Oliver Mothes1, Luise Modersohn1, Gerd Fabian Volk2,3, Carsten Klingner3,4, Otto W Witte3,4, Peter Schlattmann5, Joachim Denzler1, Orlando Guntinas-Lichius6,7. 1. Department of Computer Science, Friedrich Schiller University Jena, Jena, Germany. 2. Department of Otorhinolaryngology, Jena University Hospital, Am Klinikum 1, 07747, Jena, Germany. 3. Facial Nerve Center, Jena University Hospital, Jena, Germany. 4. Department of Neurology, Jena University Hospital, Jena, Germany. 5. Department of Medical Statistics, Computer Sciences and Data Science, Jena University Hospital, Jena, Germany. 6. Department of Otorhinolaryngology, Jena University Hospital, Am Klinikum 1, 07747, Jena, Germany. orlando.guntinas@med.uni-jena.de. 7. Facial Nerve Center, Jena University Hospital, Jena, Germany. orlando.guntinas@med.uni-jena.de.
Abstract
PURPOSE: An automated, objective, fast and simple classification system for the grading of facial palsy (FP) is lacking. METHODS: An observational single center study was performed. 4572 photographs of 233 patients with unilateral peripheral FP were subjectively rated and automatically analyzed applying a machine learning approach including Supervised Descent Method. This allowed an automated grading of all photographs according to House-Brackmann grading scale (HB), Sunnybrook grading system (SB), and Stennert index (SI). RESULTS: Median time to first assessment was 6 days after onset. At first examination, the median objective HB, total SB, and total SI were grade 3, 45, and 5, respectively. The best correlation between subjective and objective grading was seen for SB and SI movement score (r = 0.746; r = 0.732, respectively). No agreement was found between subjective and objective HB grading [Test for symmetry 80.61, df = 15, p < 0.001, weighted kappa = - 0.0105; 95% confidence interval (CI) = - 0.0542 to 0.0331; p = 0.6541]. Also no agreement was found between subjective and objective total SI (test for symmetry 166.37, df = 55, p < 0.001) although there was a nonzero weighted kappa = 0.2670; CI 0.2154-0.3186; p < 0.0001). Based on a multinomial logistic regression the probability for higher scores was higher for subjective compared to objective SI (OR 1.608; CI 1.202-2.150; p = 0.0014). The best agreement was seen between subjective and objective SB (ICC = 0.34645). CONCLUSIONS: Automated Sunnybrook grading delivered with fair agreement fast and objective global and regional data on facial motor function for use in clinical routine and clinical trials.
PURPOSE: An automated, objective, fast and simple classification system for the grading of facial palsy (FP) is lacking. METHODS: An observational single center study was performed. 4572 photographs of 233 patients with unilateral peripheral FP were subjectively rated and automatically analyzed applying a machine learning approach including Supervised Descent Method. This allowed an automated grading of all photographs according to House-Brackmann grading scale (HB), Sunnybrook grading system (SB), and Stennert index (SI). RESULTS: Median time to first assessment was 6 days after onset. At first examination, the median objective HB, total SB, and total SI were grade 3, 45, and 5, respectively. The best correlation between subjective and objective grading was seen for SB and SI movement score (r = 0.746; r = 0.732, respectively). No agreement was found between subjective and objective HB grading [Test for symmetry 80.61, df = 15, p < 0.001, weighted kappa = - 0.0105; 95% confidence interval (CI) = - 0.0542 to 0.0331; p = 0.6541]. Also no agreement was found between subjective and objective total SI (test for symmetry 166.37, df = 55, p < 0.001) although there was a nonzero weighted kappa = 0.2670; CI 0.2154-0.3186; p < 0.0001). Based on a multinomial logistic regression the probability for higher scores was higher for subjective compared to objective SI (OR 1.608; CI 1.202-2.150; p = 0.0014). The best agreement was seen between subjective and objective SB (ICC = 0.34645). CONCLUSIONS: Automated Sunnybrook grading delivered with fair agreement fast and objective global and regional data on facial motor function for use in clinical routine and clinical trials.
Entities:
Keywords:
Active appearance model; Diagnostics; Facial image analysis; Facial paralysis; Grading; Random decision forest
Authors: Robert A Gaudin; Mara Robinson; Caroline A Banks; Jennifer Baiungo; Nate Jowett; Tessa A Hadlock Journal: JAMA Facial Plast Surg Date: 2016-07-01 Impact factor: 4.611
Authors: Adel Y Fattah; Anthony D R Gurusinghe; Javier Gavilan; Tessa A Hadlock; Jeff R Marcus; Henri Marres; Charles C Nduka; William H Slattery; Alison K Snyder-Warwick Journal: Plast Reconstr Surg Date: 2015-02 Impact factor: 4.730
Authors: Adel Y Fattah; Javier Gavilan; Tessa A Hadlock; Jeffrey R Marcus; Henri Marres; Charles Nduka; William H Slattery; Alison K Snyder-Warwick Journal: Laryngoscope Date: 2014-02-27 Impact factor: 3.325
Authors: Anna-Maria Kuttenreich; Gerd Fabian Volk; Orlando Guntinas-Lichius; Harry von Piekartz; Stefan Heim Journal: Diagnostics (Basel) Date: 2022-05-04
Authors: Gerd F Volk; Benjamin Roediger; Katharina Geißler; Anna-Maria Kuttenreich; Carsten M Klingner; Christian Dobel; Orlando Guntinas-Lichius Journal: Front Rehabil Sci Date: 2021-10-14
Authors: Leonard Knoedler; Helena Baecher; Martin Kauke-Navarro; Lukas Prantl; Hans-Günther Machens; Philipp Scheuermann; Christoph Palm; Raphael Baumann; Andreas Kehrer; Adriana C Panayi; Samuel Knoedler Journal: J Clin Med Date: 2022-08-25 Impact factor: 4.964
Authors: Ala Szczepura; Nikki Holliday; Catriona Neville; Karen Johnson; Amir Jahan Khan Khan; Samuel W Oxford; Charles Nduka Journal: J Med Internet Res Date: 2020-10-05 Impact factor: 5.428