Francisco P M Oliveira1,2, Diogo Borges Faria3,4, Durval C Costa5, Miguel Castelo-Branco6, João Manuel R S Tavares7. 1. Institute for Nuclear Sciences Applied to Health (ICNAS), and Institute for Biomedical Imaging and Life Sciences (IBILI), Faculty of Medicine, University of Coimbra, Coimbra, Portugal. francisco.oliveira@fundacaochampalimaud.pt. 2. Champalimaud Centre for the Unknown, Champalimaud Foundation, Lisbon, Portugal. francisco.oliveira@fundacaochampalimaud.pt. 3. HPP Medicina Molecular, Grupo Lenitudes Saúde, Porto, Portugal. 4. School of Health Sciences, University of Aveiro, Aveiro, Portugal. 5. Champalimaud Centre for the Unknown, Champalimaud Foundation, Lisbon, Portugal. 6. Institute for Nuclear Sciences Applied to Health (ICNAS), and Institute for Biomedical Imaging and Life Sciences (IBILI), Faculty of Medicine, University of Coimbra, Coimbra, Portugal. 7. Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial, Departamento de Engenharia Mecânica, Faculdade de Engenharia, Universidade do Porto, Porto, Portugal.
Abstract
PURPOSE: This work aimed to assess the potential of a set of features extracted from [123I]FP-CIT SPECT brain images to be used in the computer-aided "in vivo" confirmation of dopaminergic degeneration and therefore to assist clinical decision to diagnose Parkinson's disease. METHODS: Seven features were computed from each brain hemisphere: five standard features related to uptake ratios on the striatum and two features related to the estimated volume and length of the striatal region with normal uptake. The features were tested on a dataset of 652 [123I]FP-CIT SPECT brain images from the Parkinson's Progression Markers Initiative. The discrimination capacities of each feature individually and groups of features were assessed using three different machine learning techniques: support vector machines (SVM), k-nearest neighbors and logistic regression. RESULTS: Cross-validation results based on SVM have shown that, individually, the features that generated the highest accuracies were the length of the striatal region (96.5%), the putaminal binding potential (95.4%) and the striatal binding potential (93.9%) with no statistically significant differences among them. The highest classification accuracy was obtained using all features simultaneously (accuracy 97.9%, sensitivity 98% and specificity 97.6%). Generally, slightly better results were obtained using the SVM with no statistically significant difference to the other classifiers for most of the features. CONCLUSIONS: The length of the striatal region uptake is clinically useful and highly valuable to confirm dopaminergic degeneration "in vivo" as an aid to the diagnosis of Parkinson's disease. It compares fairly well to the standard uptake ratio-based features, reaching, at least, similar accuracies and is easier to obtain automatically. Thus, we propose its day to day clinical use, jointly with the uptake ratio-based features, in the computer-aided diagnosis of dopaminergic degeneration in Parkinson's disease.
PURPOSE: This work aimed to assess the potential of a set of features extracted from [123I]FP-CIT SPECT brain images to be used in the computer-aided "in vivo" confirmation of dopaminergic degeneration and therefore to assist clinical decision to diagnose Parkinson's disease. METHODS: Seven features were computed from each brain hemisphere: five standard features related to uptake ratios on the striatum and two features related to the estimated volume and length of the striatal region with normal uptake. The features were tested on a dataset of 652 [123I]FP-CIT SPECT brain images from the Parkinson's Progression Markers Initiative. The discrimination capacities of each feature individually and groups of features were assessed using three different machine learning techniques: support vector machines (SVM), k-nearest neighbors and logistic regression. RESULTS: Cross-validation results based on SVM have shown that, individually, the features that generated the highest accuracies were the length of the striatal region (96.5%), the putaminal binding potential (95.4%) and the striatal binding potential (93.9%) with no statistically significant differences among them. The highest classification accuracy was obtained using all features simultaneously (accuracy 97.9%, sensitivity 98% and specificity 97.6%). Generally, slightly better results were obtained using the SVM with no statistically significant difference to the other classifiers for most of the features. CONCLUSIONS: The length of the striatal region uptake is clinically useful and highly valuable to confirm dopaminergic degeneration "in vivo" as an aid to the diagnosis of Parkinson's disease. It compares fairly well to the standard uptake ratio-based features, reaching, at least, similar accuracies and is easier to obtain automatically. Thus, we propose its day to day clinical use, jointly with the uptake ratio-based features, in the computer-aided diagnosis of dopaminergic degeneration in Parkinson's disease.
Authors: Peter S Jensen; Morten Ziebell; Glenna Skouboe; Usman Khalid; Robin de Nijs; Gerda Thomsen; Gitte M Knudsen; Claus Svarer Journal: J Nucl Med Technol Date: 2011-11-09
Authors: Robert B Innis; Vincent J Cunningham; Jacques Delforge; Masahiro Fujita; Albert Gjedde; Roger N Gunn; James Holden; Sylvain Houle; Sung-Cheng Huang; Masanori Ichise; Hidehiro Iida; Hiroshi Ito; Yuichi Kimura; Robert A Koeppe; Gitte M Knudsen; Juhani Knuuti; Adriaan A Lammertsma; Marc Laruelle; Jean Logan; Ralph Paul Maguire; Mark A Mintun; Evan D Morris; Ramin Parsey; Julie C Price; Mark Slifstein; Vesna Sossi; Tetsuya Suhara; John R Votaw; Dean F Wong; Richard E Carson Journal: J Cereb Blood Flow Metab Date: 2007-05-09 Impact factor: 6.200
Authors: Rosemary J Morton; Matthew J Guy; Ralf Clauss; Paul J Hinton; Craig A Marshall; Elizabeth A Clarke Journal: Nucl Med Commun Date: 2005-12 Impact factor: 1.690
Authors: I George Zubal; Michele Early; Olive Yuan; Danna Jennings; Kenneth Marek; John P Seibyl Journal: J Nucl Med Date: 2007-05-15 Impact factor: 10.057
Authors: Markus Wenzel; Fausto Milletari; Julia Krüger; Catharina Lange; Michael Schenk; Ivayla Apostolova; Susanne Klutmann; Marcus Ehrenburg; Ralph Buchert Journal: Eur J Nucl Med Mol Imaging Date: 2019-08-31 Impact factor: 9.236
Authors: Helen Schmitz-Steinkrüger; Catharina Lange; Ivayla Apostolova; Franziska L Mathies; Lars Frings; Susanne Klutmann; Sabine Hellwig; Philipp T Meyer; Ralph Buchert Journal: Eur J Nucl Med Mol Imaging Date: 2020-10-31 Impact factor: 9.236
Authors: Francisco Oliveira; Antoine Leuzy; João Castelhano; Konstantinos Chiotis; Steen Gregers Hasselbalch; Juha Rinne; Alexandre Mendonça; Markus Otto; Alberto Lleó; Isabel Santana; Jarkko Johansson; Sarah Anderl-Straub; Christine Arnim; Ambros Beer; Rafael Blesa; Juan Fortea; Herukka Sanna-Kaisa; Erik Portelius; Josef Pannee; Henrik Zetterberg; Kaj Blennow; Ana P Moreira; Antero Abrunhosa; Agneta Nordberg; Miguel Castelo-Branco Journal: Neuroimage Clin Date: 2018-08-19 Impact factor: 4.881