Parkinson's disease dysgraphia (PDYS), one of the earliest signs of Parkinson's disease (PD), has been researched as a promising biomarker of PD and as the target of a noninvasive and inexpensive approach to monitoring the progress of the disease. However, although several approaches to supportive PDYS diagnosis have been proposed (mainly based on handcrafted features (HF) extracted from online handwriting or the utilization of deep neural networks), it remains unclear which approach provides the highest discrimination power and how these approaches can be transferred between different datasets and languages. This study aims to compare classification performance based on two types of features: features automatically extracted by a pretrained convolutional neural network (CNN) and HF designed by human experts. Both approaches are evaluated on a multilingual dataset collected from 143 PD patients and 151 healthy controls in the Czech Republic, United States, Colombia, and Hungary. The subjects performed the spiral drawing task (SDT; a language-independent task) and the sentence writing task (SWT; a language-dependent task). Models based on logistic regression and gradient boosting were trained in several scenarios, specifically single language (SL), leave one language out (LOLO), and all languages combined (ALC). We found that the HF slightly outperformed the CNN-extracted features in all considered evaluation scenarios for the SWT. In detail, the following balanced accuracy (BACC) scores were achieved: SL-0.65 (HF), 0.58 (CNN); LOLO-0.65 (HF), 0.57 (CNN); and ALC-0.69 (HF), 0.66 (CNN). However, in the case of the SDT, features extracted by a CNN provided competitive results: SL-0.66 (HF), 0.62 (CNN); LOLO-0.56 (HF), 0.54 (CNN); and ALC-0.60 (HF), 0.60 (CNN). In summary, regarding the SWT, the HF outperformed the CNN-extracted features over 6% (mean BACC of 0.66 for HF, and 0.60 for CNN). In the case of the SDT, both feature sets provided almost identical classification performance (mean BACC of 0.60 for HF, and 0.58 for CNN).

1st Department of Neurology Faculty of Medicine and St Anne's University Hospital Masaryk University Brno Czechia

Applied Neuroscience Research Group Central European Institute of Technology CEITEC Masaryk University Brno Czechia

Department of Neurology College of Medicine University of Arizona Tucson AZ United States

Department of Neurology University of Szeged Szeged Hungary

Department of Telecommunications Brno University of Technology Brno Czechia

Escola Superior Politecnica Tecnocampus Mataró Spain

Faculty of Engineering Universidad Católica de Oriente Rionegro Colombia

Faculty of Engineering Universidad de Antioquia UdeA Medellín Colombia

Intelligent Information Systems Laboratory Faculty of Electrical Engineering and Informatics Technical University of Kosice Košice Slovakia

Pattern Recognition Lab Friedrich Alexander Universität Erlangen Germany

See more in PubMed

Agrell B., Dehlin O. (1998). The clock-drawing test. Age Ageing 27, 399–403. 10.1093/ageing/27.3.399 PubMed DOI

Alonso-Martinez C., Faundez-Zanuy M., Mekyska J. (2017). A comparative study of in-air trajectories at short and long distances in online handwriting. Cogn. Comput. 9, 712–720. 10.1007/s12559-017-9501-5 PubMed DOI PMC

Ammour A., Aouraghe I., Khaissidi G., Mrabti M., Aboulem G., Belahsen F. (2020). A new semi-supervised approach for characterizing the Arabic on-line handwriting of Parkinson's disease patients. Comput. Methods Prog. Biomed. 183:104979. 10.1016/j.cmpb.2019.07.007 PubMed DOI

Aouraghe I., Alae A., Ghizlane K., Mrabti M., Aboulem G., Faouzi B. (2020). A novel approach combining temporal and spectral features of Arabic online handwriting for Parkinson's disease prediction. J. Neurosci. Methods 339:108727. 10.1016/j.jneumeth.2020.108727 PubMed DOI

Berardelli A., Rothwell J., Thompson P., Hallett M. (2001). Pathophysiology of Bradykinesia in Parkinson's disease. Brain 124, 2131–2146. 10.1093/brain/124.11.2131 PubMed DOI

Bidet-Ildei C., Pollak P., Kandel S., Fraix V., Orliaguet J.-P. (2011). Handwriting in patients with Parkinson disease: effect of l-DOPA and stimulation of the sub-thalamic nucleus on motor anticipation. Hum. Mov. Sci. 30, 783–791. 10.1016/j.humov.2010.08.008 PubMed DOI

Brabenec L., Mekyska J., Galaz Z., Rektorova I. (2017). Speech disorders in Parkinson's disease: early diagnostics and effects of medication and brain stimulation. J. Neural Transm. 124, 303–334. 10.1007/s00702-017-1676-0 PubMed DOI

Brodal P. (2003). The Central Nervous System: Structure and Function, 3 Edn. Oxford: Oxford University Press.

Cascarano G. D., Loconsole C., Brunetti A., Lattarulo A., Buongiorno D., Losavio G., et al. . (2019). Biometric handwriting analysis to support Parkinson's disease assessment and grading. BMC Med. Inform. Decis. Making 19, 1–11. 10.1186/s12911-019-0989-3 PubMed DOI PMC

Castrillon R., Acien A., Orozco-Arroyave J. R., Morales A., Vargas J., Vera-Rodríguez R., et al. . (2019). “Characterization of the handwriting skills as a biomarker for Parkinson's disease,” in 2019 14th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2019) (Lille), 1–5. 10.1109/FG.2019.8756508 DOI

Chen T., Guestrin C. (2016). “XGBoost: a scalable tree boosting system,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. 10.1145/2939672.2939785 DOI

Collins G. S., Moons K. G. M. (2019). Reporting of artificial intelligence prediction models. Lancet 393, 1577–1579. 10.1016/S0140-6736(19)30037-6 PubMed DOI

De Stefano C., Fontanella F., Impedovo D., Pirlo G., di Freca A. S. (2019). Handwriting analysis to support neurodegenerative diseases diagnosis: a review. Pattern Recogn. Lett. 121, 37–45. 10.1016/j.patrec.2018.05.013 DOI

Diaz M., Ferrer M. A., Impedovo D., Pirlo G., Vessio G. (2019). Dynamically enhanced static handwriting representation for Parkinson's disease detection. Pattern Recogn. Lett. 128, 204–210. 10.1016/j.patrec.2019.08.018 DOI

Diaz M., Moetesum M., Siddiqi I., Vessio G. (2021). Sequence-based dynamic handwriting analysis for Parkinson's disease detection with one-dimensional convolutions and BiGRUs. Expert Syst. Appl. 168:114405. 10.1016/j.eswa.2020.114405 DOI

Dickson D. W. (2012). Parkinson's disease and Parkinsonism: neuropathology. Cold Spring Harb. Perspect. Med. 2:a009258. 10.1101/cshperspect.a009258 PubMed DOI PMC

Dounskaia N., Van Gemmert A., Stelmach G. (2000). Interjoint coordination during handwriting-like movements. Exp. Brain Res. 135, 127–140. 10.1007/s002210000495 PubMed DOI

Drotar P., Mekyska J., Rektorova I., Masarova L., Smekal Z., Faundez-Zanuy M. (2014). Analysis of in-air movement in handwriting: a novel marker for Parkinson's disease. Comput. Methods Prog. Biomed. 117, 405–411. 10.1016/j.cmpb.2014.08.007 PubMed DOI

Drotar P., Mekyska J., Rektorova I., Masarova L., Smekal Z., Faundez-Zanuy M. (2016). Evaluation of handwriting kinematics and pressure for differential diagnosis of Parkinson's disease. Artif. Intell. Med. 67, 39–46. 10.1016/j.artmed.2016.01.004 PubMed DOI

Drotar P., Mekyska J., Rektorova I., Masarova L., Smekal Z., Zanuy M. F. (2015). Decision support framework for Parkinson's disease based on novel handwriting markers. IEEE Trans. Neural Syst. Rehabil. Eng. 23, 508–516. 10.1109/TNSRE.2014.2359997 PubMed DOI

Fahn S., Elton R. L. (1987). UPDRS Development Committee (1987) Unified Parkinson's Disease Rating Scale. Recent Developments in Parkinson's Disease. Florham Park, NJ: Macmillan.

Forno L. S. (1996). Neuropathology of Parkinson's disease. J. Neuropathol. Exp. Neurol. 55, 259–272. 10.1097/00005072-199603000-00001 PubMed DOI

Galaz Z., Mucha J., Zvoncak V., Mekyska J., Smekal Z., Safarova K., et al. . (2020). Advanced parametrization of graphomotor difficulties in school-aged children. IEEE Access 8, 112883–112897. 10.1109/ACCESS.2020.3003214 DOI

Garre-Olmo J., Faúndez-Zanuy M., López-de Ipi na K., Calvó-Perxas L., Turró-Garriga O. (2017). Kinematic and pressure features of handwriting and drawing: preliminary results between patients with mild cognitive impairment, Alzheimer disease and healthy controls. Curr. Alzheimer Res. 14, 960–968. 10.2174/1567205014666170309120708 PubMed DOI PMC

Gazda M., Hireš M., Drotár P. (2021). Multiple-fine-tuned convolutional neural networks for Parkinson's disease diagnosis from offline handwriting. IEEE Trans. Syst. Man Cybernet. 1–12. 10.1109/TSMC.2020.3048892 DOI

Gil-Martin M., Montero J. M., San-Segundo R. (2019). Parkinsons disease detection from drawing movements using convolutional neural networks. Electronics 8:907. 10.3390/electronics8080907 DOI

Goetz C. G., Poewe W., Rascol O., Sampaio C., Stebbins G. T., Counsell C., et al. . (2004). Movement disorder society task force report on the Hoehn and Yahr staging scale: status and recommendations. Mov. Disord. 19, 1020–1028. 10.1002/mds.20213 PubMed DOI

Goetz C. G., Tilley B. C., Shaftman S. R., Stebbins G. T., Fahn S., Martinez-Martin P., et al. . (2008). Movement disorder society-sponsored revision of the unified Parkinson's disease rating scale (MDS-UPDRS): scale presentation and clinimetric testing results. Mov. Disord. 23, 2129–2170. 10.1002/mds.22340 PubMed DOI

Hagerty J. R., Stanley R. J., Almubarak H. A., Lama N., Kasmi R., Guo P., et al. . (2019). Deep learning and handcrafted method fusion: Higher diagnostic accuracy for melanoma dermoscopy images. IEEE J. Biomed. Health Informatics 23, 1385–1391. 10.1109/JBHI.2019.2891049 PubMed DOI

Heinzel S., Berg D., Gasser T., Chen H., Yao C., Postuma R. B., et al. . (2019). Update of the MDS research criteria for prodromal Parkinson's disease. Mov. Disord. 34, 1464–1470. 10.1002/mds.27802 PubMed DOI

Horak F. B., Dimitrova D., Nutt J. G. (2005). Direction-specific postural instability in subjects with Parkinson's disease. Exp. Neurol. 193, 504–521. 10.1016/j.expneurol.2004.12.008 PubMed DOI

Hornykiewicz O. (1998). Biochemical aspects of Parkinson's disease. Neurology 51(2 Suppl. 2), S2–S9. 10.1212/WNL.51.2_Suppl_2.S2 PubMed DOI

Hughes A., Daniel S., Lees A. (1993). The clinical features of Parkinson's disease in 100 histologically proven cases. Adv. Neurol. 60:595. PubMed

Impedovo D. (2019). Velocity-based signal features for the assessment of Parkinsonian handwriting. IEEE Signal Process. Lett. 26, 632–636. 10.1109/LSP.2019.2902936 DOI

Impedovo D., Pirlo G. (2018). Dynamic handwriting analysis for the assessment of neurodegenerative diseases: a pattern recognition perspective. IEEE Rev. Biomed. Eng. 12, 209–220. 10.1109/RBME.2018.2840679 PubMed DOI

Impedovo D., Pirlo G., Vessio G. (2018). Dynamic handwriting analysis for supporting earlier Parkinson's disease diagnosis. Information 9:247. 10.3390/info9100247 DOI

Jerkovic V. M., Kojic V., Miskovic N. D., Djukic T., Kostic V. S., Popovic M. B. (2019). Analysis of on-surface and in-air movement in handwriting of subjects with Parkinson's disease and atypical parkinsonism. Biomed. Eng. 64, 187–194. 10.1515/bmt-2017-0148 PubMed DOI

Kamble M., Shrivastava P., Jain M. (2021). Digitized spiral drawing classification for Parkinson's disease diagnosis. Measurement 16:100047. 10.1016/j.measen.2021.100047 DOI

Kamran I., Naz S., Razzak I., Imran M. (2021). Handwriting dynamics assessment using deep neural network for early identification of Parkinson's disease. Fut. Gener. Comput. Syst. 117, 234–244. 10.1016/j.future.2020.11.020 PubMed DOI

Kushki A., Schwellnus H., Ilyas F., Chau T. (2011). Changes in kinetics and kinematics of handwriting during a prolonged writing task in children with and without dysgraphia. Res. Dev. Disabil. 32, 1058–1064. 10.1016/j.ridd.2011.01.026 PubMed DOI

Lee J. Y., Kim J. W., Lee W. Y., Kim J. M., Ahn T. B., Kim H. J., et al. . (2010). Daily dose of dopaminergic medications in Parkinson's disease: clinical correlates and a posteriori equation. Neurol. Asia 15, 137–143.

Letanneux A., Danna J., Velay J.-L., Viallet F., Pinto S. (2014). From micrographia to Parkinson's disease dysgraphia. Mov. Disord. 29, 1467–1475. 10.1002/mds.25990 PubMed DOI

McLennan J., Nakano K., Tyler H., Schwab R. (1972). Micrographia in Parkinson's disease. J. Neurol. Sci. 15, 141–152. 10.1016/0022-510X(72)90002-0 PubMed DOI

Minaee S., Kafieh R., Sonka M., Yazdani S., Jamalipour Soufi G. (2020). Deep-COVID: predicting covid-19 from chest x-ray images using deep transfer learning. Med. Image Anal. 65:101794. 10.1016/j.media.2020.101794 PubMed DOI PMC

Moetesum M., Siddiqi I., Vincent N., Cloppet F. (2019). Assessing visual attributes of handwriting for prediction of neurological disorders-a case study on Parkinson's disease. Pattern Recogn. Lett. 121, 19–27. 10.1016/j.patrec.2018.04.008 DOI

Montavon G., Lapuschkin S., Binder A., Samek W., Müller K.-R. (2017). Explaining nonlinear classification decisions with deep Taylor decomposition. Pattern Recogn. 65, 211–222. 10.1016/j.patcog.2016.11.008 DOI

Morberg B. M., Malling A. S., Jensen B. R., Gredal O., Wermuth L., Bech P. (2018). The Hawthorne effect as a pre-placebo expectation in Parkinson's disease patients participating in a randomized placebo-controlled clinical study. Nordic J. Psychiatry 72, 442–446. 10.1080/08039488.2018.1468480 PubMed DOI

Mucha J., Mekyska J., Faundez-Zanuy M., Lopez-de Ipina K., Zvoncak V., Galaz Z., et al. . (2018a). “Advanced Parkinson's disease dysgraphia analysis based on fractional derivatives of online handwriting,” in 10th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). 10.1109/ICUMT.2018.8631265 DOI

Mucha J., Mekyska J., Galaz Z., Faundez-Zanuy M., Lopez-de Ipina K., Zvoncak V., et al. . (2018b). Identification and monitoring of Parkinson's disease dysgraphia based on fractional-order derivatives of online handwriting. Appl. Sci. 8:2566. 10.3390/app8122566 DOI

Mucha J., Mekyska J., Galaz Z., Faundez-Zanuy M., Zvoncak V., Safarova K., et al. . (2020). Analysis of various fractional order derivatives approaches in assessment of graphomotor difficulties. IEEE Access 8, 218234–218244. 10.1109/ACCESS.2020.3042591 DOI

Mucha J., Mekyska J., Galáž Z., Zvonč'ak V., Kiska T., Smékal Z. (2019). “Advanced analysis of online handwriting in a multilingual cohort of patients with Parkinson's disease,” in Advances in Signal Processing and Artificial Intelligence: Proceedings of the 1st International Conference on Advances in Signal Processing and Artificial Intelligence (International Frequency Sensor Association (IFSA: ) Publishing, S. L.), 144–147.

Nolazco-Flores J. A., Faundez-Zanuy M., De La Cueva V., Mekyska J. (2021). Exploiting spectral and cepstral handwriting features on diagnosing Parkinson's disease. IEEE Access 9, 141599–141610. 10.1109/ACCESS.2021.3119035 DOI

O'Reilly C. and Plamondon, R (2009). Development of a sigma-lognormal representation for on-line signatures. Pattern Recogn. 42, 3324–3337. 10.1016/j.patcog.2008.10.017 DOI

Pereira C. R., Weber S. A. T., Hook C., Rosa G. H., Papa J. P. (2016). “Deep learning-aided Parkinson's disease diagnosis from handwritten dynamics,” in 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 340–346. 10.1109/SIBGRAPI.2016.054 DOI

Pinto S., Velay J.-L. (2015). Handwriting as a marker for PD progression: a shift in paradigm. Neurodegener. Dis. Manage. 5, 367–369. 10.2217/nmt.15.29 PubMed DOI

Ponsen M. M., Daffertshofer A., Wolters E. C., Beek P. J., Berendse H. W. (2008). Impairment of complex upper limb motor function in de novo Parkinson's disease. Parkinsonism Relat. Disord. 14, 199–204. 10.1016/j.parkreldis.2007.07.019 PubMed DOI

Postuma R. B., Berg D., Adler C. H., Bloem B. R., Chan P., Deuschl G., et al. . (2016). The new definition and diagnostic criteria of Parkinson's disease. Lancet Neurol. 15, 546–548. 10.1016/S1474-4422(16)00116-2 PubMed DOI

Pullman S. (1998). Spiral analysis: a new technique for measuring tremor with a digitizing tablet. Mov. Disord. 13, 85–89. 10.1002/mds.870131315 PubMed DOI

Rios-Urrego C., Vásquez-Correa J., Vargas-Bonilla J., Nöth E., Lopera F., Orozco-Arroyave J. (2019). Analysis and evaluation of handwriting in patients with Parkinson's disease using kinematic, geometrical, and non-linear features. Comput. Methods Prog. Biomed. 173, 43–52. 10.1016/j.cmpb.2019.03.005 PubMed DOI

Rosenblum S., Samuel M., Zlotnik S., Erikh I., Schlesinger I. (2013). Handwriting as an objective tool for Parkinson's disease diagnosis. J. Neurol. 260, 2357–2361. 10.1007/s00415-013-6996-x PubMed DOI

Russakovsky O., Deng J., Su H., Krause J., Satheesh S., Ma S., et al. . (2015). ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115, 211–252. 10.1007/s11263-015-0816-y DOI

San Luciano M., Wang C., Ortega R. A., Yu Q., Boschung S., Soto-Valencia J., et al. . (2016). Digitized spiral drawing: a possible biomarker for early Parkinson's disease. PLoS ONE 11:e0162799. 10.1371/journal.pone.0162799 PubMed DOI PMC

Simonyan K., Zisserman A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv[Preprint].arXiv:1409.1556.

Thomas M., Lenka A., Kumar Pal P. (2017). Handwriting analysis in Parkinson's disease: current status and future directions. Mov. Disord. Clin. Pract. 4, 806–818. 10.1002/mdc3.12552 PubMed DOI PMC

Van Galen G. P. (1991). Handwriting: Issues for a psychomotor theory. Hum. Mov. Sci. 10, 165–191. 10.1016/0167-9457(91)90003-G PubMed DOI

Vásquez-Correa J. C., Arias-Vergara T., Orozco-Arroyave J. R., Eskofier B., Klucken J., Nöth E. (2019). Multimodal assessment of Parkinson's disease: a deep learning approach. IEEE J. Biomed. Health Inform. 23, 1618–1630. 10.1109/JBHI.2018.2866873 PubMed DOI

Vessio G. (2019). Dynamic handwriting analysis for neurodegenerative disease assessment: a literary review. Appl. Sci. 9:4666. 10.3390/app9214666 DOI

Zham P., Kumar D., Viswanthan R., Wong K., Nagao K. J., Arjunan S. P., et al. . (2019). Effect of levodopa on handwriting tasks of different complexity in Parkinson's disease: a kinematic study. J. Neurol. 266, 1376–1382. 10.1007/s00415-019-09268-2 PubMed DOI

Zham P., Kumar D. K., Dabnichki P., Poosapadi Arjunan S., Raghav S. (2017). Distinguishing different stages of Parkinson's disease using composite index of speed and pen-pressure of sketching a spiral. Front. Neurol. 8:435. 10.3389/fneur.2017.00435 PubMed DOI PMC