https://doi.org/10.31449/inf.v44i2.3177 Informatica 44 (2020) 285–286 285 Artificial Intelligence Methods for Modelling Tremor Mechanisms Vida Groznik University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies, Koper, Slovenia University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, Slovenia E-mail: vida.groznik@famnit.upr.si Thesis summary Keywords: tremor, attribute visualisation, digital spirography Received: May 28, 2020 The paper summarises a Doctoral Thesis in which we focus on two main goals: (1) building models for differentiation between three most common tremors: Parkinsonian, essential and mixed type tremor and (2) development of a novel method for attribute visualisation on series. Povzetek: ˇ Clanek povzema vsebino doktorske disertacije, v kateri se osredotoˇ camo na dva glavna cilja: (1) gradnja modelov za razloˇ cevanje med tremi najpogostejšimi tremorji: parkinsonskim, esencialnim in mešanim tipom tremorja ter (2) razvoj nove metode za vizualizacijo atributov na vrstah. 1 Introduction Tremor is an involuntary movement of the body and is one of the most common movement disorders. It is primarily associated with various diseases of the nervous system, in- cluding Parkinson’s disease. Since there are more than 20 different types of tremors, differentiation between them is important from the treatment point of view. Spirography is a diagnostic method where the subject’s task is to draw a left-twisted spiral while the doctor ob- serves the process of drawing (speed, hesitation, etc.) and the final drawing. With the development of tablets digi- talised spirography emerged, making it possible to store the course of spiral drawing and the analysis of the acquired time series. In order to increase confidence in such a system, we would need to provide an explanation of the results to doc- tors. One option is to visualize the anomalies and results onto the drawn spirals. These visualizations must make sense to physicians and, above all, they must be consistent with their medical knowledge of the domain. This paper summarises a Doctoral Thesis [1] which tries to address the need for automatic differentiation of tremors and visualisation of the decisions such system would give. 2 Diagnostic models for tremor differentiation In the thesis, we focus on differentiation between three of the most common tremors: Parkinsonian, essential and mixed type of tremor. For the purpose of building the di- agnostic models, we used the digitalised spirography for collecting the data needed. The first diagnostic model distinguishes between the three tremors, based on clinical examination data, family history and digital spirography. The process of building a model was carried out using argument-based machine learning technique which enabled us to build a decision model through the process of knowledge elicitation from the domain expert (a neurologist). The obtained model consists of thirteen rules that are medically sensible. The process of knowledge elicitation itself contributed to the higher classification accuracy of the final model in com- parison with the initial one [2, 5]. In the first diagnostic model, attributes derived from the spirography were included in more than half of the rules. This motivated us to build a model based solely on the digital spirography data. For the needs of constructing an understandable model, we first built several attributes which represented domain medical knowledge. We have built more than 500 different attributes which were used in a logistic regression to construct the final diagnostic model. The model is able to distinguish subjects with tremors from those without tremors with 90% classification accuracy. The final diagnostic model is built into the freely available PARKINSONCHECK mobile application [6]. 3 Method for attribute visualisation During the process of attribute construction, we wanted to know what our attributes were detecting. Thus, we have de- veloped a method for attribute visualisation on series. The method not only helped us with attribute construction, but it is also useful for visual interpretation of the diagnostic model’s decisions. The visualisation method and conse- quently the decision model were evaluated with the help of three independent neurology experts. The results show that 286 Informatica 44 (2020) 285–286 V . Groznik both the diagnostic model and the visualisation are mean- ingful and cover medical knowledge of the domain. Different visualisation approaches and their benefits have been published in several peer reviewed publications [3, 4, 7]. 4 Conclusion The Thesis [1] describes the development of different di- agnostic models for digitalised spirography systems. The emphasis is given to elicitation of expert’s knowledge and including that knowledge into the built-in attributes. To increase physicians’ confidence in such systems, a novel method for attribute visualisation has been proposed. The results were published in several peer-reviewed publica- tions [2, 3, 4, 5, 6, 7]. References [1] Groznik, V . (2018) Metode umetne inteligence za modeliranje mehanizmov tremorjev: doktorska dis- ertacija. Ljubljana. 139 pages, ilustr. http://eprints.fri.uni-lj.si/4134/. [2] Groznik, V ., Guid, M., Sadikov, A., Možina, M., Georgiev, D., Kragelj, V ., Ribariˇ c, S., Pirtošek, Z., and Bratko, I. (2013) Elicitation of neurological knowledge with argument-based machine learning. Artificial Intelligence in Medicine, 57:133–144. https://doi.org/10.1016/j.artmed.2012.08.003. [3] Groznik, V ., Sadikov, A., Možina, M., Žabkar, J., Georgiev, D., and Bratko, I. (2014) Attribute visual- isation for computer-aided diagnosis : a case study. In ICHI 2014: proceedings, 2014 IEEE International Conference on Healthcare Informatics, IEEE Com- puter Society, Conference Publishing Services, pp. 294–299. https://doi.org/10.1109/ICHI.2014.47. [4] Groznik, V ., Možina, M., Žabkar, J., Georgiev, D., Bratko, I., and Sadikov, A. (2015) Development, de- bugging, and assessment of PARKINSONCHECK at- tributes through visualisation. In A. Briassouli, J. Benois-Pineau, A. Hauptmann. Health monitoring and personalized feedback using multimedia data, Springer, pp. 47–71. https://doi.org/10.1007/978-3-319-17963-6_4. [5] Guid, M., Možina, M., Groznik, V ., Georgiev, D., Sadikov, A., Pirtošek, Z., and Bratko, I. (2012) ABML knowledge renement loop: a case study. In L. Chen, et al., editors, Foundations of intelligent sys- tems, vol. 7661 of Lecture notes in computer science, Lecture notes in artificial intelligence, Springer, pp. 41–50. https://doi.org/10.1007/978-3-642-34624-8_5. [6] Sadikov, A., Groznik, V ., Žabkar, J., Možina, M., Georgiev, D., Pirtošek, Z., and Bratko, I. (2014) PARKINSONCHECK smart phone app.. In T. Schaub, G. Friedrich, B. O’Sullivan, editors, ECAI 2014: pro- ceedings, Frontiers in artificial intelligence and appli- cations (Print), vol. 263, IOS Press, pp. 1213–1214. https://doi.org/10.3233/978-1-61499-419-0-1213. [7] Sadikov, A., Groznik, V ., Možina, M., Žabkar, J., Ny- holm, D., Memedi, M., Bratko, I., and Georgiev, D. (2017) Feasibility of spirography features for objec- tive assessment of motor function in Parkinson’s dis- ease. Artificial intelligence in medicine, vol. 66, pp. 54–62. https://doi.org/10.1016/j.artmed.2017.03.011.