MF Fizikos, matematikos ir biofizikos katedra (04.09)

Background/Aim Avulsion of a permanent tooth is a critical dental emergency requiring rapid intervention to optimize replantation outcomes and minimize complications. Conventional storage media maintain periodontal ligament cell viability but often lack sufficient antimicrobial protection and mechanical stabilization during transport. Hydrogels, specifically those combining gellan gum and natural additives like propolis, offer promising improvements by providing enhanced antimicrobial activity and mechanical cushioning. The aim of this study was to systematically evaluate a hydrogel-based transport medium supplemented with propolis and optimized gellan gum concentration for its ability to preserve cell viability, inhibit microbial growth, and confer mechanical stability to avulsed teeth during simulated transport.

Background and Objectives: The Canadian Occupational Performance Measure (COPM) is a widely used client-centered outcome measure designed to evaluate individuals’ self-perceived occupational performance and satisfaction with performance. Although an official Lithuanian version is available, its psychometric properties have not previously been evaluated in individuals undergoing poststroke rehabilitation. This study aimed to examine the test–retest reliability, concurrent validity, and responsiveness of the Lithuanian version of the COPM. Materials and Methods: Thirty-one individuals with subacute stroke participated in this prospective validation study. Test–retest reliability was assessed over a three-day interval using the intraclass correlation coefficient (ICC). Concurrent validity was examined by correlating COPM Performance and Satisfaction scores with the Functional Independence Measure (FIM). Responsiveness was evaluated by comparing baseline and discharge COPM scores following inpatient rehabilitation. Additional exploratory analyses assessed whether demographic or clinical characteristics were associated with COPM scores. Results: The Lithuanian version of the COPM demonstrated high test–retest reliability, with ICC values of 0.974 (95% CI: 0.947–0.988) for Occupational Performance and 0.956 (95% CI: 0.911–0.979) for Occupational Satisfaction (both p < 0.001). Concurrent validity was partially supported by weak-to-moderate correlations between the COPM and the FIM at both the initial and final evaluations. Both COPM subscales showed significant improvements following inpatient rehabilitation (p < 0.001), with a mean change of 2.37 and 2.18 points for Occupational Performance and Occupational Satisfaction, respectively. Improvements of at least 2 points, a commonly used threshold for minimal important change in COPM scores, were observed in 51.6% and 41.9% of participants, respectively. Exploratory analyses showed no significant differences according to sex, stroke type, or affected body side. Conclusions: The Lithuanian version of the COPM demonstrated high test–retest reliability, evidence partially supporting concurrent validity, and responsiveness to change in individuals undergoing poststroke rehabilitation. These findings support its use as a client-centered outcome measure for evaluating occupational performance and satisfaction alongside standardized functional assessments.

Background

Early identification of treatment response in head and neck squamous cell carcinoma (HNSCC) remains challenging, as conventional imaging mainly reflects anatomical changes rather than ongoing biological processes. Although cone-beam CT (CBCT) has limited quantitative accuracy due to artefacts and intensity variability, its routine longitudinal acquisition during radiotherapy is an opportunity to monitor dynamic intratumoural changes. This study investigates if temporal radiomic feature transitions, can reveal patterns of tumour evolution and potentially indicate treatment response, even in a setting where such information is least expected.

Methods

HNSCC patients daily longitudinal CBCT images were analysed. Tumour volumes from planning CT were propagated across treatment fractions using rigid alignment. Radiomic heterogeneity features were extracted to evaluate temporal intratumoural changes during treatment. Feature trajectories were analysed using linear mixed-effects models incorporating nonlinear time effects and tumour differentiation groups.

Results

Distinct temporal patterns were observed across radiomic features. Entropy showed nonlinear (p < 0.001) dynamics with substantial inter-patient variability, while strength demonstrated moderate temporal variation with no group discrimination. Coarseness revealed consistent temporal trajectories with significant group-dependent differences (p < 0.01) and transition points. Differences between tumour groups were driven by the timing of structural changes rather than absolute feature values, highlighting the importance of temporal pattern analysis over static assessment.

Conclusions

Longitudinal CBCT-derived radiomic features capture tumour evolution during radiotherapy, with temporal patterns providing insight beyond static metrics. Coarseness emerged as a relatively stable indicator of structural phase shifts, while entropy reflected earlier but more variable instability. These findings support time-resolved radiomic phenotyping for identifying intra-treatment response trajectories and may contribute to adaptive radiotherapy using routinely acquired imaging without additional patient burden.

Background and Objectives: Glaucoma is a neurodegenerative disease involving retinal ganglion cell loss, optic nerve damage and alterations within the central visual pathway. We evaluated visual cortex activity using functional magnetic resonance imaging (fMRI) in patients with primary open-angle glaucoma (POAG) and its associations with retinal structural and microvascular parameters assessed using optical coherence tomography (OCT) and OCT angiography (OCTA). Materials and Methods: This prospective clinical study included 19 patients with POAG and 18 healthy controls. All participants underwent ophthalmological examination, standard automated perimetry (Humphrey 24–2 SITA Standard), swept-source OCT, OCTA of the optic nerve head (ONH) (DRI-OCT Triton, Topcon, Tokyo, Japan), and visual stimulation fMRI (3T Siemens Magnetom Skyra, Siemens Healthineers, Erlangen, Germany). Retinal nerve fiber layer (RNFL) thickness, vessel density (VD), and visual cortex activity were compared between groups. Associations were assessed using Pearson’s or Spearman’s correlation according to data distribution, with Benjamini–Hochberg false discovery rate correction. Unadjusted p and adjusted q values are reported. Results: RNFL thickness was reduced in POAG patients in all four sectors. VD was lower in every sector of the superficial and deep plexus (all q < 0.05), as was outer retinal signal. Choriocapillaris findings were heterogeneous. RNFL thickness correlated with choriocapillaris VD in two of three corresponding nasal sectors, in POAG patients only. Visual cortex activation was lower in POAG patients (p = 0.046; Hedges’ g = −0.68, 95% CI −1.34 to −0.01), but not after adjustment for age (p = 0.086). No association between VD and cortical activity survived FDR correction. Conclusions: In POAG, choriocapillaris VD tracked RNFL thickness in corresponding nasal sectors, an association not detected in controls. Reduced cortical activation did not persist after adjustment for age, and no retinal–cortical microvascular association was found. These exploratory findings suggest that distinct retinal vascular layers may carry different pathophysiological information and require confirmation in larger cohorts.

Background. Radiotherapy (RT) for head and neck cancer (HNC) is associated with acute neurotoxic and functional sequelae affecting peripheral nerves, mucosal sensory pathways, speech and swallowing mechanisms. Patientreported outcome measures (PROMs) enable systematic identification of these impairments and may support family physicians in the longitudinal follow-up of cancer survivors. Aim. To assess patient-reported neurological and functional symptoms during RT in patients with HNC. Methods. A prospective observational study was conducted at the Lithuanian University of Health Sciences Oncology Hospital. Adult HNC patients undergoing active RT completed the EORTC QLQ-C30 and the head and neck-specific module (QLQ-H&N35) before, during, and at the end of treatment (T0-T2). Symptoms scoring ≥3 were considered clinically relevant, and descriptive statistics assessed symptom prevalence and dynamics. Results. Interim analysis included 19 patients with tumours of the oral cavity (n = 7), larynx (n = 5), hypopharynx (n = 4), and oropharynx (n = 3) (5.3% female (1), 94.7 % male (18)). The mean age was 64.84 ± 7.76 years. Clinically relevant emotional dysfunction was reported by 45%, while physical impairments - by 40% of patients. The most prevalent systemic symptoms were pain (70%), and fatigue (60%). Neurological and functional impairments were frequent and mostly increased from T0 to T2: xerostomia (42.1%→78.9%), sticky saliva (57.9%→78.9%), taste disturbance (15.8%→68.4%), and limited mouth opening (26.3%→57.9%). Chewing difficulties (68.4%→63.2%) and speech problems (57.9%→63.2%) were already highly prevalent at baseline and persisted throughout treatment, while neuropathic symptoms affected approximately one-third of patients (31.6%→36.8%). Smell disturbances were less common (26.3%→47.4%). Conclusions. Active radiotherapy for head and neck cancer is associated with a substantial and progressively increasing early neurological and functional symptom burden. Structured integration of patient-reported outcome measures may enhance detection of treatment-related neurofunctional toxicity, support proactive multidisciplinary management, and provide family physicians with actionable information for symptom monitoring and timely referral during post-treatment follow-up in primary care.

Background. Health literacy plays an important role in parents’ ability to understand and implement evidencebased recommendations for infant feeding. Using data from the Lithuanian WHO European Childhood Obesity Surveillance Initiative (COSI) 2023, we investigated the association between parental educational attainment, breastfeeding and exclusive breastfeeding practices. Methods. Data from 3,236 children were analyzed for breastfeeding initiation and from 2,823 children for exclusive breastfeeding. Parents were classified into two educational groups: secondary or lower education and higher education. Breastfeeding history was assessed using parental questionnaire responses regarding whether the child had ever been breastfed and whether the child had been exclusively breastfed. Associations between parental education and feeding practices were evaluated using Pearson’s chi-square test, with statistical significance set at p < 0.05. Results. Breastfeeding practices differed significantly according to parental educational attainment. Among families with secondary or lower education, 69.6% of children were breastfed, compared with 87.2% of families with higher education. Conversely, the proportion of children who had never been breastfed was more than twice as high in the lower parents’ education group (10.6% vs. 4.5%). Also, the proportion of children that were breastfed less than one month was higher (19.8% vs. 8.4%) in the lower education group. Similar disparities were observed for exclusive breastfeeding. Among families with secondary or lower education, 44.9% of children were exclusively breastfed, compared with 65.0% of families with higher education. The proportion of children who had never been exclusively breastfed was substantially higher in the lower education group (39.6% vs. 27.3%). All associations were statistically significant (p < 0.05). Conclusions. Although health literacy was not measured directly, these findings suggest that educational inequalities may influence the adoption of recommended infant-feeding practices. Strengthening family health literacy and targeted breastfeeding education among socially disadvantaged groups may help reduce inequalities and improve early-life health outcomes.

Background. Artificial intelligence and digital health tools are reshaping clinical workflows, decision-making and remote patient care. In response to this rapid integration, international and Lithuanian institutions provide guidance to support the safe and effective use of digital and AI technologies. These frameworks emphasize the need for strong digital, data governance and AI competencies across health professional roles. Aim. To identify the digital and AI-related competencies required in the era of artificial intelligence and to examine how healthcare professionals currently align with these competencies. Methods. The study consisted of two components. First, key strategic documents were reviewed to identify digital and AI related competencies relevant to modern healthcare. Second, the current competency landscape was analyzed using open data from the Health Technology Competency Questionnaire, developed by the Health Tech Accelerator working group and published on the Lithuanian Open Data Portal (data.gov.lt). A total of 107 respondents were included (55 physicians and 52 nurses-midwives). Descriptive statistics and the Chi-square test for homogeneity were applied to examine differences across professional groups. Results. Strategic documents most frequently highlighted competencies related to digital literacy, data governance, AI literacy, critical evaluation of algorithmic outputs, ethical digital practice and continuous digital upskilling. Respondent data showed strong basic digital skills and moderate understanding of information management, while AI literacy and critical evaluation of results remained limited. All statistically significant differences favored physicians, who demonstrated higher understanding of AI concepts, precision medicine principles, digital system use, digital communication tools and digital transformation processes. No differences were observed in continuous upskilling. Conclusion. The study shows that healthcare professionals have good basic digital skills, but more advanced AIrelated abilities are still uneven, with physicians scoring higher than nurses-midwives; this highlights the need to strengthen advanced digital and AI skills across all roles.

Background. Patient-reported outcomes (PROs) assessed with the EORTC QLQ-C30 provide information on symptoms, functioning and quality of life during radiotherapy. Repeated assessment may support patient-centred monitoring. Aim. To evaluate changes in PROs during radiotherapy and assess whether longitudinal modelling identifies associations missed by individual time-point analyses. Methods. Data from 34 patients with head and neck cancer were analysed at three time points: first day (T1), middle (T2) and last day (T3) of radiotherapy. PROs were assessed using the EORTC QLQ-C30 and transformed to a 0–100 scale. Repeated-measures ANOVA or Friedman test, Spearman correlations and linear mixed models (LMM) were used to assess changes over time and associations with disease stage and mean tumour dose. Results. Univariate tests showed significant changes in Global Health Status at T2 and T3 (p = 0.031) but no correlations with mean tumour dose. In contrast, the LMM identified significant effects of time and mean tumour dose (p = 0.003 and p = 0.026), indicating changes over the treatment course. Although traditional statistical methods showed no statistically significant differences in Physical Functioning mean scores either across time points (p = 0.065) or between tumour stage groups at each time point (all p > 0.05), the LMM revealed significant effects of time (p = 0.005) and disease stage (p = 0.005), indicating meaningful changes across the treatment course. While Friedman test showed a significant overall change in Nausea/Vomiting (p = 0.042), no pairwise comparisons remained significant after correction, and no correlations with mean tumour dose were observed. In contrast, the LMM identified significant effects of time (p < 0.001) and a dose by time interaction (p = 0.002). Conclusions. Repeated QLQ-C30 assessment captured changes in global health, physical functioning and nausea/ vomiting. Longitudinal modelling identified additional associations, supporting repeated PRO assessment during radiotherapy.

Background/Objectives: This is the first study in Eastern Europe to integrate skin and iris phenotypic characteristics with selected polymorphisms in genes related to skin aging, melanogenesis and skin pigmentation, and inflammatory regulation in relation to cutaneous melanoma (CM) risk. Methods: This case–control study included 180 patients with primary CM identified from the Hospital of the LSMU Kauno klinikos database and 181 controls without a personal history of skin cancer selected from a population-based study in Kaunas. All participants underwent dermatological and ophthalmological examinations and completed questionnaires assessing sun-exposure habits, including severe sunburn history, and Fitzpatrick skin phototype. The total-body number of common melanocytic nevi (CMN) >2 mm in diameter and the number of iris freckles were evaluated using standardized protocols. Five selected genetic polymorphisms, including TYR rs1126809, SLC45A2 rs26722, SLC45A2 rs16891982, IL6 rs1800795, and TNF-α rs1800629, were genotyped. Associations with CM were assessed using multivariable logistic regression. Results: We identified three phenotypical factors independently associated with increased odds of CM (OR; 95% CI): a total-body CMN count >15 (7.89; 4.29–14.53), ≥4 iris freckles (3.70; 2.04–6.74), and a history of severe sunburn (2.76; 1.50–5.07). Compared with Fitzpatrick skin phototype IV, phototypes II (0.35; 0.15–0.82) and III (0.28; 0.11–0.70) were associated with decreased CM odds. In the multivariable model the heterozygous SLC45A2 rs16891982 genotype was associated with increased CM odds (6.64; 1.48–29.83), while the IL6 rs1800795 Mono G genotype revealed an inverse association with CM (0.39; 0.16–0.97). Conclusions: A total-body CMN count >15, ≥4 iris freckles, and severe sunburn history revealed strong independent associations with increased CM risk. However, associations involving SLC45A2 rs16891982 and IL6 rs1800795 in this Lithuanian case–control study were model-dependent. These findings highlight the potential value of integrating selected genetic markers with phenotypic characteristics and sun-exposure history in the development of personalized CM risk prediction models; however, validation in larger populations is required.

Head and neck squamous cell carcinoma (HNSCC) exhibits substantial inter-patient variability in treatment response. Cone-beam computed tomography (CBCT) is routinely acquired during radiotherapy but remains underutilised for longitudinal tumour characterisation. This study aimed to develop a registration-consistent longitudinal CBCT framework for quantifying treatment-induced tumour texture and to evaluate the temporal stability of radiomic features as candidate imaging biomarkers. Unlike conventional pre/post-treatment comparisons, longitudinal intra-treatment trajectories remain insufficiently characterised.