Article
Computer Science, Interdisciplinary Applications
Jasleen Kaur Sethi, Mamta Mittal
Summary: This research investigates the effectiveness of a feature selection method based on LASSO for predicting air quality in Delhi and surrounding cities, identifying meteorological factors and pollutant concentrations as the most important influencing factors, and suggesting preventive measures to improve air quality.
EARTH SCIENCE INFORMATICS
(2021)
Article
Business, Finance
Elyas Elyasiani, Hadi Movaghari
Summary: This study examines the determinants of cash holdings in US firms using robust regression and variable selection techniques. The findings suggest that financial leverage is a key determinant of cash holdings, while the effectiveness of corporate governance mechanisms in managing cash is limited. Additionally, the predictors for cash holdings were significantly impacted by the financial crisis of 2008. The study has important implications for corporate cash managers, shareholders, monetary policy authorities, and scholars.
INTERNATIONAL REVIEW OF ECONOMICS & FINANCE
(2022)
Article
Multidisciplinary Sciences
Dan Zhang, Guotao Wu, Lichao Yang, Qiang Wu, Lianwen Yuan
Summary: In this study, we identified 5 m6A RNA methylated regulators that are significantly associated with colorectal cancer patients' overall survival. We developed a risk score model that can predict patient outcomes and distinguish cases based on gene expression data. These findings highlight the importance of these regulators in predicting patient prognosis and diagnosing colorectal cancer.
Article
Oncology
Lingge Yang, Yuan Wu, Huan Xu, Jingnan Zhang, Xinjie Zheng, Long Zhang, Yongfang Wang, Weiyu Chen, Kai Wang
Summary: This study aimed to establish a lncRNA-based model for predicting overall survival in LUAD patients. The model showed good predictive value and was found to be associated with immunotherapy and anti-angiogenic therapy sensitivity.
FRONTIERS IN ONCOLOGY
(2022)
Article
Automation & Control Systems
Tinghua Wang, Zhenwei Hu, Hanming Liu
Summary: Feature selection is a challenging and important task in machine learning and data mining. This paper proposes a unified view of supervised, unsupervised, and semi-supervised feature selection methods. The unified framework, based on the Hilbert-Schmidt independence criterion, shows that these methods share the same objective. It also provides a clear statistical interpretation and enables efficient computation of the global optimal solution through solving a Lasso optimization problem. Additionally, a new unsupervised feature selection algorithm is proposed and demonstrated with benchmark examples.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Construction & Building Technology
Xian Zheng, Rui Li, Yilong Han, Rui Xue, Ju Bai
Summary: Public-private partnerships (PPPs) are increasingly used for financing urban rail transit (URT) projects. This study examines the factors influencing government's choice of operators for URT projects. The findings identify key factors such as project operation duration, government-business relationships, and ownership type that significantly impact operator selection. The study contributes to URT project management knowledge and provides a systematic approach for local governments to select suitable operating partners.
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT
(2023)
Article
Energy & Fuels
Wai Yan Shum, Ning Ma, Xiaomei Lin, Tingting Han
Summary: China, one of the biggest energy consumers and carbon emitters in the world, sees carbon emissions mainly driven by economic growth and energy consumption, with income growth being the least significant factor among others. The ranking helps policymakers focus on critical contributors to carbon emissions and provides flexibility in policy interventions.
FRONTIERS IN ENERGY RESEARCH
(2021)
Article
Food Science & Technology
Leiming Yuan, Xueping Fu, Xiaofeng Yang, Xiaojing Chen, Guangzao Huang, Xi Chen, Wen Shi, Limin Li
Summary: This study developed a rapid and reliable method for non-destructive in-site determination of egg freshness using visible near-infrared (Vis-NIR) spectroscopy combined with interval partial least square regression (iPLS). Semi-transmittance spectra were collected from the waist of eggs and monitored every two days. A series of iPLS models were constructed based on spectral intervals to predict the Haugh unit (HU) of eggs, and the least absolute shrinkage and selection operator (Lasso) was used for model selection and fusion regression. The optimal model achieved the lowest root mean squared error of prediction (RMSEP) and outperformed other models.
Article
Obstetrics & Gynecology
Bohan Li, Hua Duan, Sha Wang, Yazhu Li
Summary: The study found that in patients with endometriosis, there is a progressive increase in changes of ferroptosis-related genes during the development of endometrium. A diagnostic model established by nine ferroptosis-associated genes can distinguish endometriosis patients from healthy patients, with the model having an area under the ROC curve of 0.979, indicating the clinical value of ferroptosis-associated genes in endometriosis diagnosis.
REPRODUCTIVE BIOMEDICINE ONLINE
(2021)
Article
Environmental Sciences
Angel Adhikari, Cristian R. Montes, Alicia Peduzzi
Summary: Recent advancements in laser scanning technology have shown great potential for accurately characterizing forests. However, the high correlation between lidar metrics poses a challenge when using them to predict forest attributes. This study compared four modeling methods using aerial lidar data and found that the ALASSO method performed the best, achieving high R-2 values for volume, basal area, and dominant height. The ALASSO method was also less biased compared to the other methods.
Article
Mathematics
Javier Sanchez Garcia, Salvador Cruz Rambaud
Summary: Vector autoregressions (VARs) and their variants are standard models in economic and financial research, but estimating high-dimensional models presents challenges. This paper explores machine learning regularization methods as an alternative to traditional methods, finding that they perform better in forecasting and impulse response analysis. Regularization structures allowing for high-dimensional models outperform standard Bayesian methods in nowcasting and forecasting.
Article
Engineering, Biomedical
Giulia Noaro, Giacomo Cappon, Martina Vettoretti, Giovanni Sparacino, Simone Del Favero, Andrea Facchinetti
Summary: This study introduces a new machine-learning model for improving MIB calculation in T1D therapy, achieving better results by considering CGM-Delta G and easily measurable features, particularly the LASSO(Q) model.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Dingliang Wang, Xuezhi Yang, Xuenan Liu, Likun Ma, Longwei Li, Wenjin Wang
Summary: This study proposed a novel noninvasive blood pressure estimation method using single-channel PPG signals, demonstrating improved accuracy compared to previous methods. The approach effectively rejected weak or redundant features and learned complex nonlinear relationships.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Oncology
Lei Lei, Li-Xin Du, Ying-Long He, Jian-Peng Yuan, Pan Wang, Bao-Lin Ye, Cong Wang, ZuJun Hou
Summary: This study proposes a feature selection method based on LASSO with dictionary learning for hepatocellular carcinoma grading. By learning a dictionary, the method enhances the high-information part and suppresses the low discriminative information, resulting in improved accuracy and discrimination in feature selection.
FRONTIERS IN ONCOLOGY
(2023)
Article
Cardiac & Cardiovascular Systems
Weihua Shi, Xiaoli Li, Yongxing Su, Dezhao Liu, Liying Wu, Shuo Li, Wenxiu He, Guoqiang Zhong, Zhiyuan Jiang
Summary: In this study, we investigated the significance of immune cell infiltration in atrial fibrillation (AF) and identified the potential Hub genes involved in the regulation of immune cell infiltration in AF. The results indicated that PILRA gene was closely related to multiple types of immune cell infiltration and may be a novel target for intervention in AF.
FRONTIERS IN CARDIOVASCULAR MEDICINE
(2023)
Article
Chemistry, Multidisciplinary
Hanna M. G. Barriga, Isaac J. Pence, Margaret N. Holme, James J. Doutch, Jelle Penders, Valeria Nele, Michael R. Thomas, Marta Carroni, Molly M. Stevens
Summary: This study combines SPARTA with SAXS/SANS techniques to investigate the relationship between LNP composition, internal structure, morphology, and PLD interactions. The findings suggest that the structure of the LNP lipid membrane can be used to control enzyme interactions, providing an opportunity for designing enzyme-responsive LNP solutions.
ADVANCED MATERIALS
(2022)
Article
Dermatology
Sara Moradi Tuchayi, Ying Wang, Alla Khodorova, Isaac J. Pence, Conor L. Evans, R. Rox Anderson, Ethan A. Lerner, Clifford J. Woolf, Lilit Garibyan
Summary: We developed a neural selective and injectable cryoneurolysis method using ice slurry, which effectively reduces cutaneous pain caused by skin disease. This method shows comparable effects on nerve structure and function compared to an FDA-approved cryoneurolysis device. The decrease in mechanical pain from slurry injection, although less profound than that of the FDA-approved device, lasts longer without inducing dysesthesia.
JOURNAL OF INVESTIGATIVE DERMATOLOGY
(2023)
Article
Surgery
Giju Thomas, Colleen M. Kiernan, Parker A. Willmon, Ezekiel Haugen, Amy N. Luckenbaugh, Daniel A. Barocas, Naira Baregamian, Anita Mahadevan-Jansen, Carmen C. Solorzano
Summary: This study introduces a novel label-free approach using near-infrared autofluorescence (NIRAF) detection to enhance intraoperative adrenal gland visualization, showing promising results in distinguishing adrenal cortex and surrounding structures, as well as differentiating malignant adrenal tumors from benign and healthy adrenal tissues in a cohort of 55 patients.
WORLD JOURNAL OF SURGERY
(2023)
Article
Surgery
Naira Baregamian, Konjeti R. Sekhar, Evan S. Krystofiak, Maria Vinogradova, Giju Thomas, Emmanuel Mannoh, Carmen C. Solorzano, Colleen M. Kiernan, Anita Mahadevan-Jansen, Naji Abumrad, Michael L. Freeman, Vivian L. Weiss, Jeffrey C. Rathmell, W. Kimryn Rathmell
Summary: By using fine needle aspiration, the researchers have successfully established 3-dimensional endocrine organoid models that accurately simulate the complex tumor microenvironment of thyroid, parathyroid, and adrenal neoplasms, while maintaining cytokine production and near-infrared autofluorescence properties.
Article
Medicine, Research & Experimental
Ryan H. Belcher, Giju Thomas, Parker A. Willmon, Jean-Nicolas Gallant, Naira Baregamian, Monica E. Lopez, Carmen C. Solorzano, Anita Mahadevan-Jansen
Summary: Compared to adult patients, pediatric patients undergoing thyroid surgery are more likely to develop hypoparathyroidism due to inadvertent injury or devascularization of the parathyroid gland. This study evaluates the utility and accuracy of near-infrared-autofluorescence (NIRAF) for parathyroid identification in pediatric patients. The results indicate that NIRAF detection can be a valuable and non-invasive technique for identifying parathyroid glands during neck operations in children.
Article
Spectroscopy
Rekha Gautam, Rafay Ahmed, Ezekiel Haugen, Mustafa Unal, Sean Fitzgerald, Sasidhar Uppuganti, Anita Mahadevan-Jansen, Jeffry S. Nyman
Summary: This study investigated the feasibility of using spatially offset Raman spectroscopy (SORS) to acquire Raman bands related to bone fracture resistance. The results showed that autoclaving of femur mid-shafts reduced the yield stress of cortical beams. Autoclaving also affected the Raman characteristics of the organic matrix, but changes in Raman properties related to bone strength could still be detected with SORS.
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY
(2023)
Article
Chemistry, Analytical
Rafay Ahmed, Mustafa Unal, Rekha Gautam, Sasidhar Uppuganti, Shrey Derasari, Anita Mahadevan-Jansen, Jeffry S. Nyman
Summary: The fracture resistance of bone is determined by the hierarchical arrangement of minerals, collagen fibrils, non-collagenous proteins, and water. Raman spectroscopy can detect changes in the protein structure and composition, making it a valuable tool for studying bone properties.
Proceedings Paper
Engineering, Electrical & Electronic
Shree Krishnamoorthy, Urbashi Basu, Kiangwei Kho, Rekha Gautam, Francesca Di Croce, Walter Messina, Cleitus Antony, Paul Townsend, Fergus P. McCarthy, Stefan Andersson-Engels, Ray Burke
Summary: Intra-partum hypoxia is the main cause of death in 2 out of 10,000 infants. Monitoring hypoxia during childbirth can prevent infant mortality and reduce the risk of cerebral palsy in 10-20% of surviving babies. Current monitoring techniques are either indirect or intermittently invasive. To address this, a non-invasive, continuous sensor based on LW-NIR spectroscopic technique is being developed to detect fetal hypoxia through multiple biomarkers.
PHOTONIC INSTRUMENTATION ENGINEERING X
(2023)
Proceedings Paper
Engineering, Biomedical
Hui Ma, Sanathana Konugolu Venkata Sekar, Rekha Gautam, Stefan Andersson-Engels
Summary: In this project, a beam-shaping system is demonstrated to manipulate the illumination patterns at the distal tip of the multimode fiber for endoscopic applications. This system has the potential to miniaturize the structured illumination system and the endoscope geometry, enabling surgical guidance and diagnosis.
ADAPTIVE OPTICS AND WAVEFRONT CONTROL FOR BIOLOGICAL SYSTEMS IX
(2023)
Article
Chemistry, Analytical
Siddra Maryam, Sanathana Konugolu Venkata Sekar, M. Daniyal Ghauri, Edward Fahy, Marcelo Saito Nogueira, Huihui Lu, Flavien Beffara, Georges Humbert, Richeal Ni Riordain, Patrick Sheahan, Ray Burke, Kiang Wei Kho, Rekha Gautam, Stefan Andersson-Engels
Summary: Early diagnosis of oral cancer is critical for improving patient survival rates. Raman spectroscopy has shown promise as a non-invasive method for identifying early-stage oral cancer biomarkers. However, its widespread use is limited due to weak signals that require costly sensitive detectors. In this research, a customised Raman system is developed to reduce costs for different applications and demonstrate the potential for complete oral cancer screening.
Article
Chemistry, Analytical
Giju Thomas, Sean T. T. Fitzgerald, Rekha Gautam, Fuyao Chen, Ezekiel Haugen, Pratheepa Kumari Rasiah, Wilson R. R. Adams, Anita Mahadevan-Jansen
Summary: Biochemical insights into varying breast cancer phenotypes can be obtained through Raman spectroscopy, which provides high specificity in measuring the biochemical differences. In this study, stainless steel substrates were found to deliver stronger Raman signals and improved spectral characterization compared to CaF2 substrates.
ANALYTICAL METHODS
(2023)
Article
Chemistry, Analytical
Yuxiao Wei, Isaac J. Pence, Anna Wiatrowski, Julia B. Slade, Conor L. Evans
Summary: Pharmaceutical development of solid-state formulations requires chemical imaging techniques to analyze the solid-state properties for ensuring drug efficacy and stability. Current chemical imaging techniques have high chemical and spatial resolution but lack the necessary measurement speed for pharmaceutical production and quality assurance processes. To address this issue, fast chemical imaging using stimulated Raman scattering can quantitatively analyze the degradation and distribution of drugs at a faster speed and higher resolution.
Article
Chemistry, Analytical
Rekha Gautam, Danielle Mac Mahon, Grainne Eager, Hui Ma, Claudia Nunzia Guadagno, Stefan Andersson-Engels, Sanathana Konugolu Venkata Sekar
Summary: This study presents the design and validation of gelatin-based phantoms for simulating tissue optical properties, serving as a calibration and benchmarking tool for multimodal spectroscopy devices.
Proceedings Paper
Optics
Hui Ma, Sanathana Konugolu Venkata Sekara, Pranav Lanka, Carrie O'Flynn, Patrick Henn, Stefan Andersson-Engels, Rekha Gautam
Summary: By implementing the dual-wavelength inverse Spatially Offset Raman Spectroscopy (SORS) technique, we can comprehensively assess the chemical composition and microstructure of bones, thereby improving the accuracy of osteoporosis diagnosis.
QUANTUM TECHNOLOGY: DRIVING COMMERCIALISATION OF AN ENABLING SCIENCE III
(2022)