4 results on '"Stephen Sukumaran"'
Search Results
2. Comparing algorithms for assessing upper limb use with inertial measurement units.
- Author
-
Subash, Tanya, David, Ann, Reeta Janet Surekha, Stephen Sukumaran, Gayathri, Sankaralingam, kamalesh kumar, Selvaraj Samuel, Magimairaj, Henry Prakash, Malesevic, Nebojsa, Antfolk, Christian, S. K. M., Varadhan, Melendez-Calderon, Alejandro, and Balasubramanian, Sivakumar
- Subjects
RANDOM measures ,MACHINE learning ,RANDOM forest algorithms ,FUNCTIONAL training ,MACHINE performance - Abstract
The various existing measures to quantify upper limb use from wrist-worn inertial measurement units can be grouped into three categories: 1) Thresholded activity counting, 2) Gross movement score and 3) machine learning. However, there is currently no direct comparison of all these measures on a single dataset. While machine learning is a promising approach to detecting upper limb use, there is currently no knowledge of the information used by machine learning measures and the data-related factors that influence their performance. The current study conducted a direct comparison of the 1) thresholded activity counting measures, 2) gross movement score,3) a hybrid activity counting and gross movement score measure (introduced in this study), and 4) machine learning measures for detecting upper-limb use, using previously collected data. Two additional analyses were also performed to understand the nature of the information used by machine learning measures and the influence of data on the performance of machine learning measures. The intra-subject random forest machine learning measure detected upper limb use more accurately than all other measures, confirming previous observations in the literature. Among the non-machine learning (or traditional) algorithms, the hybrid activity counting and gross movement score measure performed better than the other measures. Further analysis of the random forest measure revealed that this measure used information about the forearm’s orientation and amount of movement to detect upper limb use. The performance of machine learning measures was influenced by the types of movements and the proportion of functional data in the training/testing datasets. The study outcomes show that machine learning measures perform better than traditional measures and shed some light on how these methods detect upper-limb use. However, in the absence of annotated data for training machine learning measures, the hybrid activity counting and gross movement score measure presents a reasonable alternative. We believe this paper presents a step towards understanding and optimizing measures for upper limb use assessment using wearable sensors. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
3. 1162. Epidemiology of Carbapenem-Resistant Pseudomonas aeruginosa Identified Through the Emerging Infections Program (EIP), United States, 2016–2017
- Author
-
Julian Grass, Sandra Bulens, Wendy Bamberg, Sarah J Janelle, Patrick Stendel, Jesse T Jacob, Chris Bower, Stephen Sukumaran, Lucy E Wilson, Elisabeth Vaeth, Linda Li, Ruth Lynfield, Paula Snippes Vagnone, Ginette Dobbins, Erin C Phipps, Emily B Hancock, Ghinwa Dumyati, Rebecca Tsay, Rebecca Pierce, P Maureen Cassidy, Nicole West, Marion A Kainer, Daniel Muleta, Jacquelyn Mounsey, Davina Campbell, Richard Stanton, Maria S Karlsson, and Maroya Spalding Walters
- Subjects
Abstracts ,Infectious Diseases ,Oncology ,B. Poster Abstracts - Abstract
Background Pseudomonas aeruginosa is intrinsically resistant to many commonly used antimicrobials and carbapenems are often required to treat infections. We describe the epidemiology and crude incidence of carbapenem-resistant P. aeruginosa(CRPA) in the EIP catchment area. Methods From August 1, 2016 through July 31, 2017, we conducted laboratory- and population-based surveillance for CRPA in selected metropolitan areas in Colorado, Georgia, Maryland, Minnesota, New Mexico, New York, Oregon, and Tennessee. We defined an incident case as the first isolate of P. aeruginosa-resistant to imipenem, meropenem, or doripenem from the lower respiratory tract, urine, wounds, or normally sterile sites identified from a resident of the EIP catchment area in a 30-day period. Patient charts were reviewed. A random sample of isolates was screened at CDC for carbapenemases using the modified carbapenem inactivation method (mCIM) and real-time PCR. Results During the 12-month period, we identified 3,042 incident cases among 2,154 patients. The crude incidence rate was 21.2 (95% CI, 20.4–21.9) per 100,000 persons and varied by site (range: 7.7 in Oregon to 31.1 in Maryland). The median age of patients was 64 years (range: Conclusion The burden of CRPA varied by EIP site. Most cases occurred in persons with healthcare exposures and underlying conditions. The majority of isolates were susceptible to at least one first-line antimicrobial. Carbapenemase producers were rare; a more specific phenotypic definition would greatly facilitate surveillance for these isolates. Disclosures All authors: No reported disclosures.
- Published
- 2018
4. Comparison of 30- and 90-Day Mortality Rates in Patients with Cultures Positive for Carbapenem-resistant Enterobacteriaceae and Acinetobacter in Atlanta, 2011–2015
- Author
-
Mary Elizabeth Sexton, Chris Bower, Jesse T. Jacob, and Stephen Sukumaran
- Subjects
Gerontology ,Klebsiella ,Carbapenem ,animal structures ,biology ,business.industry ,Mortality rate ,food and beverages ,Carbapenem-resistant enterobacteriaceae ,Enterobacter ,Acinetobacter ,biology.organism_classification ,Microbiology ,Acinetobacter baumannii ,body regions ,Abstracts ,Infectious Diseases ,Oncology ,Oral Abstract ,medicine ,In patient ,business ,medicine.drug - Abstract
Background Carbapenem-resistant Enterobacteriaceae (CRE) and Acinetobacter baumannii (CRAB) pose a threat to public health, but comparisons of disease burden are limited. We compared survival in patients following cultures positive for CRE or CRAB. Methods The Georgia Emerging Infections Program performs active population-based and laboratory-based surveillance for CRE and CRAB in metropolitan Atlanta, GA. Using standard CDC definitions, we included patients who had incident carbapenem-nonsusceptible E. coli, Klebsiella spp., Enterobacter spp., or Acinetobacter baumannii isolated from urine only (noninvasive infection) or a sterile site (invasive infection) between 8/2011 and 12/2015. Death dates, verified by Georgia Vital Statistics records, were used to calculate 30- and 90-day mortality rates. We used the chi-square test for mortality rates and the log-rank test for survival analysis to 90 days to compare patients with invasive CRAB, noninvasive CRAB, invasive CRE, and noninvasive CRE. Results There were 535 patients with CRE (87 invasive, 448 noninvasive) and 279 (78 invasive, 201 noninvasive) with CRAB. Nearly all patients with CRE and CRAB had healthcare exposures (97.2% vs. 100%) and most were immunosuppressed (62.6% vs. 56.3%). Both 30-day (24.4% vs. 18.3%, p = 0.04) and 90-day (37.6% vs. 30.5%, p = 0.04) mortality were higher in patients with CRAB than CRE. Patients with invasive infections were more likely to die at 90 days than those with noninvasive infections (53.3% vs. 38.4%, p < 0.0001). Overall mortality rates for invasive infection were similar between CRAB and CRE at 30 (44.9% vs. 34.5% p = 0.2) and 90 days (59.0% vs. 48.3%, p = 0.2). Using survival analysis at 90 days, invasive CRAB had the worst outcomes, followed by invasive CRE, noninvasive CRAB and noninvasive CRE (p < 0.0001, see Figure). Conclusion Ninety -day mortality for invasive infections with CRE and CRAB was ~50%, and patients with CRAB had lower survival than those with CRE, suggesting that prevention efforts may need to prioritize CRAB as highly as CRE in facilities with endemic CRAB. With the high proportion of healthcare exposures and immunosuppression, these infections may signify poor prognosis or directly contribute to mortality. Disclosures All authors: No reported disclosures.
- Published
- 2017
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.