Advanced electroencephalographic system-based neurophysiological methods to elucidate a mechanism of cognitive dysfunction associated with hepatolenticular degeneration
DOI: 10.32364/2587-6821-2024-8-10-1
E.V. Ovchinnikova1,2,3, N.A. Schneider2,4, A.A. Ovchinnikova1, R.F. Nasyrova2, S.A. Gulyaev5
1Far Eastern Federal University, Vladivostok, Russian Federation
2V.M. Bekhterev National Research Medical Center for Psychiatry and Neurology, St. Petersburg, Russian Federation
3Primorye Territory Clinical Hospital No. 1, Vladivostok, Russian Federation
4Prof. V.F. Voino-Yasenetsky Krasnoyarsk State Medical University, Krasnoyarsk, Russian Federation
5National Research Nuclear University MEPhI, Moscow, Russian Federation
Background: the information revolution induced the growth of neurodegenerative diseases and neurotic conditions associated with cognitive deficits. The studies were focused on genetic cerebral abnormalities. Hepatolenticular degeneration (HLD) or Wilson's disease whose pathogenesis is based on copper metabolism disorder caused by ATP7B gene mutations, is one of these conditions. Decreased excretion of copper leads to the metal accumulation in organs and the brain associated with various clinical manifestations. Motor defects of extrapyramidal origin and mental disorders are a core of HLD's neurological manifestations. Early manifestations of HLD-associated cognitive dysfunctions may lead to significantly limited information uptake and processing, in turn, reducing the quality of life of patients. Elucidation of the mechanisms of cognitive dysfunctions associated with the abnormality is of particular interest both for HLD diagnosis and development of advanced rehabilitation systems for patients with cognitive deficits.
Aim: to study the mechanisms of cognitive dysfunction development by advanced electroencephalographic (EEG) system-based neurophysiological methods in patients with a genetically confirmed diagnosis of HLD.
Patients and Methods: study object — 24 European subjects with an established neurological stage of genetically confirmed HLD. 24 healthy subjects of the same age and gender were enrolled in a control group. A neuropsychological examination was performed by MOCA and HADS. A neurophysiological EEG was carried out with a 52-channel computer-based bio-amplifier (sLORETA v20210701 software package).
Results: it was found that cognitive deficits are based on disorders of program development and activity planning associated with impaired mental alertness, as well as high levels of anxiety and depression whose severity does not depend on a type of the ATP7B gene mutation. Functional changes demonstrate an impact on cortical areas responsible for visual analysis and face identification. The indicators of 8 classes of EEG microstates show similarity in distribution profiles of bioelectric activity for all the ATP7B gene mutation types.
Conclusion: the findings indicate an impact on cortical structures of impulses arising from sources in damaged basal ganglia.
Keywords: hepatolenticular degeneration, Wilson's disease, electroencephalography, neurophysiology, cognitive dysfunction.
For citation: Ovchinnikova E.V., Schneider N.A., Ovchinnikova A.A., Nasyrova R.F., Gulyaev S.A. Advanced electroencephalographic system-based neurophysiological methods to elucidate a mechanism of cognitive dysfunction associated with hepatolenticular degeneration. Russian Medical Inquiry. 2024;8(10):547–553 (in Russ.). DOI: 10.32364/2587-6821-2024-8-10-1
Background
In recent decades, there has been an increase in neurodegenerative diseases and pathological conditions with cognitive deficits. The study of genetic brain disorders has become a particularly salient field of research in recent years. Hepatolenticular degeneration (HLD), otherwise known as Wilson's disease, is a condition that arises from a defect in copper metabolism. This defect is caused by mutations in the ATP7B gene. A decrease in the excretion of copper results in an accumulation of copper within various organs and the brain, which can lead to the development of diverse clinical manifestations. These manifestations may include liver failure, neurological disorders, and psychiatric conditions [1]. The predominant neurological manifestations of HLD encompass motor defects of extrapyramidal origin and psychiatric disorders. Cortical lesions are less prevalent and result in the extrapyramidal cortical variant of the disease. The manifestations of this condition include epileptic seizures and severe dementia, which has prompted research into the primary origin of the disease. The initial indications of cognitive dysfunction in HLD have the potential to result in considerable impediments to information acquisition and processing, consequently diminishing the quality of life experienced by the affected individual. Elucidation of the mechanisms underlying cognitive dysfunction in this disorder is of particular interest, not only for diagnostic purposes, but also for the development of contemporary rehabilitation tools for cognitive deficits [2].
To address this challenge, it is imperative to select technologies that facilitate volumetric evaluation of brain structures implicated in cognitive function execution. In clinical settings, their evaluation is predominantly informed by the assessment system initially developed by A.R. Luria in the mid-1950s. Despite its apparent simplicity, the system is designed to identify the resulting component of the cognitive process rather than to elucidate the mechanism of higher nervous function. Furthermore, the experiment demands a substantial time investment and is susceptible to the current state of the subject, which may include anxiety and depression.
The emergence of sophisticated cognitive functions is contingent upon rapid cerebral processes, necessitating the development of methodological approaches that are capable of recording them. Given that the activity of brain networks reflects the results of neurophysiological methods, especially those realized on the basis of modern electroencephalographic (EEG) systems, the improvement of these methods has become the basis of modern research.
In 1990, D. Lehmann pioneered a method for recording the activity of specific neuronal groups by analyzing the EEG signal. Through meticulous observation of individual variants in the surface potential of the head, he arrived at the conclusion that these variants reflect the formation of stable neuronal pools that synchronously generate electrical activity in the brain. These variants can be represented by a sequence of temporally fixed patterns with a duration of approximately 40 to 120 milliseconds [3]. These configurations are referred to as EEG microstates. The duration of a particular microstate is indicative of the co-preservation and stability of its neuronal assembly, while the frequency of registration is associated with the activity (activation) of neuronal generators during the execution of the studied brain function [4]. However, the registration of individual EEG microstates did not allow for the determination of the sources of the registered activity or the specificity of this process to the thinking process. This issue was addressed during the period of 1994 to 1997 by R.D. Pascual-Marqui and colleagues [5], who developed a system for addressing the inverse problem of EEG. The model is predicated on the integration of dipole localization with a layer-by-layer model of the head. This technique was termed low-resolution electromagnetic tomography (LORETA).
By carefully analyzing the sequence of EEG microstates, the researcher was able not only to assess the characteristics of large brain networks, but also to compare them with each other. The results of the research revealed that the EEG signals show a high variability between individuals, which indicates a low correlation between them. Therefore, a comparative analysis of events is essential. However, comparison of EEG recordings has shown that the lack of a singular starting point for the event leads to phase shifts in the EEG signal, making it difficult to achieve high accuracy. Consequently, the utilization of conventional signal correlation and coherence methods became impractical. To address this challenge, D. Lehmann proposed a methodology involving the segmentation of continuous EEG data into distinct components using clustering techniques. This approach resulted in the generation of an array of discrete recording segments with analogous electrophysiological characteristics (termed "microstates") while preserving the stability of key parameters. In the course of simultaneous studies with EEG-correlated functional magnetic resonance imaging (fMRI), I. Neuner et al. [6] concluded that the LORETA technology reveals separate standard networks with a special "electrophysiological subscript", created by a combination of different brain rhythms reflecting one or another presumed function. The study of autonomously functioning components of a cohesive brain network that are directly related to individual cognitive characteristics has the potential to provide a diagnostic approach to assessing the activity of specific brain structures. Cluster analysis techniques have been shown to identify up to 39 separate EEG microstates. However, it has been found that maximum representativeness can be achieved in only 2-8 first classes. These classes are responsible for the realization of the basic and most stable brain functions. Impairment of these functions is clinically manifested by severe mental disorders [7].
In 1999, the originator of the LORETA technology presented a neurophysiological instrument analogous to classical functional imaging techniques, such as positron emission tomography and fMRI. This instrument incorporated high temporal resolution statistical parametric mapping into the template-based neuroanatomy quantification technique developed at the Montreal Neurological Institute (MNI) Imaging Center.
A novel approach for the neurophysiological study of individual neural structures in the brain has been developed. This approach utilizes a combination of technologies, including individual EEG microstate extraction and EEG inverse problem solving. Having sufficient spatial accuracy with a smaller time delay for execution, it became a tool to study the activity of neural structures that is not inferior in information content to the methods of neuroimaging already widely used in the clinic.
The functionality of the brain is achieved through the activity of either individual neural networks or autonomously operating components of a brain network. These components are capable of manifesting their activity not only in the mode of active functioning but also in periods of relative rest. These components are directly related to individual-personal characteristics of human thinking [8].
In 2018, J.M. Kernbach et al. [9] proposed a division of the default network into networks of "internal" and "external" systems, designed to process incoming and outgoing information separately. The main neural network of the passive brain functioning includes the posterior medial cortex, the medial prefrontal cortex, and the temporoparietal junction, integrating areas of information evaluation with areas of information analysis. This structure is regarded as a type of "internal system," exhibiting elevated activity during rest periods and a decline in activity during experimental tasks [10]. Given that the state of passive relaxed wakefulness with eyes closed is reproduced in different people with greater representativeness than separate types of stress tests, the analysis of passive brain network activity should be recognized as the most convincing [10].
The observed alterations in their characteristics are predominantly influenced by structural changes in the anatomical structures that generate neural networks. Consequently, in the absence of an organic substrate, the analysis of the rate of representation of each of the distinguished EEG microstates may not differ from conditionally normal values. This phenomenon has been observed in patients with genetic variants of epilepsy, when researchers are unable to detect organic changes using neuroimaging techniques due to the high excitability of cortical neurons. In such conditions, the disease induces alterations in the sequence of excitation of cortical structures [11, 12].
However, the validity of these assumptions remains to be substantiated by a substantial body of empirical evidence.
The aim of this study was to investigate the mechanisms of cognitive dysfunction in patients with genetically confirmed HLD using neurophysiological methods based on modern EEG systems.
Patients and Methods
The study group comprised 24 European patients (13 women and 11 men) with genetically confirmed neurological stage of HLD. The mean age of the participants was 29.76 years (±4.21 years). All patients exhibited active daily living (ADL) and were carriers of the ATP7B gene in its homozygous state. The genetic alterations included the His1069Gln mutation in 10 patients, the 2304insC mutation in 7 patients, the Glu1064Lys mutation in 4 patients, the 3402delC mutation in 2 patients, and the Gly710Ser mutation in 1 patient. The comparison group consisted of 24 age- and sex-matched healthy individuals.
The clinical study was grounded in the diagnostic criteria of the 2001 Leipzig Scale, with consideration given to the severity of the clinical manifestations.
The molecular genetic study was designed to isolate causal mutations in the ATP7B gene by Sanger sequencing of the ATP7B gene.
A neuropsychological examination was conducted using the Mini-Mental Status Examination (MMSE) and the Hospital Anxiety and Depression Scale (HADS) to ascertain the severity of cognitive impairment, anxiety levels, and depressive symptoms.
The neurophysiological examination included an EEG performed on a 52-channel bioamplifier of domestic manufacture with a base frequency of the analog-digital converter of 500 Hz. This configuration facilitated the acquisition of reliable data within the range of 1 to 250 Hz, with no loss of information. The obtained information was processed using the sLORETA v20210701 software package (University of Zurich; Switzerland), as well as by implementing technological prototypes using the EEGLAB and BRAINSTORM interpreted software packages, implemented under the control of the MATLAB system (Mathworks ver. 98, USA).
The severity of the neurological manifestations of HLD was determined by considering the deficiencies in ADL and the sum of the scores of seven signs (dysarthria, tremor, ataxia, rigidity/bradykinesia, chorea/dystonia, cognitive deficit with respect to MoCA, HADS, epileptic seizures). The evaluation of these signs was conducted using a 3-point scale, with 0 points indicating the absence of signs, 1 point indicating 1-2 signs, 2 points indicating 3-4 signs, and 3 points indicating 5-7 signs.
The following scores were used to differentiate the severity of clinical manifestations: mild (1 point and no ADL deficiencies), moderate (2-3 points but no ADL deficiencies), and severe (2-3 points with ADL deficiencies).
The statistical analysis was conducted using SPSS Statistics v. 23.0 (IBM, USA). The normality of distribution was assessed using the Kolmogorov-Smirnov test, while the significance of differences was determined using the chi-squared test.
Results
The core clinical manifestations exhibited by all patients included extrapyramidal motor signs and mental disturbances associated with organ failure symptoms. The predominant motor abnormality was complex hyperkinesia, manifesting as dystonia accompanied by myoclonus and choreiform movements, nonrhythmic tremor, ballism, and tonic dystonia. However, in each specific case, an individual rhythm of muscle twitches, their complexity, and the unique character of signs of organ failure were reported. Depending on the leading syndrome, the following forms of HLD were identified: tremor (10 patients), arrhythmic hyperkinetic (9 patients), rigid tremor (2 patients), and extrapyramidal cortical (3 patients).
The most precise clinical diagnosis of HLD was substantiated by the findings of abdominal ultrasound, brain MRI, and micro-ophthalmoscopy. The rate of abdominal organ and renal abnormalities detected by ultrasound was 75% (up to 80% in patients with mild and 100% in patients with moderate HLD).
A total of 39% of patients exhibited specific abnormalities as indicated by brain MRI results. The highest detection rate was observed in patients with moderate disease and the His1069Gln mutation. No structural abnormalities of the cerebral cortex were reported.
The Kayser-Fleischer ring was identified through ophthalmic examination in 32% of patients, with nearly equivalent prevalence observed across different severity levels of HLD.
Jaundice was reported in 14% of patients.
The severity of clinical manifestations of HLD was classified as mild in 14 patients (mean age 29.9 years) and moderate in 10 patients (mean age 24.7 years).
The predominant neuropsychological impairments included disturbances in planning any kind of activity, concentration of attention, and memory dysfunction, as well as impaired oral counting. These impairments resulted in errors in task performance, behavioral disturbances, and neurotic reactions (see Table 1). The impairment of the action plan was reported with a significantly higher frequency in patients with moderate HLD.
Table 2 presents the findings from clinical assessments of anxiety and cognitive function, employing the Mini-Mental State Examination (MMSE) and the Hospital Anxiety and Depression Scale (HADS).
The severity of anxiety and cognitive dysfunction is contingent upon the severity of HLD.
Tables 3 and 4 and Figure 1 illustrate the characteristics of activity of individual brain structures based on cluster analysis of EEG microstates in patients with different mutations in the ATP7B gene.
A notable finding was the identification of significant abnormalities in the contribution of individual EEG microstates to the total scalp potential in carriers of the Glu1064Lys mutation. These abnormalities were characterized by a substantial increase in the contribution of class I and II microstates, accompanied by a concurrent decrease in the contribution of class IV, V, VI, and VIII microstates. No significant differences were identified between patients with other mutations and the control group (see Figure 1).
The rate of registration of individual classes of EEG microstates differed significantly in carriers of the Glu1064Lys mutation due to a significant increase in the contribution of classes I and II and a decrease in the contribution of classes IV, V, and VI. No significant differences were identified between patients with other mutationsand the control group (see Table 3).
As illustrated in Table 4, cluster analysis indicated the maximum activity of EEG microstates in zones 2 and 7 of class I, which are responsible for brain function implementation (impairment of these zones is manifested by severe mental disorders) with decrease of activity in zone of class IV responsible for gnosis only in carriers of Glu1064Lys mutation. In carriers of other mutations, the activity of brain structures in the same zones did not differ from those of healthy subjects.
As demonstrated in Figure 2, the distribution of rhythmic activity on the scalp surface when solving the inverse EEG task for each of the eight classes of EEG microstates in HLD clearly demonstrates the presence of two main peaks in the visual cortical areas (Brodmann areas 18, 19, and 37). Conversely, the activity of Brodmann area 7 (the tertiary zone of information processing, or gnosis) exhibited a marked absence. A substantial discrepancy from the conditional norm was identified exclusively among patients bearing the His1069Gln mutation, in contrast to those carrying the Glu1064Lys, 2304insC, and 3402delC mutations, where the discrepancy proved to be non-significant.
Discussion
The neuropsychological study using the MoCA and HADS scales has demonstrated that the underlying cause of the cognitive deficit in HLD is the impairment of program construction and activity planning in the context of attentional concentration deficits and changes in the speed of its distribution. These elements have the potential to contribute to errors in the execution of test tasks. The presence of anxiety and depression has been observed in all patients, irrespective of the specific mutation type in the ATP7B gene and the severity of clinical manifestations. The severity of cognitivedeficit is within average values.
The severity of cognitive impairments and anxiety levels does not correlate with the specific mutation in the ATP7B gene. Instead, these symptoms are associated with the intensity of the clinical manifestations associated with HLD. The manifestations of these conditions reach their maximal level in patients with moderate severity, which is in agreement with the data of F.T. Kirk et al. [13].
The analysis of eight classes of EEG microstates in HLD also revealed no association between the distribution of bioelectrical activity and the severity of clinical manifestations of HLD or the type of mutation in the ATP7B gene.
The presence of functional abnormalities was found to be associated with a specific mutation in the ATP7B gene, thereby demonstrating a clear impact on the cortical regions involved in visual analysis and face recognition. This finding is consistent with the observations reported by L. Henderson et al. [14].
The elevated levels of subclinical anxiety and the average levels of cognitive deficit parameters, in the absence of MRI imaging of structural defects in the cerebral cortex, indicate that the cognitive defect in all patients is caused by changes in the activity of the basal ganglia, with the secondary effect of pathological impulses on cortical structures. Consequently, the characteristics of the cognitive deficit in different types of mutations in the ATP7B gene become analogous, which is consistent with the findings of S. Shribman et al. [15] and corroborates the hypothesis of N. Gutierrez-Avila et al. [16] on the development of a "syndrome of frontal-striatal relationship disorders" in GLD. Thissyndrome is a specific neurocognitive defect, characterized by impaired verbal and visual memory, accompanied by slowthinking and pronunciation difficulties.
Basal ganglia lesions result in a range of motor dysfunction symptoms, including local and generalized dystonia, as well as nonrhythmic tremor, other fast and slow hyperkinesia. These symptoms can garner attention from others and compel patients to maintain a state of constant readiness to resist the activity of their body [17, 18].
As posited by T. Litwin et al. [19], elevated anxiety levels have been demonstrated to precipitate the onset of psychosis.
Conclusions
In summary, the present study illustrates the extensive potential inherent in the utilization of contemporary neurophysiological methodologies founded upon EEG systems. This approach facilitates the analysis of the involvement of brain structures in the execution of complex cognitive tasks, the role of specific regions in the development of various cognitive disorders, the severity of these disorders, and the nature of damage to cortical areas responsible for cognitive function. This phenomenon is of particular interest for two primary reasons. Firstly, it facilitates the early diagnosis of diseases associated with the impairment of higher cortical functions. Secondly, it fosters the development of modern rehabilitation tools for patients with cognitive disorders. The method of EEG microstate evaluation, a technique employed for the study of cognitive processes, facilitates the examination of brain structures across diverse clinical models. In this regard, HLD emerges as a prominent clinical model for investigating the underlying mechanisms of cognitive disorders.
About the authors:
Elena V. Ovchinnikova — postgraduate student, V.M. Bekhterev National Research Medical Center for Psychiatry and Neurology; 3, Bekhterev str., St. Petersburg, 192019, Russian Federation; Assistant of the Department of Clinical Medicine, Far Eastern Federal University; 10, Ajax Village, Russky Island, Vladivostok, 690922, Russian Federation; neurologist, Primorye Territory Clinical Hospital No. 1; 57, Aleutskaya str., Vladivostok, 690091, Russian Federation; ORCID iD 0000-0002-4106-1163
Natalia A. Schneider — Dr. Sc. (Med.), Chief Scientific Officer, Deputy Head of the Institute of Personalized Psychiatry and Neurology, V.M. Bekhterev National Research Medical Center for Psychiatry and Neurology; 3, Bekhterev str., St. Petersburg, 192019, Russian Federation; Professor, Prof. V.F. Voino-Yasenetsky Krasnoyarsk State Medical University; 1, Partizan Zheleznyak str., Krasnoyarsk, 660022, Russian Federation; ORCID iD 0000-0002-2840-837X
Anna A. Ovchinnikova — Dr. Sc. (Med.), Professor of the Department of Clinical Medicine, Far Eastern Federal University; 10, Ajax Village, Russky Island, Vladivostok, 690922, Russian Federation; ORCID iD 0000-0002-6336-8166
Regina F. Nasyrova — Dr. Sc. (Med.), Professor, Chief Scientific Officer, Head of the Institute of Personalized Psychiatry and Neurology, V.M. Bekhterev National Research Medical Center for Psychiatry and Neurology; 3, Bekhterev str., St. Petersburg, 192019, Russian Federation; ORCID iD 0000-0003-1874-9434
Sergey A. Gulyaev — Dr. Sc. (Med.), Head of the Department of Simulation Technologies, Engineering Physics Institute of Biomedicine, Assistant Professor of the Department of Fundamental Medicine; National Research Nuclear University MEPhI; 31, Kashirskoe road, Moscow, 115409, Russian Federation; ORCID iD 0000-001-9122-7144
Contact information: Elena V. Ovchinnikova, e-mail: ovchinnikovaelv@mail.ru
Financial Disclosure: no authors have a financial or property interest in any material or method mentioned.
There is no conflict of interest.
Received 04.08.2024.
Revised 27.08.2024.
Accepted 19.09.2024.
2. Lucena-Valera A., Ruz-Zafra P., Ampuero J. Wilson's disease: overview. Med Clin (Barc). 2023;160(6):261–267. DOI: 10.1016/j.medcli.2022.12.016
3. Mishra A., Englitz B., Cohen M.X. EEG microstates as a continuous phenomenon. Neuroimage. 2020;208:116454. DOI: 10.1016/j.neuroimage.2019.116454
4. Milz P., Faber P.L., Lehmann D. et al. The functional significance of EEG microstates-Associations with modalities of thinking. Neuroimage. 2016;125:643–656. DOI: 10.1016/j.neuroimage.2015.08.023
5. Pascual-Marqui R.D., Michel C.M., Lehmann D. Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain. Int J Psychophysiol. 1994;18(1):49–65. DOI: 10.1016/0167-8760(84)90014-x
6. Neuner I., Arrubla J., Werner C.J. et al. The default mode network and EEG regional spectral power: a simultaneous fMRI-EEG study. PLoS One. 2014;9(2):e88214. DOI: 10.1371/journal.pone.0088214
7. Sun Q., Zhou J., Guo H. et al. EEG Microstates and Its Relationship With Clinical Symptoms in Patients With Schizophrenia. Front Psychiatry. 2021;12:761203. DOI: 10.3389/fpsyt.2021.761203
8. Buckner R.L., DiNicola L.M. The brain's default network: updated anatomy, physiology and evolving insights. Nat Rev Neurosci. 2019;20(10):593–608. DOI: 10.1038/s41583-019-0212-7
9. Kernbach J.M., Yeo B.T.T., Smallwood J. et al. Subspecialization within default mode nodes characterized in 10.000 UK Biobank participants. Proc Natl Acad Sci USA. 2018;115(48):12295–12300. DOI: 10.1073/pnas.1804876115
10. Di Plinio S., Ebisch S.J.H. Probabilistically Weighted Multilayer Networks disclose the link between default mode network instability and psychosis-like experiences in healthy adults. NeuroImage. 2022;257:119291. DOI: 10.1016/j.neuroimage.2022.119291
11. Keator L.M., Yourganov G., Faria A.V. et al. Application of the dual stream model to neurodegenerative disease: Evidence from a multivariate classification tool in primary progressive aphasia. Aphasiology. 2022;36(5):618–647. DOI: 10.1080/02687038.2021.1897079
12. Chang I.J., Hahn S.H. The genetics of Wilson disease. Handb Clin Neurol. 2017;142:19–34. DOI: 10.1016/B978-0-444-63625-6.00003-3
13. Kirk F.T., Munk D.E., Laursen T.L. et al. Cognitive impairment in stable Wilson disease across phenotype. Metab Brain Dis. 2021;36(7):2173–2177. DOI: 10.1007/s11011-021-00804-6
14. Henderson L., Dittrich W.H. Preparing to react in the absence of uncertainty: I. New perspectives on simple reaction time. Br J Psychol. 1998;89(Pt4):531–554. DOI: 10.1111/j.2044-8295.1998.tb02702.x
15. Shribman S., Burrows M., Convery R. et al. Neuroimaging Correlates of Cognitive Deficits in Wilson's Disease. Mov Disord. 2022;37(8):1728–1738. DOI: 10.1002/mds.29123
16. Gutierrez-Avila N., Zuniga-Marguez J., Burgos-Torres N. et al. Dysexecutive Syndrome in a Patient with Wilson's Disease. Psychology. 2014;5(1):47–52. DOI: 10.4236/psych.2014.51009
17. Wang G., Jin P., Ma X. et al. Cognitive and gait in Wilson's disease: a cognitive and motor dual-task study. Front Neurol. 2023;14:1243122. DOI: 10.3389/fneur.2023.1243122
18. Tian L., Dong T., Hu S. et al. Radiomic and clinical nomogram for cognitive impairment prediction in Wilson's disease. Front Neurol. 2023;14:1131968. DOI: 10.3389/fneur.2023.1131968
19. Litwin T., Dzieżyc K., Karliński M. et al. Psychiatric disturbances as a first clinical symptom of Wilson's disease — case report. Psychiatr Pol. 2016;50(2):337–344. DOI: 10.12740/PP/45218
This work is licensed under a Creative Commons «Attribution» 4.0 License.
