Abstract
Background
Chemosensory loss is a symptom of Parkinson’s disease starting already at preclinical stages. Their appearance without an identifiable etiology therefore indicates a possible early symptom of Parkinson’s disease. Supervised machine-learning was used to identify parameters that predict Parkinson’s disease among patients having sought medical advice for chemosensory symptoms.
Methods
Olfactory, gustatory and demographic parameters were analyzed in 247 patients who had reported for chemosensory symptoms. Unsupervised machine-learning, implanted as so-called fast and frugal decision trees, was applied to map these parameters to a diagnosis of Parkinson’s disease queried for in median 9 years after the first interview.
Results
A symbolic hierarchical decision rule-based classifier was created that comprised d = 5 parameters, including scores in tests of odor discrimination, odor identification and olfactory thresholds, the age at which the chemosensory loss has been noticed, and a familial history of Parkinson’s disease. The rule set provided a cross-validated negative predictive performance of Parkinson’s disease of 94.1%; however, its balanced accuracy to predict the disease was only 58.9% while robustly above guessing.
Conclusions
Applying machine-learning techniques, a classifier was developed that took the shape of a set of six hierarchical rules with binary decisions about olfaction-related features or a familial burden of Parkinson’s disease. Its main clinical strength lies in the exclusion of the possibility of developing Parkinson’s disease in a patient with olfactory or gustatory loss.
Similar content being viewed by others
References
Ansari KA, Johnson A (1975) Olfactory function in patients with Parkinson's disease. J Chron Dis. 28:493–497
Korten J, Meulstee J (1980) Olfactory disturbances in parkinsonism. Clin Neurol Neurosurg 82:113–118
Ward CD, Hess WA, Calne DB (1983) Olfactory impairment in Parkinson's disease. Neurology 33:943–946
Serby M, Corwin J, Conrad P, Rotrosen J (1985) Olfactory dysfunction in Alzheimer's disease and Parkinson's disease [letter]. Am J Psychiatry 142(6):781–782
Doty RL, Deems D, Steller S (1988) Olfactory dysfunction in Parkinson's disease: a general deficit unrelated to neurologic signs, disease stage, or disease duration. Neurology 38:1237–1244
Haehner A, Masala C, Walter S, Reichmann H, Hummel T (2019) Incidence of Parkinson's disease in a large patient cohort with idiopathic smell and taste loss. J Neurol 266(2):339–345
Heilmann S, Hüttenbrink KB, Hummel T (2004) Local and systemic administration of corticosteroids in the treatment of olfactory loss. Am J Rhinol 18:29–33
Lötsch J, Hummel T (2019) A machine-learned analysis suggests non-redundant diagnostic information in olfactory subtests. IBRO Rep 6:64–73
Kobal G, Hummel T, Sekinger B, Barz S, Roscher S, Wolf SR (1996) "Sniffin' Sticks": screening of olfactory performance. Rhinology 34:222–226
Hummel T, Sekinger B, Wolf SR, Pauli E, Kobal G (1997) 'Sniffin' sticks': olfactory performance assessed by the combined testing of odor identification, odor discrimination and olfactory threshold. Chem Senses 22(1):39–52
Hummel T, Kobal G, Gudziol H, Mackay-Sim A (2007) Normative data for the "Sniffin' Sticks" including tests of odor identification, odor discrimination, and olfactory thresholds: an upgrade based on a group of more than 3,000 subjects. Eur Arch Otorhinolaryngol 264(3):237–243
Frasnelli J, Landis BN, Heilmann S, Hauswald B, Huttenbrink KB, Lacroix JS et al (2003) Clinical presentation of qualitative olfactory dysfunction. Eur Arch Otorhinolaryngol 11:11–13
Landis BN, Welge-Luessen A, Bramerson A, Bende M, Mueller CA, Nordin S et al (2009) "Taste Strips"—a rapid, lateralized, gustatory bedside identification test based on impregnated filter papers. J Neurol 256(2):242–248
Hummel T, Hummel C, Welge-Luessen A (2013) Assessment of olfaction and gustation. In: Hummel T, Welge-Luessen A (eds) Management of smell and taste disorders: a practical guide for clinicians. Thieme, Stuttgart, pp 58–75
R Development Core Team. R: a language and environment for statistical computing. 2008.
Cover T, Hart P (1967) Nearest neighbor pattern classification. IEEE Trans Inf Theor 13(1):21–27
Torgo L (2010) Data mining with R: learning with case studies. Chapman & Hall, London
Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1):41–55
Student (1908) The probable error of a mean. Biometrika 6(1):1–25
Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics 1:80–83
Pearson K (1900) On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Philos Mag Ser 5(50):157–175
Newell A, Simon HA (1976) Computer science as empirical inquiry: symbols and search. Commun ACM 19(3):113–126
Gigerenzer G, Todd PM (1999) Fast and frugal heuristics: the adaptive toolbox. Simple heuristics that make us smart. Evolution and cognition. Oxford University Press, New York, pp 3–34
Martignon L, Katsikopoulos KV, Woike JK (2008) Categorization with limited resources: a family of simple heuristics. J Math Psychol 52(6):352–361
Marewski JN, Gigerenzer G (2012) Heuristic decision making in medicine. Dialogues Clin Neurosci 14(1):77–89
Gigerenzer G, Czerlinski J, Martignon L. How good are fast and frugal heuristics? In: Shanteau J, Mellers BA, Schum DA, editors. Decision science and technology: reflections on the contributions of ward edwards. Boston: Springer US; 1999. p. 81–103.
Phillips N, Neth H, Woike J, Gaissmaer W. FFTrees: generate, visualise, and evaluate fast-and-frugal decision trees. R package version 1.4.0. 2018.
Phillips ND, Neth H, Woike JK, Gaissmaier W (2017) FFTrees: a toolbox to create, visualize, and evaluate fast-and-frugal decision trees. Judgm Decis Mak 12(4):344–368
Tillé Y, Matei A (2016) Sampling: survey sampling. R package version 2.8. https://CRAN.R-project.org/package=sampling
Ultsch A, Lötsch J (2015) Computed ABC analysis for rational selection of most informative variables in multivariate data. PLoS ONE 10(6):e0129767
Wild A (1997) Best practice in inventory management. Wiley, New York
Altman DG, Bland JM (1994) Diagnostic tests 2: predictive values. BMJ 309(6947):102
Altman DG, Bland JM (1994) Diagnostic tests. 1: sensitivity and specificity. BMJ 308(6943):1552
Martignon L, Vitouch O, Takezawa M, Forster M, Hardman D, Macchi L. Thinking: psychological perspective on reasoning, judgment, and decision making. In: Hardman D, Macchi L, editors. 2003. pp.189–211.
Martignon L, Katsikopoulos KV, Woike JK, editors. Categorization with limited resources: a family of simple heuristics. 2008.
Lotsch J, Kringel D, Hummel T. Machine learning in human olfactory research. Chem Senses. 2018.
Ross GW, Petrovitch H, Abbott RD, Tanner CM, Popper J, Masaki K et al (2008) Association of olfactory dysfunction with risk for future Parkinson's disease. Ann Neurol 63(2):167–173
Postuma RB, Gagnon JF, Vendette M, Desjardins C, Montplaisir JY (2011) Olfaction and color vision identify impending neurodegeneration in rapid eye movement sleep behavior disorder. Ann Neurol 69(5):811–818
Ponsen MM, Stoffers D, Twisk JW, Wolters E, Berendse HW (2009) Hyposmia and executive dysfunction as predictors of future Parkinson's disease: a prospective study. Mov Disord 24(7):1060–1065
Cox D (1972) Regression models and life tables. J R Stat Soc B 34:187–220
Lötsch J, Reichmann H, Hummel T (2008) Different odor tests contribute differently to the evaluation of olfactory loss. Chem Senses 33(1):17–21
Hedner M, Larsson M, Arnold N, Zucco GM, Hummel T (2010) Cognitive factors in odor detection, odor discrimination, and odor identification tasks. J Clin Exp Neuropsychol 32(10):1062–1067
Mahlknecht P, Iranzo A, Hogl B, Frauscher B, Muller C, Santamaria J et al (2015) Olfactory dysfunction predicts early transition to a Lewy body disease in idiopathic RBD. Neurology 84(7):654–658
Mogensen UB, Ishwaran H, Gerds TA (2012) Evaluating random forests for survival analysis using prediction error curves. J Stat Softw 50(11):1–23
Berg D, Postuma RB, Adler CH, Bloem BR, Chan P, Dubois B et al (2015) MDS research criteria for prodromal Parkinson's disease. Mov Disord 30(12):1600–1611
Postuma RB, Gagnon JF, Bertrand JA, Genier Marchand D, Montplaisir JY (2015) Parkinson risk in idiopathic REM sleep behavior disorder: preparing for neuroprotective trials. Neurology 84(11):1104–1113
Gustafsson H, Nordstrom A, Nordstrom P (2015) Depression and subsequent risk of Parkinson disease: a nationwide cohort study. Neurology 84(24):2422–2429
Abbott RD, Petrovitch H, White LR, Masaki KH, Tanner CM, Curb JD et al (2001) Frequency of bowel movements and the future risk of Parkinson's disease. Neurology 57(3):456–462
Schrag A, Horsfall L, Walters K, Noyce A, Petersen I (2015) Prediagnostic presentations of Parkinson's disease in primary care: a case-control study. Lancet Neurol 14(1):57–64
Postuma RB, Lang AE, Gagnon JF, Pelletier A, Montplaisir JY (2012) How does parkinsonism start? Prodromal parkinsonism motor changes in idiopathic REM sleep behaviour disorder. Brain 135(Pt 6):1860–1870
Iranzo A, Lomena F, Stockner H, Valldeoriola F, Vilaseca I, Salamero M et al (2010) Decreased striatal dopamine transporter uptake and substantia nigra hyperechogenicity as risk markers of synucleinopathy in patients with idiopathic rapid-eye-movement sleep behaviour disorder: a prospective study [corrected]. Lancet Neurol 9(11):1070–1077
Tremblay C, Durand Martel P, Frasnelli J (2017) Trigeminal system in Parkinson's disease: a potential avenue to detect Parkinson-specific olfactory dysfunction. Parkinsonism Relat Disord 44:85–90
Lötsch J, Lerch F, Djaldetti R, Tegeder I, Ultsch A (2018) Identification of disease-distinct complex biomarker patterns by means of unsupervised machine-learning using an interactive R toolbox (Umatrix). BMC Big Data Anal. https://doi.org/10.1186/s41044-018-0032-1
Pearson K (1904) Report on certain enteric fever inoculation statistics. BMJ 3:1243–1246
Ultsch A, editor. Pareto density estimation: a density estimation for knowledge discovery. Innovations in classification, data science, and information systems. In: Proceedings 27th annual conference of the German Classification Society (GfKL). Berlin: Springer; 2003.
Funding
This work has been funded by the Landesoffensive zur Entwicklung wissenschaftlich-ökonomischer Exzellenz (LOEWE), LOEWE-Zentrum für Translationale Medizin und Pharmakologie (JL). We also would like to thank the Deutsche Forschungsgemeinschaft for support (DFG HU 441/18-1 to TH). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
The authors have declared that no competing interests exist.
Ethical approval
The study has been approved by the Ethics Committee of the Medical Faculty of the Technical University of Dresden, Germany and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
Informed consent
All persons gave their informed consent prior to their inclusion in the study. Details that might disclose the identity of the subjects under study are omitted.
Rights and permissions
About this article
Cite this article
Lötsch, J., Haehner, A. & Hummel, T. Machine-learning-derived rules set excludes risk of Parkinson’s disease in patients with olfactory or gustatory symptoms with high accuracy. J Neurol 267, 469–478 (2020). https://doi.org/10.1007/s00415-019-09604-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00415-019-09604-6