS). In S2MS framework, both the input and output labels are noisy images. Specifically, one single channel image was used as output while images of other single channels and channel-sum image were used as input to train the network, which can fully use the spectral data information without extra cost. The simulation results based on the AAPM Low-dose CT Challenge database showed that the proposed S2MS model can suppress the noise and preserve details more effectively in comparison with the traditional DL models, which has potential to improve the image quality of PCCT in clinical applications.