Vijay R Ramakrishnan1, Jaron Arbet2, Jess C Mace3, Krithika Suresh2, Stephanie Shintani Smith4, Zachary M Soler5, Timothy L Smith3. 1. Department of Otolaryngology-Head and Neck Surgery, University of Colorado, Aurora, CO, USA. 2. Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado-Denver Anschutz Medical Campus, Aurora, CO, USA. 3. Department of Otolaryngology-Head and Neck Surgery, Oregon Health Sciences University, Portland, OR, USA. 4. Department of Otolaryngology-Head and Neck Surgery, Northwestern University, Chicago, IL, USA. 5. Department of Otolaryngology-Head and Neck Surgery, Medical University of South Carolina, Charleston, SC, USA.
Abstract
OBJECTIVE: Compare machine learning (ML)-based predictive analytics methods to traditional logistic regression in classification of olfactory dysfunction in chronic rhinosinusitis (CRS-OD) and identify predictors within a large multi-institutional cohort of refractory CRS patients. METHODS: Adult CRS patients enrolled in a prospective, multi-institutional, observational cohort study were assessed for baseline CRS-OD using a smell identification test (SIT) or brief SIT (bSIT). Four different ML methods were compared to traditional logistic regression for classification of CRS normosmics versus CRS-OD. RESULTS: Data were collected for 611 study participants who met inclusion criteria between 2011 April and 2015 July. Thirty-four percent of enrolled patients demonstrated olfactory loss on psychophysical testing. Differences between CRS normosmics and those with smell loss included objective disease measures (CT and endoscopy scores), age, sex, prior surgeries, socioeconomic status, steroid use, polyp presence, asthma, and aspirin sensitivity. Most ML methods performed favorably in terms of predictive ability. Top predictors include factors previously reported in the literature, as well as several socioeconomic factors. CONCLUSION: Olfactory dysfunction is a variable phenomenon in CRS patients. ML methods perform well compared to traditional logistic regression in classification of normosmia versus smell loss in CRS, and are able to include numerous risk factors into prediction models. Several actionable features were identified as risk factors for CRS-OD. These results suggest that ML methods may be useful for current understanding and future study of hyposmia secondary to sinonasal disease, the most common cause of persistent olfactory loss in the general population.
OBJECTIVE: Compare machine learning (ML)-based predictive analytics methods to traditional logistic regression in classification of olfactory dysfunction in chronic rhinosinusitis (CRS-OD) and identify predictors within a large multi-institutional cohort of refractory CRS patients. METHODS: Adult CRS patients enrolled in a prospective, multi-institutional, observational cohort study were assessed for baseline CRS-OD using a smell identification test (SIT) or brief SIT (bSIT). Four different ML methods were compared to traditional logistic regression for classification of CRS normosmics versus CRS-OD. RESULTS: Data were collected for 611 study participants who met inclusion criteria between 2011 April and 2015 July. Thirty-four percent of enrolled patients demonstrated olfactory loss on psychophysical testing. Differences between CRS normosmics and those with smell loss included objective disease measures (CT and endoscopy scores), age, sex, prior surgeries, socioeconomic status, steroid use, polyp presence, asthma, and aspirin sensitivity. Most ML methods performed favorably in terms of predictive ability. Top predictors include factors previously reported in the literature, as well as several socioeconomic factors. CONCLUSION: Olfactory dysfunction is a variable phenomenon in CRS patients. ML methods perform well compared to traditional logistic regression in classification of normosmia versus smell loss in CRS, and are able to include numerous risk factors into prediction models. Several actionable features were identified as risk factors for CRS-OD. These results suggest that ML methods may be useful for current understanding and future study of hyposmia secondary to sinonasal disease, the most common cause of persistent olfactory loss in the general population.
Authors: Edward El Rassi; Jess C Mace; Toby O Steele; Jeremiah A Alt; Zachary M Soler; Rongwei Fu; Timothy L Smith Journal: Int Forum Allergy Rhinol Date: 2015-12-01 Impact factor: 3.858
Authors: Sarah A Gitomer; Cynthia R Fountain; Todd T Kingdom; Anne E Getz; Stefan H Sillau; Rohit K Katial; Vijay R Ramakrishnan Journal: Otolaryngol Head Neck Surg Date: 2016-03-15 Impact factor: 3.497
Authors: Cezmi A Akdis; Claus Bachert; Cemal Cingi; Mark S Dykewicz; Peter W Hellings; Robert M Naclerio; Robert P Schleimer; Dennis Ledford Journal: J Allergy Clin Immunol Date: 2013-04-12 Impact factor: 10.793
Authors: Rodney J Schlosser; Timothy L Smith; Jess C Mace; Jeremiah Alt; Daniel M Beswick; Jose L Mattos; Spencer Payne; Vijay R Ramakrishnan; Zachary M Soler Journal: Int Forum Allergy Rhinol Date: 2020-01 Impact factor: 3.858