Digital Commons Network^™

Brain Imaging With Portable Low-Field Mri, Rachel Beekman, Francis X Shen, Gordon Sze, Pamela Schaefer, Juan Eugenio Iglesias, Matthew S Rosen, Kevin N Sheth

Faculty, Staff and Student Publications

The advent of portable, low-field MRI (LF-MRI) heralds new opportunities in neuroimaging. Low power requirements and transportability have enabled scanning outside the controlled environment of a conventional MRI suite, enhancing access to neuroimaging for indications that are not well suited to existing technologies. Maximizing the information extracted from the reduced signal-to-noise ratio of LF-MRI is crucial to developing clinically useful diagnostic images. Progress in electromagnetic noise cancellation and machine learning reconstruction algorithms from sparse k-space data as well as new approaches to image enhancement have now enabled these advancements. Coupling technological innovation with bedside imaging creates new prospects in …

The Y Of Cancer Sex Differences, Jiexi Li, Ronald A Depinho

Faculty, Staff and Student Publications

No abstract provided.

Greater Male Variability In Daily Energy Expenditure Develops Through Puberty, Lewis G Halsey, Vincent Careau, Philip N Ainslie, Heliodoro Alemán-Mateo, Lene F Andersen, Leonore Arab, Linda Bandini, Kweku Bedu-Addo, Ellen E Blaak, Stephane Blanc, Alberto G Bonomi, Carlijn V C Bouten, Pascal Bovet, Soren Brage, Maciej S Buchowski, Nancy Butte, Stephan G Camps, Regian Casper, Graeme L Close, Lisa H Colbert, Jamie A Cooper, Richard Cooper, Prasangi Dabare, Sai Krupa Das, Peter S W Davies, Sanjoy Deb, Christine Delisle Nyström, William Dietz, Lara R Dugas, Simon Eaton, Ulf Ekelund, Asmaa El Hamdouchi, Sonja Entringer, Terrence Forrester, Barry W Fudge, Melanie Gillingham, Annelies H Goris, Michael Gurven, Hinke Haisma, Catherine Hambly, Daniel Hoffman, Marije B Hoos, Sumei Hu, Noorjehan Joonas, Annemiek Joosen, Peter Katzmarzyk, Kitty P Kempen, Misaka Kimura, William E Kraus, Wantanee Kriengsinyos, Rebecca Kuriyan, Robert F Kushner, Estelle V Lambert, Pulani Lanerolle, Christel L Larsson, Nader Lessan, Marie Löf, Corby Martin, Eric Matsiko, Gerwin A Meijer, James C Morehen, James P Morton, Aviva Must, Marian Neuhouser, Theresa A Nicklas, Robert M Ojiambo, Kirsi H Pietilainen, Yannis P Pitsiladis, Jacob Plange-Rhule, Guy Plasqui, Ross L Prentice, Roberto Rabinovich, Susan B Racette, David A Raichen, Eric Ravussin, Leanne Redman, John J Reilly, Rebecca Reynolds, Susan Roberts, Jennifer C Rood, Dulani Samaranayake, Luís B Sardinha, Albertine J Scuitt, Analiza M Silva, Srishti Sinha, Anders M Sjödin, Eric Stice, Albert Stunkard, Samuel S Urlacher, Mauro Eduardo Valencia, Giulio Valenti, Ludo M Van Etten, Edgar A Van Mil, Jeanine A Verbunt, Jonathan C K Wells, George Wilson, Brian Wood, Tsukasa Yoshida, Xueying Zhang, Alexia Murphy-Alford, Cornelia Loechl, Amy Luke, Herman Pontzer, Jennifer Rood, Hiroyuki Sagayama, Klaas R Westerterp, William W Wong, Yosuke Yamada, John R Speakman

Faculty, Staff and Student Publications

There is considerably greater variation in metabolic rates between men than between women, in terms of basal, activity and total (daily) energy expenditure (EE). One possible explanation is that EE is associated with male sexual characteristics (which are known to vary more than other traits) such as musculature and athletic capacity. Such traits might be predicted to be most prominent during periods of adolescence and young adulthood, when sexual behaviour develops and peaks. We tested this hypothesis on a large dataset by comparing the amount of male variation and female variation in total EE, activity EE and basal EE, at …

Feasibility And Clinical Acceptability Of Automation-Assisted 3d Conformal Radiotherapy Planning For Patients With Cervical Cancer In A Resource-Constrained Setting, Sikudhani Muya, Jerry Ndumbalo, Sarah Kutika Nyagabona, Shaid Yusuf, Dong Joo Rhee, Benjamin Li, Surbhi Grover, Mary Feng, I-Chow Hsu, Elia Mmbaga, Katherine Van Loon, Laurence Court, Melody Xu

Faculty, Staff and Student Publications

Purpose: The Ocean Road Cancer Institute (ORCI) in Tanzania began offering 3D conformal radiation therapy (3DCRT) in 2018. Steep learning curves, high patient volume, and a limited workforce resulted in long radiation therapy (RT) planning workflows. We aimed to establish the feasibility of implementing an automation-assisted cervical cancer 3DCRT planning system.

Materials and methods: We performed chart abstractions on 30 patients with cervical cancer treated with 3DCRT at ORCI. The Radiation Planning Assistant (RPA) generated a new automated set of contours and plans on the basis of anonymized computed tomography images. Each were assessed for edit time requirements, dose-volume safety …

Making Waves: Integrating Wastewater Surveillance With Dynamic Modeling To Track And Predict Viral Outbreaks, Tin Phan, Samantha Brozak, Bruce Pell, Jeremiah Oghuan, Anna Gitter, Ruy M Ribeiro, Kristina D Mena, Alan S Perelson, Yang Kuang, Fuqing Wu

Faculty, Staff and Student Publications

Wastewater surveillance has proved to be a valuable tool to track the COVID-19 pandemic. However, most studies using wastewater surveillance data revolve around establishing correlations and lead time relative to reported case data. In this perspective, we advocate for the integration of wastewater surveillance data with dynamic within-host and between-host models to better understand, monitor, and predict viral disease outbreaks. Dynamic models overcome emblematic difficulties of using wastewater surveillance data such as establishing the temporal viral shedding profile. Complementarily, wastewater surveillance data bypasses the issues of time lag and underreporting in clinical case report data, thus enhancing the utility and …

Lineage Plasticity And Stemness Phenotypes In Prostate Cancer: Harnessing The Power Of Integrated "Omics" Approaches To Explore Measurable Metrics, Souzana Logotheti, Eugenia Papadaki, Vasiliki Zolota, Christopher Logothetis, Aristidis G Vrahatis, Vasiliki Tzelepi

Faculty, Staff and Student Publications

Prostate cancer (PCa), the most frequent and second most lethal cancer type in men in developed countries, is a highly heterogeneous disease. PCa heterogeneity, therapy resistance, stemness, and lethal progression have been attributed to lineage plasticity, which refers to the ability of neoplastic cells to undergo phenotypic changes under microenvironmental pressures by switching between developmental cell states. What remains to be elucidated is how to identify measurements of lineage plasticity, how to implement them to inform preclinical and clinical research, and, further, how to classify patients and inform therapeutic strategies in the clinic. Recent research has highlighted the crucial role …

Clonal Evolution And Hierarchy In Myeloid Malignancies, Koichi Takahashi, Tomoyuki Tanaka

Faculty, Staff and Student Publications

Myeloid malignancies, a group of hematopoietic disorders that includes acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and myeloproliferative neoplasms (MPNs), are caused by the accumulation of genetic and epigenetic changes in hematopoietic stem and progenitor cells (HSPCs) over time. Despite the relatively low number of genomic drivers compared with other forms of cancer, the process by which these changes shape the genomic architecture of myeloid malignancies remains elusive. Recent advancements in clonal hematopoiesis research and the use of cutting-edge single cell technologies have shed new light on the developmental process of myeloid malignancies. In this review, we delve into the …

Key Genetic Determinants Driving Esophageal Squamous Cell Carcinoma Initiation And Immune Evasion, Kyung-Pil Ko, Yuanjian Huang, Shengzhe Zhang, Gengyi Zou, Bongjun Kim, Sohee Jun, Karen J Dunbar, Gizem Efe, Anil K Rustgi, Hiroshi Nakagawa, Jae-Il Park

Faculty, Staff and Student Publications

BACKGROUND & AIMS: Despite recent progress in identifying aberrant genetic and epigenetic alterations in esophageal squamous cell carcinoma (ESCC), the mechanism of ESCC initiation remains unknown.

METHODS: Using CRISPR/Cas 9-based genetic ablation, we targeted 9 genes (TP53, CDKN2A, NOTCH1, NOTCH3, KMT2D, KMT2C, FAT1, FAT4, and AJUBA) in murine esophageal organoids. Transcriptomic phenotypes of organoids and chemokine released by organoids were analyzed by single-cell RNA sequencing. Tumorigenicity and immune evasion of organoids were monitored by allograft transplantation. Human ESCC single-cell RNA sequencing data sets were analyzed to classify patients and find subsets relevant to organoid models and immune evasion.

RESULTS: We …

Vascularized Hepatocellular Carcinoma On A Chip To Control Chemoresistance Through Cirrhosis, Inflammation And Metabolic Activity, Alican Özkan, Danielle L Stolley, Erik N K Cressman, Matthew Mcmillin, Thomas E Yankeelov

Faculty, Staff and Student Publications

Understanding the effects of inflammation and cirrhosis on the regulation of drug metabolism during the progression of hepatocellular carcinoma (HCC) is critical for developing patient-specific treatment strategies. In this work, we created novel three-dimensional vascularized HCC-on-a-chips (HCCoC), composed of HCC, endothelial, stellate, and Kupffer cells tuned to mimic normal or cirrhotic liver stiffness. HCC inflammation was controlled by tuning Kupffer macrophage numbers, and the impact of cytochrome P450-3A4 (CYP3A4) was investigated by culturing HepG2 HCC cells transfected with CYP3A4 to upregulate expression from baseline. This model allowed for the simulation of chemotherapeutic delivery methods such as intravenous injection and transcatheter …

Paradoxes Of Cellular Sumoylation Regulation: A Role Of Biomolecular Condensates?

Faculty, Staff and Student Publications

Protein SUMOylation is a major post-translational modification essential for maintaining cellular homeostasis. SUMOylation has long been associated with stress responses as a diverse array of cellular stress signals are known to trigger rapid alternations in global protein SUMOylation. In addition, while there are large families of ubiquitination enzymes, all small ubiquitin-like modifiers (SUMOs) are conjugated by a set of enzymatic machinery comprising one heterodimeric SUMO-activating enzyme, a single SUMO-conjugating enzyme, and a small number of SUMO protein ligases and SUMO-specific proteases. How a few SUMOylation enzymes specifically modify thousands of functional targets in response to diverse cellular stresses remains an …

Machine Independence Of Ultrasound-Based Quantification Of Vitreous Echodensities, J Sebag

Faculty, Staff and Student Publications

PURPOSE: Quantitative ultrasound (QUS) provides objective indices of Vision Degrading Myodesopsia (VDM) that correlate with contrast sensitivity (CS). To date, QUS methods were only tested on a single ultrasound machine. Here, we evaluate whether QUS measurements are machine independent.

METHODS: In this cross-sectional study, 47 eyes (24 subjects; age = 53.2 ± 14.4 years) were evaluated with Freiburg acuity contrast testing (%Weber), and ultrasonography using 2 machines: one with a 15-MHz single-element transducer and one with a 5-ring, 20-MHz annular-array. Images were acquired from each system in sequential scans. Artifact-free, log-compressed envelope data were processed to yield three parameters (mean …

Cell-Based Noninvasive Prenatal Testing (Cbnipt)-A Review On The Current Developments And Future Prospects, Mohamad Ali Maktabi, Liesbeth Vossaert, Ignatia B Van Den Veyver

Center for Medical Ethics and Health Policy Staff Publications

Considering the diagnostic limitations of cfDNA-based noninvasive prenatal testing (NIPT), scientists have long been interested in isolating and analyzing rare intact fetal and trophoblast cells from maternal blood or endocervical samples to diagnose fetal genetic conditions. These cells may be scarce and difficult to isolate, but they are a direct source of pure fetal genetic material. In this review, we summarize the history of cell-based NIPT, present an updated review on its current developments, evaluate its genetic diagnostic potential, and discuss its future prospects for clinical use.

Restoration Management Of Phosphorus Pollution On Lowland Fen Peatlands: A Data Evidence Review From The Somerset Levels And Moors, Sean Comber, Paul Lunt, Mark Taylor, Natasha Underwood, Ry Crocker, Rob Schindler

School of Geography, Earth and Environmental Sciences

Eutrophication of wetlands caused by urban, industrial and agricultural run-off is an important environmental problem. Eutrophication is characterized by excessive plant and algal growth due to the increased availability of one or more growth “limiting nutrients”, in freshwater generally considered to be controlled by the bioavailability of phosphorus (P). The Somerset Levels and Moors (SLMs) catchments are subject to intensive agriculture and wastewater inputs which leads to nutrient contamination of the inflow waters, to the extent that they fail Water Framework Directive Good Status targets for P concentrations. In 2021, Natural England downgraded the status of the SLMs Sites of …

Transcranial Focused Ultrasound-Mediated Neurochemical And Functional Connectivity Changes In Deep Cortical Regions In Humans, Siti N. Yaakub, Tristan A. White, Jamie Roberts, Eleanor Martin, Lennart Verhagen, Charlotte J. Stagg, Stephen Hall, Elsa F. Fouragnan

School of Psychology

Low-intensity transcranial ultrasound stimulation (TUS) is an emerging non-invasive technique for focally modulating human brain function. The mechanisms and neurochemical substrates underlying TUS neuromodulation in humans and how these relate to excitation and inhibition are still poorly understood. In 24 healthy controls, we separately stimulated two deep cortical regions and investigated the effects of theta-burst TUS, a protocol shown to increase corticospinal excitability, on the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) and functional connectivity. We show that theta-burst TUS in humans selectively reduces GABA levels in the posterior cingulate, but not the dorsal anterior cingulate cortex. Functional connectivity increased following TUS …

Enhancing The Co2 Trapping Capacity Of Saudi Arabian Basalt Via Nanofluid Treatment: Implications For Co2 Geo-Storage, Muhammad Ali, Nurudeen Yekeen, Mirhasan Hosseini, Ghazanfer R. Abbasi, Amer Alanazi, Alireza Keshavarz, Thomas Finkbeiner, Hussein Hoteit

Research outputs 2022 to 2026

Mineralization reactions in basaltic formations have gained recent interest as an effective method for CO2 geo-storage in order to mitigate anthropogenic greenhouse gas emissions. The CO2/rock interactions, including interfacial tension and wettability, are crucial factors in determining the CO2 trapping capacity and the feasibility of CO2 geological storage in these formations. The Red Sea geological coast in Saudi Arabia has many basaltic formations, and their wetting characteristics are rarely reported in the literature. Moreover, organic acid contamination is inherent in geo-storage formations and significantly impacts their CO2 geo-storage capacities. Hence, to reverse the organic effect, the influence of various SiO2 …

Harnessing The Power Of Computational Fluid Dynamics For Flow Coefficient And Rain Resistance Improvement Of Type 1 Natural Ventilators, Amin Roostaee, Mehdi Khiadani, Hussein A. Mohammed, Abdellah Shafieian

Research outputs 2022 to 2026

Type 1 Natural Ventilators act like a shield protecting the building from rain intrusion while allowing air to flow from outside. Despite their wide applications, they are far from the ideal design, mainly due to a lack of data on their air discharge coefficient and rain defense effectiveness under different climatic conditions. According to ventilation standards, since experimental testing of different geometries and configurations of these ventilators are complex, difficult, and costly, this study aims to implement Computational Fluid Dynamics (CFD) tools to investigate their performance. To do this, a combination of 9 various louvre panels was designed with 1, …

Pymaivar: An Open-Source Python Suit For Audio-Image Representation In Human Action Recognition, Muhammad B. Shaikh, Douglas Chai, Syed M. S. Islam, Naveed Akhtar

Research outputs 2022 to 2026

We present PyMAiVAR, a versatile toolbox that encompasses the generation of image representations for audio data including Wave plots, Spectral Centroids, Spectral Roll Offs, Mel Frequency Cepstral Coefficients (MFCC), MFCC Feature Scaling, and Chromagrams. This wide-ranging toolkit generates rich audio-image representations, playing a pivotal role in reshaping human action recognition. By fully exploiting audio data's latent potential, PyMAiVAR stands as a significant advancement in the field. The package is implemented in Python and can be used across different operating systems.

Establishing Physical And Chemical Mechanisms Of Polymerization And Pyrolysis Of Phenolic Resins For Carbon-Carbon Composites, Ivan Gallegos, Josh Kemppainen, Jacob R. Gissinger, Malgorzata Kowalik, Adri Van Duin, Kristopher E. Wise, S. Gowtham, Gregory Odegard

Michigan Tech Publications

The complex structural and chemical changes that occur during polymerization and pyrolysis critically affect material properties but are difficult to characterize in situ. This work presents a novel, experimentally validated methodology for modeling the complete polymerization and pyrolysis processes for phenolic resin using reactive molecular dynamics. The polymerization simulations produced polymerized structures with mass densities of 1.24 ± 0.01 g/cm3 and Young's moduli of 3.50 ± 0.64 GPa, which are in good agreement with experimental values. The structural properties of the subsequently pyrolyzed structures were also found to be in good agreement with experimental X-ray data for the phenolic-derived carbon …

Carbohydrates Generated Via Hot Water As Catalyst For Co2 Reduction Reaction, Yang Yang, Heng Zhong, Jiong Cheng, Yun Hang Hu, Richard Lee Smith Jr, Fangming Jin

Michigan Tech Publications

Combining terrestrial biomass with submarine-type hydrothermal environments for CO2 reduction is a possible approach for realizing new energies while achieving sustainable circulation of carbon. Herein, carbohydrateenabled CO2 reduction based on NaHCO3 conversion to formate revealed that hydrothermal environments facilitated direct hydrogen transfer from carbohydrates (glucose, cellulose) to CO2/NaHCO3 with hot water (250–300 °C, 5–20 MPa) acting as homogeneous catalyst in absence of any conventional catalysts giving CO2/ NaHCO3 reduction efficiencies as high as 76% for cellulose. Time-resolved operando hydrothermal DRIFTS spectra of glycolaldehyde in hot water (250 °C, autogenous pressure) verified that water catalyzed NaHCO3 reduction by converting the -CHO …

A Numerical Investigation Realizing The Flow Structure And Heat Transfer Performance Of The Potential Impinging Jets, Mohammed Sami Uddin Khan

Theses

The demand for improvement in the performance of gas turbines has led to the consideration of flows at increasingly high temperatures, but this introduces challenges in terms of maintaining their structural integrity and preventing overheating. To respond to these challenges, gas turbine manufacturers have turned to internal cooling, and jet impingement provides an effective solution for cooling the leading edge of the blades of gas turbines. In this study, the author numerically simulated the cooling performance of the leading edge of the blades of a gas turbine under constant heat flux by using five configurations of jet impingement: a steady …

Dual-Site Photoplethysmography Sensing For Noninvasive Continuous-Time Blood Pressure Monitoring Using Artificial Neural Network, Anas Mohmmad Rabab’Ah

Theses

Millions of people worldwide struggle from high blood pressure, often known as hypertension, and it is a major health concern that can lead to serious cardiovascular diseases, including heart attacks and many other consequences. Blood pressure monitoring that is reliable and accurate is crucial to the detection and management of hypertension. Although invasive techniques, such as arterial catheterization, are considered to be the most accurate means of evaluating blood pressure, they can be painful, time-consuming and carry a risk of complications.
/="/">This thesis presents the development of a real time non-invasive blood pressure monitoring system based on commercially available …

Upgrading Biomass To Bio-Oil: Catalytic Hydrodeoxygenation And Pyrolysis Of Vanillin And Anisole Biomass Model Compounds, Ahmad J. Yaghi

Theses

The purpose of this project is to investigate capacity of selected formulations of catalysts toward the hydrodeoxygenation (HDO) of biomass model compounds in the gas phase with the underlying aim to produce hydrocarbon cuts that can be directly used as transportation fuels. The first part of this project is to devoted to illustrate procedures for the synthesis of potent (HDO) catalysts. Synthesized catalysts encompass nickel (Ni), cobalt (Co) and palladium (Pd) supported on zeolite, cerium oxide (CeO₂) or alumina (Al₂O₃). Various characterization techniques such as X-ray Diffraction (XRD), Temperature Programmed Reduction (TPR) were deployed …

On Predicting Esg Ratings Using Dynamic Company Networks, Gary Ang, Zhiling Guo, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Environmental, social and governance (ESG) considerations play an increasingly important role due to the growing focus on sustainability globally. Entities, such as banks and investors, utilize ESG ratings of companies issued by specialized rating agencies to evaluate ESG risks of companies. The process of assigning ESG ratings by human analysts is however laborious and time intensive. Developing methods to predict ESG ratings could alleviate such challenges, allow ESG ratings to be generated in a more timely manner, cover more companies, and be more accessible. Most works study the effects of ESG ratings on target variables such as stock prices or …

Carbon-Aware Mine Planning With A Novel Multi-Objective Framework, Nurul Asyikeen Binte Azhar, Aldy Gunawan, Shih-Fen Cheng, Erwin Leonardi

Research Collection School Of Computing and Information Systems

The logistical complication of long-term mine planning involves deciding the sequential extraction of materials from the mine pit and their subsequent processing steps based on geological, geometrical, and resource constraints. The net present value (NPV) of profit over the mine's lifespan usually forms the sole objective for this problem, which is considered as the NP-hard precedence-constrained production scheduling problem (PCPSP) as well. However, increased pressure for more sustainable and carbon-aware industries also calls for environmental indicators to be considered. In this paper, we enhance the generic PCPSP formulation into a multi-objective optimization (MOO) problem whereby carbon cost forms an additional …

Constrained Multiagent Reinforcement Learning For Large Agent Population, Jiajing Ling, Arambam James Singh, Duc Thien Nguyen, Akshat Kumar

Research Collection School Of Computing and Information Systems

Learning control policies for a large number of agents in a decentralized setting is challenging due to partial observability, uncertainty in the environment, and scalability challenges. While several scalable multiagent RL (MARL) methods have been proposed, relatively few approaches exist for large scale constrained MARL settings. To address this, we first formulate the constrained MARL problem in a collective multiagent setting where interactions among agents are governed by the aggregate count and types of agents, and do not depend on agents’ specific identities. Second, we show that standard Lagrangian relaxation methods, which are popular for single agent RL, do not …

Threshold Attribute-Based Credentials With Redactable Signature, Rui Shi, Huamin Feng, Yang Yang, Feng Yuan, Yingjiu Li, Hwee Hwa Pang, Robert H. Deng

Research Collection School Of Computing and Information Systems

Threshold attribute-based credentials are suitable for decentralized systems such as blockchains as such systems generally assume that authenticity, confidentiality, and availability can still be guaranteed in the presence of a threshold number of dishonest or faulty nodes. Coconut (NDSS'19) was the first selective disclosure attribute-based credentials scheme supporting threshold issuance. However, it does not support threshold tracing of user identities and threshold revocation of user credentials, which is desired for internal governance such as identity management, data auditing, and accountability. The communication and computation complexities of Coconut for verifying credentials are linear in the number of each user's attributes and …

K-St: A Formal Executable Semantics Of The Structured Text Language For Plcs, Kun Wang, Jingyi Wang, Christopher M. Poskitt, Xiangxiang Chen, Jun Sun, Peng Cheng

Research Collection School Of Computing and Information Systems

Programmable Logic Controllers (PLCs) are responsible for automating process control in many industrial systems (e.g. in manufacturing and public infrastructure), and thus it is critical to ensure that they operate correctly and safely. The majority of PLCs are programmed in languages such as Structured Text (ST). However, a lack of formal semantics makes it difficult to ascertain the correctness of their translators and compilers, which vary from vendor-to-vendor. In this work, we develop K-ST, a formal executable semantics for ST in the K framework. Defined with respect to the IEC 61131-3 standard and PLC vendor manuals, K-ST is a high-level …

Experimental Comparison Of Features, Analyses, And Classifiers For Android Malware Detection, Lwin Khin Shar, Biniam Fisseha Demissie, Mariano Ceccato, Naing Tun Yan, David Lo, Lingxiao Jiang, Christoph Bienert

Research Collection School Of Computing and Information Systems

Android malware detection has been an active area of research. In the past decade, several machine learning-based approaches based on different types of features that may characterize Android malware behaviors have been proposed. The usually-analyzed features include API usages and sequences at various abstraction levels (e.g., class and package), extracted using static or dynamic analysis. Additionally, features that characterize permission uses, native API calls and reflection have also been analyzed. Initial works used conventional classifiers such as Random Forest to learn on those features. In recent years, deep learning-based classifiers such as Recurrent Neural Network have been explored. Considering various …

Generative Model-Based Testing On Decision-Making Policies, Zhuo Li, Xiongfei Wu, Derui Zhu, Mingfei Cheng, Siyuan Chen, Fuyuan Zhang, Xiaofei Xie, Lei Ma, Jianjun Zhao

Research Collection School Of Computing and Information Systems

The reliability of decision-making policies is urgently important today as they have established the fundamentals of many critical applications, such as autonomous driving and robotics. To ensure reliability, there have been a number of research efforts on testing decision-making policies that solve Markov decision processes (MDPs). However, due to the deep neural network (DNN)-based inherit and infinite state space, developing scalable and effective testing frameworks for decision-making policies still remains open and challenging.In this paper, we present an effective testing framework for decision-making policies. The framework adopts a generative diffusion model-based test case generator that can easily adapt to different …

Edge Distraction-Aware Salient Object Detection, Sucheng Ren, Wenxi Liu, Jianbo Jiao, Guoqiang Han, Shengfeng He

Research Collection School Of Computing and Information Systems

Integrating low-level edge features has been proven to be effective in preserving clear boundaries of salient objects. However, the locality of edge features makes it difficult to capture globally salient edges, leading to distraction in the final predictions. To address this problem, we propose to produce distraction-free edge features by incorporating cross-scale holistic interdependencies between high-level features. In particular, we first formulate our edge features extraction process as a boundary-filling problem. In this way, we enforce edge features to focus on closed boundaries instead of those disconnected background edges. Second, we propose to explore cross-scale holistic contextual connections between every …