Reinytė, Karolina

Postoperative delirium is a common complication following sub-thalamic nucleus deep brain stimulation surgery in Parkinson's disease patients. Postoperative delirium has been shown to prolong hospital stays, harm cognitive function, and negatively impact outcomes. Utilizing radiomics as a predictive tool for identifying patients at risk of delirium is a novel and personalized approach. This pilot study analyzed preoperative T1-weighted and T2-weighted magnetic resonance images from 34 Parkinson's disease patients, which were used to segment the thalamus, amygdala, and hippocampus, resulting in 10,680 extracted radiomic features. Feature selection using the minimum redundancy maximal relevance method identified the 20 most informative features, which were input into eight different machine learning algorithms. A high predictive accuracy of postoperative delirium was achieved by applying regularized binary logistic regression and linear discriminant analysis and using 10 most informative radiomic features. Regularized logistic regression resulted in 96.97% (±6.20) balanced accuracy, 99.5% (±4.97) sensitivity, 94.43% (±10.70) specificity, and area under the receiver operating characteristic curve of 0.97 (±0.06). Linear discriminant analysis showed 98.42% (±6.57) balanced accuracy, 98.00% (±9.80) sensitivity, 98.83% (±4.63) specificity, and area under the receiver operating characteristic curve of 0.98 (±0.07). The feed-forward neural network also demonstrated strong predictive capacity, achieving 96.17% (±10.40) balanced accuracy, 94.5% (±19.87) sensitivity, 97.83% (±7.87) specificity, and an area under the receiver operating characteristic curve of 0.96 (±0.10). However, when the feature set was extended to 20 features, both logistic regression and linear discriminant analysis showed reduced performance, while the feed-forward neural network achieved the highest predictive accuracy of 99.28% (±2.71), with 100.0% (±0.00) sensitivity, 98.57% (±5.42) specificity, and an area under the receiver operating characteristic curve of 0.99 (±0.03). Selected radiomic features might indicate network dysfunction between thalamic laterodorsal, reuniens medial ventral, and amygdala basal nuclei with hippocampus cornu ammonis 4 in these patients. This finding expands previous research suggesting the importance of the thalamic-hippocampal-amygdala network for postoperative delirium due to alterations in neuronal activity.

Introduction Parkinson's disease, the second most common neurodegenerative disorder, is marked by progressive loss of dopaminergic neurons. Dopaminergic disruption causes severe motor symptoms (resting tremor, rigidity, bradykinesia, postural instability), cognitive deficits, and reduced quality of life in aging. While incurable, treatments such as deep brain stimulation can alleviate symptoms. We aim to develop personalized virtual brain models for Parkinson's patients that explicitly incorporate dopaminergic modulation. Building on our earlier work, we now refine these models by integrating individual structural connectivity, advanced neural mass formulations, and EEG-derived biomarkers, enabling more precise characterization of patient-specific dynamics. Methods We introduced a modular framework to capture dopaminergic regulation at the neural mass scale (Depannemaecker, 2024; Casagrande, 2025). Targeting D1-type receptor dynamics, it enables investigation of how fluctuations in dopamine availability shape population responses. The framework adopts a mean-field formulation (Chen & Campbell, 2022), providing a tractable yet biologically plausible description of macroscopic activity, well-suited to study dopamine-mediated modulation of basal ganglia circuits in health and disease.To identify biomarkers of brain dynamics in Parkinson’s disease (PD), we analyzed resting-state EEG acquired before and after DBS electrode implantation. Preoperatively, patients were tested ON and OFF L-DOPA; postoperatively, six months later, they were assessed with DBS ON and OFF, under both medication conditions. EEG features—Amplitude Envelope Correlation (AEC) and Phase Locking Value (PLV)—were extracted, and significant effects identified with permutation ANOVA. Results Our framework reveals that dopaminergic modulation critically reshapes the dynamical repertoire of neural mass models. By systematically varying dopamine input levels, we identified transitions between qualitatively distinct regimes, including fixed-point convergence, sustained oscillatory activity (limit cycles), and bursting dynamics. Progressive increases in dopaminergic tone induced a marked expansion of quiescent states, accompanied by reduced oscillatory frequencies across both fast spiking and inter-burst domains. Importantly, these model-derived regimes correspond to electrophysiological features observed in EEG and DBS recordings from Parkinsonian patients, thereby providing a mechanistic account of dopamine-dependent neural dynamics. Analysis of EEG data showed that the most significant changes were found in the reduction in the Beta band for both PLV and AEC features, as reported in previous studies. Discussion Our results show that dopaminergic tone critically shapes neural mass dynamics, inducing transitions between quiescent, oscillatory, and bursting regimes. The observed reduction in oscillatory frequency and expansion of quiescent states parallel electrophysiological signatures in Parkinsonian EEG and DBS data, particularly in the Beta band. Next, we will integrate model-derived features with empirical connectivity into a unified framework, extending earlier approaches (Angiolelli, 2025) by incorporating DBS effects. This strategy advances toward personalized dynamical models to guide understanding and optimization of neuromodulation therapies.

Parkinson’s disease (PD) is a neurodegenerative disorder marked by motor impairments, often accompanied by pathological oscillations in brain networks (notably excessive beta-band activity). Understanding these alterations is crucial for improving diagnosis and guiding therapies such as deep brain stimulation (DBS). In this study, we combined a computational model with source-reconstructed EEG to examine how dopaminergic modulation and DBS affect brain network activity in PD. We created a neural mass model that accounts for dopamine-driven modulation. The model suggests that increasing dopamine reduces pathological oscillations and promotes more stable brain activity. To test this prediction, we analyzed high-density EEG from a PD patient recorded before DBS (on/off medication) and six months after DBS implantation (on/off stimulator, on/off medication). EEG signals were source-reconstructed using the patient’s MRI, and functional connectivity metrics (amplitude envelope correlation, AEC; phase-locking value, PLV) were calculated across standard frequency bands. Our results showed that beta-band functional connectivity was markedly reduced after DBS – the largest change among all frequency bands. Beta-band coupling between cortical regions (especially frontal and temporal areas) significantly decreased postDBS, whereas changes in delta, theta, alpha, and gamma bands were minor. This reduction in pathological beta synchrony aligns with our model’s prediction that enhanced dopaminergic signaling suppresses excessive network oscillations and corresponds to improved motor function with DBS therapy. These findings highlight beta-band connectivity as a key biomarker of PD network changes and demonstrate that integrating computational modeling with EEG connectivity analysis yields mechanistic insight into DBS effects, informing personalized neuromodulation strategies. To further support clinical translation, these models will be embedded into The Virtual Brain platform to create individualized digital twins of Parkinson’s disease patients.

Although Artificial Intelligence and its applications in medicine are growing rapidly, its integration into mental health remains at an early stage. The aim of the study is to evaluate psychology students’ attitudes and the significance of brief interventions towards Artificial Intelligence systems and their application in mental healthcare. The study involved 62 psychology students (M= 23.19±4.69; 85.5% women). Thirty-one participants tested the Artificial Intelligence-based emotional support app Wysa, and thirty-one others watched a presentation on the use of Artificial Intelligence in providing psychological assistance, based on the latest scientific research. Attitudes towards Artificial Intelligence were assessed before and after the interventions using the General Attitudes towards Artificial Intelligence Scale and an adapted version of the Questionnaire for Attitudes Toward Medical Application of Artificial Intelligence to examine psychologists’ attitudes towards the use of artificial intelligence in their work. During interventions, facial expression analysis software FaceReader was used to assess participants’ emotions. Following a scientific presentation, participants showed significative increases in positive attitudes, compared to those who used Wysa. While improvements in negative attitudes were noted, these did not differ significantly between groups. Positive changes in perceived Artificial Intelligence advantages were positively associated with feelings of surprise and fear, and negatively with contempt and disgust. Perceived Artificial Intelligence disadvantages correlated positively with contempt. The scientific presentation helped students develop more positive attitudes toward Artificial Intelligence, suggesting that education on Artificial Intelligence is important in shaping future psychologists’ view on new technologies. Emotional responses (particularly surprise, fear, disgust, and contempt) played a significant role in these attitude changes.

Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterised by motor and non-motor symptoms, and is associated with altered brain network dynamics. Understanding large-scale functional changes is crucial for improving diagnosis, treatment planning, and monitoring of disease progression or intervention effects. Electroencephalography (EEG) combined with source reconstruction and magnetic resonance imaging (MRI) data facilitates the analysis of cortical dynamics with greater anatomical precision. The present study investigates differences in preoperative and postoperative brain activity in a PD patient using source-constructed EEG. The objective of this study is to extract and compare the connectivity and dynamical features of brain networks to identify functional alterations associated with the deep brain stimulation (DBS) treatment. The resting state EEG dataset of PD patients was collected before and after DBS electrode implantation. The patients were treated at the Department of Neurosurgery, Lithuanian University of Health Sciences Hospital in Kaunas, Lithuania, 2024–2025. Preoperatively, EEG recordings were obtained when the patients were off L-DOPA medication, and postoperatively with the DBS electrode activated and off L-DOPA medication. Post-surgery EEGs were recorded six months later after the surgical intervention. All EEG recordings were acquired using a 64-channel ANT Neuro device. The study protocol was approved by the Kaunas Regional Biomedical Research Ethics Committee (No. BE-2-115), informed consent was obtained from every study participant. EEG preprocessing was performed using Matlab EEGLab and Makoto’s Miyakoshi pipeline. Brain sources were reconstructed utilising processed EEGs in Brainstorm, using patient MRI data to create Boundary Element Method (BEM) surfaces and an individualized head model via OpenMEEG. Electrode positions were manually adjusted, EEG data was concatenated, and downsampled to the Desikan-Killiany atlas. Source scouts were then extracted and resegmented to the 3-second epochs. The EEG features Amplitude Envelope Correlation (AEC) and Phase Lock Value (PLV) were calculated. A Wilcoxon permutation test (1000 permutations) was used to assess the statistical significance at the edge level, focusing on connections between brain regions. The resulting p-values were MaxT corrected to control the family-wise error rate, and the effect size was calculated. All statistically significant edges exhibited an effect size greater than 0.5. The most significant regions were selected by ranking PLV and AEC metrics. The top 15 of both features were compared to find regions with most statistically significant edges. A normalised sum of all statistically significant edges per EEG band was also calculated for comparison. The most significant changes were found in the Beta band for both AEC and PLV features, as reported in previous studies [1]. Specifically, the normalised number of the significant AEC features in the following bands were as follows: Delta: 0.00, Theta: 1.18, Alpha: 2.71, Beta: 15.82, Gamma: 5.85; the number of the PLV features were: Delta: 0.00, Theta: 1.56, Alpha: 3.53, Beta: 13.62, Gamma: 9.68. The regions with the most significant edges were the Fusiform Gyrus on both left and right side, Pericalcarine Cortex, Parahippocampal Gyrus, Supramarginal Gyrus, Temporal Pole, Pars Opercularis, Precentral Gyrus, and Superior Frontal Gyrus on the left hemisphere, and Caudal Middle Frontal Gyrus and Superior Temporal Gyrus on the right hemisphere. The findings of this study indicate that the most significant alterations in brain connectivity occurred in the Beta frequency band for both AEC and PLV metrics [1]. In contrast, other frequency bands showed comparatively minor alterations. These results suggest that DBS-induced modulation predominantly affects the Beta band connectivity within these critical brain regions, which may have implications for understanding the neural mechanisms underlying DBS efficacy in Parkinson’s disease

Background/Objectives: Synesthesia is an unusual neurological condition when stimulation of one sensory modality automatically triggers an additional sensory sensation in an additional unstimulated modality. In this study, we investigated a case of sound–color synesthesia in a female with impaired vision. After confirming a positive case of synesthesia, we aimed to determine the sound features that played a key role in the subject’s sound perception and color development. Methods: We applied deep neural networks and a benchmark of binary logistic regression to classify blue and pink synesthetically voice-evoked color classes using 136 voice features extracted from eight study participants’ voice recordings. Results: The minimum Redundancy Maximum Relevance algorithm was applied to select the 20 most relevant voice features. The recognition accuracy of 0.81 was already achieved using five features, and the best results were obtained utilizing the seventeen most informative features. The deep neural network classified previously unseen voice recordings with 0.84 accuracy, 0.81 specificity, 0.86 sensitivity, and 0.85 and 0.81 F1-scores for blue and pink classes, respectively. The machine learning algorithms revealed that voice parameters, such as Mel-frequency cepstral coefficients, Chroma vectors, and sound energy, play the most significant role. Conclusions: Our results suggest that a person’s voice’s pitch, tone, and energy affect different color perceptions.

Įvadas. Nors medicinoje dirbtinis intelektas (DI) ir jo pritaikymas vystosi greitai, ir yra naudingas tiek medicinos personalui, tiek pacientams (Liuir kt., 2018), dirbtinio intelekto integracija į psichikos sveikatos priežiūrą vis dar yra ankstyvoje stadijoje (Fiske ir kt., 2019). Tyrimo tikslas –įvertinti psichologijos studentų požiūrį ir trumpų intervencijų reikšmę į psichologinę pagalbą teikiančias dirbtinio intelekto sistemas.Metodai.Tyrime dalyvavo 53 psichologijos studentai nuo 18 iki 43 metų (M=23,40 ± 5,024; 83% moterys) -27 studentai išmėgino DI paremtą emocinės pagalbos programėlę „Wysa“, o 26 studentai stebėjo moksliniais duomenimis grįstą prezentaciją apie DI panaudojimą psichologinei pagalbai teikti. Požiūris į DI vertintas prieš ir po intervencijų Bendro požiūrio į dirbtinį intelektą skale ir Oh ir kt. (2019) klausimynu, pritaikytu tirti psichologų požiūrį į DI taikymą savo darbe. Abiejų intervencijų metu, emocijų vertinimui taikyta automatizuota veido išraiškų matavimo sistema „FaceReader“.Rezultatai. Lyginant grupes nustatyti statistiškai reikšmingi emocijų skirtumai -Wysa grupėje stebėti aukštesni „pykčio“, „paniekos“, „pasibjaurėjimo“ ir „laimės“ įverčiai (Mann-Whitney U p= 0,04; 0,01; <0,001; <0,001), o pristatymo grupėje –aukštesnis „neutralios“ būsenos lygis (Mann-Whitney U p<0,00). XXI JMPK28Lyginant požiūrį prieš ir po intervencijų, nustatyta, kad „Bendro požiūrio į DI psichologijoje“ ir „DI naudojimo psichologijoje privalumų“ pokytis buvo statisiškai reikšmingas ir priklausė nuo grupės (kartotinių matavimų ANOVA, atitinkamai F(1,51) = 14, 176, p < 0.001; F(1,51) = 6,779, p= 0,012) –pristatymą stebėjusių studentų tarpe, požiūrio įverčiai po intervencijos buvo statistiškai reikšmingai aukštesni (F(1,25) = 28,580, p< 0,001; F(1,25) = 6,882, p= 0,015). Po intervencijos, prezentaciją stebėję dalyviai demonstravo ir reikšmingai aukštesnius „Teigiamo požiūrio į DI“ įverčius (F(1,25) = 18,496, p= 0,017). Programėlę naudojusių dalyvių tarpe reikšmingi požiūrio į DI pokyčiai nenustatyti.Išvados.Po prezentacijos apie dirbtinio intelekto panaudojimą psichologinei pagalbai teikti, stebėti teigiami požiūrio į DI psichologijoje pokyčiai, o DI paremtos emocinės pagalbos programėlės išbandymas dalyvių požiūrio nepakeitė. Tai gali lemti didesnis studentų pasitikėjimas mokslu grįstais duomenimis apie DI nei praktiniu metodų išbandymu.

Įvadas. Nors medicinoje dirbtinis intelektas (DI) ir jo pritaikymas vystosi greitai, ir yra naudingas tiek medicinos personalui, tiek pacientams (Liu ir kt., 2018), dirbtinio intelekto integracija į psichikos sveikatos priežiūrą vis dar yra ankstyvoje stadijoje (Fiske ir kt., 2019). Tyrimo tikslas – įvertinti psichologijos studentų požiūrį į psichologinę pagalbą teikiančias DI sistemas ir požiūrio pokytį po trumpų intervencijų. Metodai. Tyrime dalyvavo 35 psichologijos studentai (amžius=23,11 ± 4,993; 85,7% moterys) - 18 studentų išmėgino DI paremtą emocinės pagalbos programėlę „Wysa“, o 17 studentų stebėjo moksliniais duomenimis grįstą prezentaciją apie DI panaudojimą psichologijoje. Požiūris į DI vertintas prieš ir po intervencijų Bendro požiūrio į dirbtinį intelektą skale ir Oh ir kt. (2019) klausimynu, pritaikytu tirti psichologų požiūrį į DI taikymą savo darbe. Abiejų intervencijų taikymo metu taip pat naudota ir automatizuota emocijų vertinimo sistema „FaceReader“. Rezultatai. Lyginant grupes nustatyti statistiškai reikšmingi emocijų skirtumai - Wysa grupėje stebėti aukštesni „pykčio“, „paniekos“, „pasibjaurėjimo“, „laimės“ ir „baimės“ įverčiai (Mann-Whitney U p=0.02; 0.03; <0.001; <0.001; 0.09), o pristatymo grupėje – aukštesnis „neutralios“ būsenos lygis (Mann-Whitney U p<0.00). Lyginant požiūrio įverčius prieš ir po intervencijų, nustatyta, kad prieš intervencijas „Bendras požiūris į DI psichologijoje“ tarp grupių nesiskyrė (Mann-Whitney U p=0.252), o po intervencijų, pristatymą stebėjusių tiriamųjų tarpe, „Bendras požiūris į DI psichologijoje“ buvo statistiškai reikšmingai aukštesnis, lyginant su „Wysa“ grupe (MannWhitney U p=0.010). Analizuojant pokyčius grupėse nustatyta, kad po intervencijos, prezentaciją stebėję dalyviai demonstravo reikšmingai aukštesnius „Bendro požiūrio į DI psichologijoje“, „DI naudojimo psichologijoje privalumų“ ir „Teigiamo požiūrio į DI“ įverčius (Vilkoksono ženklų kriterijus p=0.001; 0.013; 0.017). Programėlę naudojusių dalyvių tarpe reikšmingi požiūrio į DI pokyčiai nestebėti. Diskusija ir išvados. Po prezentacijos apie dirbtinio intelekto panaudojimą psichologijoje, stebėti teigiami požiūrio į DI psichologijoje pokyčiai, o DI paremtos emocinės pagalbos programėlės išbandymas dalyvių požiūrio nepakeitė – tai galėjo lemti didesnis studentų pasitikėjimas mokslu grįstais duomenimis.

Introduction. While artificial intelligence (AI) and its application in medicine are developing rapidly, its integration into mental health care is nascent. The prevalence of negative psychological states create a need to further investigate the potential effectiveness of AI and its application methods. The aim of the study is to evaluate psychology students’ attitudes and the significance of brief interventions towards AI systems and their application in mental healthcare. Methods. The study included 53 psychology students – the first group (n = 27) utilized Wysa, a mobile app offering emotional support through AI-driven chat conversations; second group (n = 26) watched a recorded presentation on AI in psychology. Attitudes towards AI were evaluated pre- and post-interventions in both groups using The General Attitudes towards Artificial Intelligence Scale and Oh et al. (2019) questionnaire. During interventions, facial expression analysis software “FaceReader” was used to assess participants’ emotions. Results. Significant differences in emotions emerged between intervention groups, with the Wysa group exhibiting higher levels of “angry”, “contempt”, “disgusted” and “happy” (Mann-Whitney p = 0.04; 0.01; < 0.001; < 0.001, correspondingly) while the presentation group demonstrated a higher prevalence of the “neutral” state (Mann-Whitney p < 0.001). Comparing attitudes before and after interventions revealed a statistically significant change in both “Perception of AI in psychology” and “Advantages of using AI in psychology” dependent on the group (Repeated measures ANOVA p < 0.001; 0.012). Among the students observing the presentation, attitude scores post-intervention were significantly higher (p < 0.001; 0.015). Participants of the Wysa group did not show significant changes in the attitudes on AI technologies. This outcome could potentially be attributed to the greater confidence in science-based data about AI than in testing practical methods. Conclusions. The presentation intervention elicited significant positive changes in attitudes towards AI in psychology, whereas the Wysa intervention did not yield significant alterations in attitudes among participants.