Publications

Prediction of postoperative cardiac events in multiple surgical cohorts using a multimodal and integrative decision support system

Published in Scientific Reports, 2022

Postoperative patients are at risk of life-threatening complications such as hemodynamic decompensation or arrhythmia. Automated detection of patients with such risks via a real-time clinical decision support system may provide opportunities for early and timely interventions that can significantly improve patient outcomes. We utilize multimodal features derived from digital signal processing techniques and tensor formation, as well as the electronic health record (EHR), to create machine learning models that predict the occurrence of several life-threatening complications up to 4 hours prior to the event. In order to ensure that our models are generalizable across different surgical cohorts, we trained the models on a cardiac surgery cohort and tested them on vascular and non-cardiac acute surgery cohorts. The best performing models achieved an area under the receiver operating characteristic curve (AUROC) of 0 …

Recommended citation: R. B. Kim, O. P. Alge, G. Liu, B. E. Biesterveld, G. Wakam, A. M. Williams, M. R. Mathis, K. Najarian and J. Gryak, "Prediction of postoperative cardiac events in multiple surgical cohorts using a multimodal and integrative decision support system", Sci Rep, vol. 12, no. 1, p. 11347, Dec. 2022, doi: 10.1038/s41598-022-15496-w. https://www.nature.com/articles/s41598-022-15496-w

Predicting Poor Sleep Quality in Fibromyalgia with Wrist Sensors

Published in EMBC 2020, 2020

Fibromyalgia is a musculoskeletal disorder characterized by chronic, widespread muscle pain. This condition is associated with disturbed sleep, which has a direct impact on patient quality of life. Patient-reported outcomes are frequently used to assess sleep quality, but show modest correlations with objective measures of sleep, such as polysomnography. Working towards our goal of an automated ambulatory system of assessing sleep quality, we use features from blood volume pulse (BVP) and electrodermal activity (EDA) collected with a wearable device during sleep. We compare these measurements between individuals with fibromyalgia who experienced poor sleep and individuals in a control group who experienced refreshing sleep. By applying Learning Using Concave and Convex Kernels (LUCCK) and Support Vector Machines (SVM), we achieve mean Area Under the Receiver Operating Characteristic …

Recommended citation: O. Alge, S. M. Reza Soroushmehr, J. Gryak, A. Kratz, and K. Najarian, "Predicting Poor Sleep Quality in Fibromyalgia with Wrist Sensors", in 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada, Jul. 2020, pp. 4290–4293. doi: 10.1109/EMBC44109.2020.9176386. https://ieeexplore.ieee.org/abstract/document/9176386/

Automated Classification of Osteosarcoma and Benign Tumors using RNA-seq and Plain X-ray

Published in EMBC 2020, 2020

Osteosarcoma is a prominent bone cancer that typically affects adolescents or people in late adulthood. Early recognition of this disease relies on imaging technologies such as x-ray radiography to detect tumor size and location. This paper aims to differentiate osteosarcoma from benign tumors by analyzing both imaging and RNA-seq data through a combination of image processing and machine learning. In experimental results, the proposed method achieved an Area Under the Receiver Operator Characteristic Curve (AUC) of 0.7272 in three-fold cross-validation, and an AUC of 0.9015 using leave-one-out cross-validation.

Recommended citation: O. Alge, L. Lu, Z. Li, Y. Hua, J. Gryak and K. Najarian, "Automated Classification of Osteosarcoma and Benign Tumors using RNA-seq and Plain X-ray", 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada, 2020, pp. 1165-1168, doi: 10.1109/EMBC44109.2020.9176104. https://ieeexplore.ieee.org/abstract/document/9176104/

Classifying Osteosarcoma Using Meta-Analysis of Gene Expression

Published in BIBM 2018, 2018

Meta-analysis of gene expression provides the opportunity to compare gene expression across different platforms. In this paper, we use a meta-analysis of RNA-seq data collected by the SJTU team and publicly available microarray data to build a Random Forest classification model. The Random Forest model had average accuracy of 74.1% for cross-validation in the training set and achieved accuracy of 80.0% on the testing set.

Recommended citation: O. Alge, J. Gryak, Y. Hua, and K. Najarian, "Classifying Osteosarcoma Using Meta-Analysis of Gene Expression", in 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Madrid, Spain, Dec. 2018, pp. 2400–2404. doi: 10.1109/BIBM.2018.8621119. https://ieeexplore.ieee.org/abstract/document/8621119/

Differential signaling during macropinocytosis in response to M-CSF and PMA in macrophages

Published in Frontiers in Physiology, 2015

The cellular movements that construct a macropinosome have a corresponding sequence of chemical transitions in the cup-shaped region of plasma membrane that becomes the macropinosome. To determine the relative positions of type I phosphatidylinositol 3-kinase (PI3K) and phospholipase C (PLC) in this pathway, we analyzed macropinocytosis in macrophages stimulated by the growth factor macrophage-colony-stimulating factor (M-CSF) and by the diacylglycerol (DAG) analogue phorbol 12-myristate 13-acetate (PMA). In cells stimulated with M-CSF, microscopic imaging of fluorescent probes for intracellular lipids indicated that the PI3K product phosphatidylinositol (3,4,5)-trisphosphate (PIP3) appeared in cups just prior to DAG. We then tested the hypothesis that PMA and DAG function after PI3K and prior to Ras and protein kinase C (PKC) during macropinosome formation in macrophages. Although the PI3K target Akt was activated by M-CSF, the Akt inhibitor MK-2206 did not inhibit macropinocytosis. The phospholipase C (PLC) inhibitor U73122 blocked macropinocytosis by M-CSF but not PMA. Macropinocytosis in response to M-CSF and PMA was inhibited by the Ras inhibitor farnesyl thiosalicylate (FTS), by the PKC inhibitor Calphostin C and by the broad specificity inhibitor rottlerin. These studies support a model in which M-CSF stimulates PI3K in macropinocytic cups, and the resulting increase in PIP3 activates PLC, which in turn generates DAG necessary for activation of PKC, Ras and the late stages of macropinosome closure.

Recommended citation: S. Yoshida, I. Gaeta, R. Pacitto, L. Krienke, O. Alge, B. Gregorka, J. A. Swanson, "Differential signaling during macropinocytosis in response to M-CSF and PMA in macrophages" Front. Physiol. , vol. 6, Jan. 2015, doi: 10.3389/fphys.2015.00008. https://www.frontiersin.org/articles/10.3389/fphys.2015.00008/full