Danijel Marinčić

Mr. Danijel Marinčić


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 May 2026 in Technologies
Verified
Journal Article
Open access
PDF file
PNPetar Nakić
Petar Nakić
Marija HabijanMarija Habijan
Marija Habijan
DMDanijel Marinčić
Danijel Marinčić
MMMarko Martinović
Marko Martinović
Accurate multi-class segmentation of the prostate in T2-weighted magnetic resonance imaging (MRI) into the peripheral zone (PZ), central gland (CG) and tumour is essential for targeted biopsy guidance and treatment planning. We present DSBANet, an encoder–decoder architecture that combines a pretrained ResNet-50 encoder, Atrous Spatial Pyramid Pooling, Multi-Scale Attention Fusion on skip connections, a Feature Fusion Module, deep supervision and boundary refinement. We evaluate eight architectures across three input dimensionalities (2D, 2.5D, 3D), yielding 24 models trained under identical conditions on the Prostate158 dataset. DSBANet achieves the best anatomy segmentation with PZ DSC of 0.8176 and CG DSC of 0.7888 among 2D models. To address the severe class imbalance of the tumour class, we further train DSBANet 2D with a class-weighted cross-entropy term and tumour-positive slice oversampling, raising per-case tumour DSC from 0.003 to 0.170 (a sixty-fold absolute improvement). A systematic eight-variant ablation study, evaluated under matched-pairs effect-size analysis, identifies the SE-Residual blocks and skip-connection attention as the largest contributors to tumour segmentation, while every architectural component contributes a directionally consistent gain.

16 February 2026 in Engineering Proceedings
Verified
Conference Paper
Open access
PDF file
MSMatej Spajić
Matej Spajić
Marija HabijanMarija Habijan
Marija Habijan
DMDanijel Marinčić
Danijel Marinčić
Irena GalićIrena Galić
Irena Galić
Semantic segmentation is a critical task in computer vision, enabling dense classification of image regions. This work investigates the effectiveness of the DeepLabv3+ architecture for binary semantic segmentation using annotated image data. A pretrained ResNet-101 backbone is employed to extract deep features, while Atrous Spatial Pyramid Pooling (ASPP) and a decoder module refine the segmentation outputs. The dataset provides per-image annotations indicating class presence, which are leveraged to approximate segmentation masks for training purposes. Various data augmentation techniques and training strategies were applied to support effective learning and reduce overfitting. Experimental results on the MHIST dataset show that the proposed pipeline achieves strong performance despite the lack of pixel-level annotations, with a mean Intersection-over-Union (mIoU) of 0.76 and a mean Dice coefficient of 0.84. These confirm the potential of weakly supervised segmentation using class-aware CAMs and deep pretrained encoders for structured pixel-level prediction tasks in medical imaging.

12 October 2026 in Diseases
Verified
Journal Article
Open access
PDF file
ZBZvonimir Bosnic
Zvonimir Bosnic
HPHrvoje Pitlovic
Hrvoje Pitlovic
IŠIvana Šofić
Ivana Šofić
MMMaja Miletić
Maja Miletić
DMDanijel Marinčić
Danijel Marinčić
TBTatjana Bačun
Tatjana Bačun
Background: Falls in older women are multifactorial events driven by the complex interaction of age-related loss of physiological reserve, frailty, sarcopenia, chronic low-grade inflammation, multimorbidity, and medication exposure. Older women represent a substantial proportion of fall-related injuries and fractures, yet available evidence is highly heterogeneous. Since many studies focus on mixed-sex cohorts, we must differentiate women-specific findings from general geriatric data. Methods: A structured narrative review was conducted to integrate biological, clinical, and pharmacological evidence on falls in older women, focusing particularly on primary care. The initial literature search covered PubMed, Scopus, Web of Science, and Embase from January 2000 to March 2025. To ensure the integration of recent high-level evidence, a supplementary search and backward citation tracking were subsequently executed through 31 August 2026, aligning with peer-review recommendations. Women-specific evidence was prioritized over mixed-sex cohorts. Given that the review aims for an interpretive synthesis rather than quantitative effect estimation, a meta-analysis and formal risk-of-bias assessments were omitted. Discussion: Frailty and sarcopenia are consistently associated with falls and functional decline in older adults. In contrast, inflammaging and metabolic dysregulation are more strongly supported as contributors to frailty, sarcopenia, and functional vulnerability than as independently established causal determinants of falls. Medication-related fall susceptibility may arise through several mechanisms, including sedation, impaired attention and psychomotor function, orthostatic hypotension, bradycardia, electrolyte and glucose disturbances, gait impairment, and adverse skeletal effects. Psychotropic agents, anticholinergic drugs, opioids, selected cardiovascular drugs, and other fall-risk-increasing drugs warrant particular attention in patients with reduced physiological reserve. Medication review and appropriate deprescribing should be considered components of multidomain fall-prevention strategies rather than stand-alone interventions. Conclusions: The principal contribution of this review is not the identification of new individual fall-risk factors, but the integration of established biological vulnerability, functional impairment, and pharmacological burden within a sex-sensitive primary-care framework. This approach may help clinicians consider frailty, medication exposure, and women-specific modifiers simultaneously during fall-risk assessment. The proposed framework is hypothesis-generating and should not be interpreted as a prospectively validated mechanistic or prediction model.